British Journal of Haernatology. 1992, 82, 5 7 5 - 5 8 3
Mechanisms of inhibition of T cell IL-2 secretion by factor VIII concentrates TUBBS,*TREVOR B A R R O W C L I F F EB,E~ R N A R D MAHONA N D ROBINTHORPEDivisions of lmrnunobiology and *Haernatology, NIBSC, South Mimms, P o t t m Bar, Herts. MEEN11 W A D H W A , P A U L A D I L G E R , J I L L
Received 5 February 7 992; accepted for publication 27 June 1992
Summary. We have continued our previous study of the inhibitory effects of factor VIII concentrates on IL-2 secretion by T cells. Experiments with an extended range of products confirm our previous conclusion that some but not all low, intermediate and high purity concentrates possess inhibitory activity on IL-2 secretion. The inhibition occurs almost immediately after addition of factor VIII concentrate and it was not possible to adsorb inhibitory activity with activated or non-activated cells: this suggests that the mechanism of inhibition involves interference with early T cell activation events rather than simple blocking of cell surface components by inhibitory molecules. The inhibitory components were shown to reside in different molecular weight fractions of concentrates. A strongly inhibitory component of approximately 200 kD and a minor species of approximately 60 kD were identified in strongly inhibitory concentrates. Some products contained a
dialysable inhibitory substance which is most likely a salt as it was also present in some formulation buffers. The proportions of the inhibitory components varied widely between products. We have found that the pattern of inhibition using in vitro systems reflects that observed using a mouse in vivo antigen challenge method. In addition we have shown that the previously reported concentrate mediated inhibition of lectin induced low affinity IL-2 receptor (CD25) is mainly a consequence of diminished IL-2 secretion rather than a ‘direct’ effect on CD2 5 expression. Considering the wide variation between products of the same purity group, caution should be exercised in drawing conclusions concerning the immunosuppressive effects of a particular type of concentrate in haemophilia patients from study with only one product from that group.
Since the occurrence of HIV-1 infection and AIDS in haemophiliacs, resulting from the use of HIV contaminated factor VIII concentrates, there has been intensive study of the immune system in both HIV seropositive and seronegative patients. It is now recognized that, in addition to the immunosuppressive effects of HIV infection, an apparently milder form of immunosuppression can occur in HIV negative haemophiliacs (Ludlam et al, 1985; Madhok et al, 1986; Madhok. 1989; Cuthbert et al, 1992). The clinical significance of this form of immunosuppression, manifested as deficient responses to skin antigen tests and involving both T-cell and B-cell dysfunction (Madhok et al. 1990, 1991a, b), remains uncertain. Although the immunological changes are thought to be related in some way to repeated alloantigen stimulation from high-dose
factor VIII therapy, the exact mechanism, and the agent(s) responsible for the immunosuppressive effects are unknown. In vitro studies have shown that factor VIII Concentrates can be potent inhibitors of lymphocyte proliferation and interleukin-2 (IL-2)secretion by T lymphocytes (Froebelef al. 1983; Lederman et al, 1986; Thorpe et al. 1989). In our previous studies we found no obvious relationship between concentrate inhibitory activity and protein concentration. Although very high purity factor VIII products were all noninhibitory, the degree of inhibition with intermediate-purity products varied widely (between 7% and 98%) at the same factor VTII concentration (Thorpe et al, 1989). In the present study we have carried out further investigations on the mechanism by which factor VIII concentrates inhibit IL-2 secretion in vitro, including analyses of the molecular weight of the inhibitory component(s). effect of concentrates on in vivo antigen stimulated IL-2 secretion, and studies of the effect of concentrates on IL-2 receptor expression.
Correspondence: Dr M. Wadhwa, Division of Immunobiology. NIBSC. Blanche Lane, South Mimms, Potters Bar, Herts. EN6 3QG.
575
(8mls of 2 x 106/ml)
(0.8mls of 42IU/ml) Incubated overnight at 37OC
Supernatants assayed for IL-2
MATERIALS A N D METHODS
Factor VIII concentrates. The details of the concentrates used in the study are: Intermediate-purity concentrates (spec. act. 0 , 5 1 0 IU/mg protein): Dry heated (80°C. 72 h ) A, from Bio-Products Laboratory, Elstree. U.K.. and B. from Scottish National Blood Transfusion Service, Edinburgh, U.K.: Solvent/detergent treated C. from Centre Regionale de Transfusion de Sang. Lille. France, D. from Octapharma GmbH. Germany, and E, from Alpha Therapeutic, Los Angeles. U.S.A. Heated in organic solvent F. from Alpha Therapeutic, Los Angeles. U.S.A. Pasteurized G. from Behringwerke. Marburg. Germany. High-purity concentrates prepared by chromatography (spec. act. 50-2 50 IU/mg protein): Solvent/ detergent treated H. from CRTS, Lille, and J, from Octapharma. Pasteurized K. from Behringwerke. Very high-purity concentrates, prepared by rnonoc~ona~ antibody absorption (spec. act. 2 3000 IU/mg protein before addition of an albumin stabilizer): Solvent/detergent treated I,, from BPL. Elstree. Pasteurized M, from Armour Pharmaceuticals, Kankakee. {J.S.A. Recombinant N, from Cutter Laboratories. Berkeley. IJ.S.A.. and P. from Baxter laboratories. 1.0s Angeles. U.S.A. Stimulation of IL-2 secretion using Irrrkat cells. The Jurkat human T cell lymphoblastoid line was maintained at a density of 10i-lOh/ml in RPMI 1640 medium containing 10% heat inactivated fetal calf serum (culture medium). For stimulation of IL-2 secretion, Jurkat cells were suspended at a density of 2 x lohcells/ml in culture medium and cultured in the presence of 1% PHA (Wellcome)with or without factor VIII concentrates in flat-sided tubes at 3 7°C in a humidified atmosphere containing 5% C02. In each assay. control cultures without PHA were included simultaneously. The supernatants were harvested after 18-24 h of culture and assayed for 1L-2 as described below. Measurement of IG2. IL-2 was measured using a bioassay based on proliferation of the IL-2 dependent mouse cytotoxic lymphocyte cell line, CTLL-2 (Gilliset al. 1 9 78: Wadhwa et al. 1991). This line was unaffected by factor VIII concentrates (Thorpe et al. 1989). For this, 50 pl of cell suspension ( 5 x lo3cells) was added to 50 pI of test samples or control medium and cultured for 18 h in 96-well microtitre plates at 3 7°C in a humidified CO2 incubator. A titration of an 11,-2 working standard calibrated
Fig 1 . Outline of adsorption experiment using unstirnulated Jurkatcells.
directly against the WHO 1st international standard for IL-2 ( 8 6 /504) was included in each assay. The cultures were then pulsed for 4 h with 3H-thymidine (18.5 kBq/well), harvested onto filter mats and the radioactivity incorporated into DNA estimated by scintillation counting. The percentage inhibition was calculated using parallel line analysis (Finney, 19 78). Dialysis of factor VlIl concentrates. Products were dialysed u s n g tubing size 1-8/32 (Medicell International Ltd) for 4 h at 4°C with three changes of phosphate buffered saline. The dialysed and non-dialysed products (at 5 IU/ml) were tested for inhibition of IL-2 secretion using the Jurkat cell-line bioassay. 4dsorption experiments. These experiments were performed in two different ways to determine if the inhibitory component could be adsorbed or internalized by Jurkat cells. ( 1 ) Unstimulated cells were used to establish if these could adsorb the inhibitory component using the protocol outlined in Fig 1. In some experiments, factor VIII concentrate was incubated with three successive batches of fresh unstimulated cells (6 ml of 2 x 10b/ml)over 3 h at 3 7°C instead of an extended overnight incubation with only one batch of cells. The supernatant (containing factor VIII concentrate) from the final incubation step was added to fresh Jurkat cells with PHA to test for remaining inhibitory effects on IL-2 secretion. (2) PHA stimulated Jurkat cells were used in an attempt to adsorb out the inhibitory component of the factor VIII concentrates. Factor VIII concentrate was added to three successive batches of stimulated cells (5 ml of 2 x lOh/rnl) as in Fig 2 . Appropriate controls were included in all the experiments. €:$feet of fuctor VIII concentrates on expression of low affinity TL-.? receptor (CD25). To further elucidate the possible mechanism of inhibition of K-2 secretion and T cell proliferation, the effect of factor VIII concentrates on IL-2 receptor expression was examined using peripheral blood mononuclear cells (PBMC). PBMCs were separated from blood by centrifugation over Ficol/Hypaque at 400 g for 30 min (Lymphoprep. Nycomed). Mononuclear cells were removed from the interface. washed three times in RPMI 1640 and resuspended to a density of 2 x lob cells/ml in culture medium. The cells were cultured with PHA in the presence or absence of factor VIII concentrates in flat-sided tubes for 72 h
FVIII Concentrate Inhibition of IL-2 Secretion
5 77
factor VIII concentrate ( 0 . 5 ~ 1 1 of~ 42IU/ml) -I-
PHA stimulated washed Jurkat cells (51111sof 2 x
I
l o 6 cells/ml)
Incubated 2hr at 37OC
Centrifugation
Discard
cells
PHA stimulated Jurkat cells (6 x
lo6)
Supernatant assayed for IL-2 activity Fig 2. Schematic diagram of adsorption experiment using stimulated Jurkat cells.
at 37OC in a C 0 2 incubator. Appropriate controls (medium or factor VIII concentrate alone) were also included in the assay. Following this period, the expression of the cell surface CD2 5 antigen was detected by incubation with an anti CD25 monoclonal antibody, anti-tac (Uchiyama et al, 1981) and FITC conjugated goat anti-mouse Ig and subsequent analysis using a FACScan flow cytometer (Becton-Dickinson).In some experiments, cells were cultured with exogenous IL-2 (100 IU/ml) and PHA in the presence or absence of factor VIII concentrates for 72 h and the expression of CD25 monitored as described above. Appropriate controls (treated cells and cells incubated with an irrelevant antibody of the same subclass as the anti-tac antibody) were included in all experiments. Effect of factor VlIl concentrates on the intermediate affinity IL-2 receptor (75 kD).This was assessed by flow cytometry using the protocol described above but with a different antibody, TU2 7 which recognizes the p7 5 IL-2 receptor chain (Ohashi et al, 1989). Fractionation of factor VIII concentrates. Four factor VIII concentrates of different purity were fractionated by gel filtration using a Sepharose CL-4B column (Pharmacia). Concentrates (total factor VIII content xl000 IU) were applied to the column and eluted overnight with TBS (50 mM Tris/l50 mM sodium chloride buffer pH 7.4) at 30 ml/h collecting 5 ml fractions. The fractions were monitored for protein by measuring their absorbance at 280 nm and any inhibitory effects on IL-2 secretion by Jurkat cells. Some fractions were further analysed by SDS-PAGE to determine the protein composition more precisely.
Stimulation of IL-2 secretion by in vivo antigenir challenge using sheep red blood cells (SRBC). Balb/c mice were injected in their hind footpads with 10 pl of a 10%suspension of sheep red blood cells. After 3 d the popliteal lymph nodes were aseptically removed, washed and the cells suspended in RPMI 1640 supplemented with 3% FCS. Viability was assessed by trypan blue exclusion, and the cells cultured at 5 x 10h/ml with or without factor VIII concentrate in flat-sided tubes at 37°C in a humidified C02 incubator. Supernatants from lymphocyte suspensions were harvested at 24 h by centrifugation and stored at - 20°C until assayed for IL-2 activity as described above.
RESULTS Inhibition of IL-2 secretion The intermediate-purity products tested in the present study represent most of the types of concentrate available currently in the U.K. and Europe and include those prepared using solvent/detergent methods of viral inactivation. Results with this group of products showed a similar diversity of inhibitory activity as in the previous study; at 1 IU/ml the inhibitory activity ranged from 30% to S5%, and at 2.5 IU/ml from 2 5% to 99% (Table I). Three of the products from the current group which had also been included in the previous study showed almost identical inhibitory activities in the two studies, demonstrating consistency of both production methods and assay methodology. There was little batch to batch variation in inhibitory activity.
Meenu Wadhwa et a1
5 78
Table 1. Inhibition of IL-2 secretion by concentrates.
( 5 -20%) was removed by dialysis. However, for product G, and for the conventional high-purity products H and J, the proportion was higher, 40-60%. These products, as well as product K, contain stabilizers such as citrate which are added to prevent factor VIII degradation during viral inactivation procedures. The formulation buffer, Y, which is used for products G and K was also tested: as shown in Fig 3 it also inhibited IL-2 secretion and most of this inhibitory activity was removed by dialysis. A s can be seen from Fig 3 , there were no concentrates in which the inhibitory activity was totally removed by dialysis, and in all cases the activity remained above 50% at the concentration tested.
Inhibition of I L 2 secretion (%I Concentrate type
2.5 IU/ml
Code ~~
Intermediate purity
C
25 82 99
u
84
A R
1 ILl!rnl
~
85 30
52 70
E
F Conventional high-purity
G
99 99
H
79
i3
J
Very-high-purity
K
90
I,
0 0
bl
Kiwtics of inhibition by factor VlIl concentrates To determine the kinetics of inhibition of IL-2 secretion, factor VIlI concentrate was added to Jurkat cells a t various time ini-ervals after stimulation with PHA. Maximum inhibition WBS produced only when factor VIII was added simultaneously with the mitogen. Delayed exposure to factor VIII for 2 h or more resulted in a progressive attenuation of the inhibitory effect on IL-2 secretion induced by PHA (Fig 4). This suggests that the inhibitory effect of factor VIII concentrate is exerted almost immediately on addition of the concentrate, i.e. IL-2 levels detected reflect the IL-2 which is secreted and accumulates in the culture medium prior to addition of inhibitory factor VIII concentrate.
41 43 32
N
0
P
0
The very-high-purity products. including the two rDNA derived concentrates. gave no inhibitory activity at any of the concentrations tested, up to 5 ll'/ml. However, somewhat surprisingly, the high-purity products prepared by conventional technology, with specific activities over 100 IC'/mg. had substantial inhibitory activity: 30-45'j& at 1 IU/ml and SO-SO% at 2 . 5 IlJ/ml (Table I). As shown in a subsequent section (see below). a considerable portion of this activity is eliminated by dialysis. Eflec'ts of dialysis on inhibition of 11,-2 secretion Because some factor VIII concentrates are stabilized with various buffers, high salt, etc., it is possible that these could influence IL-2 secretion. Accordingly we carried out a careful and thorough dialysis of several of the inhibitory products. after first optimizing the method. As shown in Fig 3 . the effects of dialysis differed considerably among the products. For most of the intermediate-purity concentrates. only a relatively small proportion of the total inhibitory activity
Adsorption of fmtor Vlll concentrates with resting or activated cr,'ls does riot eliminate the inhibitory component(s) from factor I.'M concentrates Resting or activated Jurkat cells were used in an attempt to adsorb the inhibitory components from the concentrates. In both cases the inhibitory component(s) from the concentrate could not be removed by adsorption onto cell surfaces after a n 01ernight incubation even when three successive batches of cells were used. The supernatants (containing factor VIII concentrate) obtained from both experiments were still capable of exerting a n unreduced inhibitory effect on IL-2 secretion when added to fresh Jurkat cells in the presence of PlIA (Fig 5A. B). In addition, unstimulated Jurkat cells previously incubated with the concentrate (in a n attempt to
C__i Before 2:'
Dialysis
A J FaCtorVII concenlrate Y .Albumtn C wale
"
A
B
C
O
E
F
G
H
J
Y
Fig 3. The effect of dialysis on inhibition of interleukin-2 secretion by factor VIII concentrates.
FVIll Concentrate lnhibition of IL-2 Secretion
5 79
reduction in the percentage of cells expressing the CD25 antigen was seen. This effect was dose-dependent with maximal inhibition occurring when factor VIII was added at a concentration of 4 IU/ml; the mean percentage of CD2 5 cells was 44.9f8.5 as compared to 81.5 k8.8 in the absence of the concentrate. A representative experiment is shown in Fig 7. When a less inhibitory concentrate was used, the expression of IL-2 receptor was not significantly diminished. As the decreased expression of the IL-2 receptor may be due to inhibition of IL-2 secretion, experiments were performed to test whether addition of exogenous IL-2 could reverse the reduction in IL-2 receptor expression caused by inhibitory concentrates. Addition of IL-2 (100 IU/ml) could partially reverse the inhibitory effect of factor VIII concentrate (4 IU/ ml) on IL-2 receptor expression: the mean percentage of CD25+ cells in these experiments was 66.2&2.2% (Fig 7). +
0‘ 0
5
I
I
I
10
15
25
20
Time of F.VIII Addition (Hours)
Fig 4. Time course of inhibition of IL-2 secretion from Jurkat cells by factor VIII concentrates. The data shown is from an experiment in which product F was used at a concentration of 4 IU/ml.
adsorb the inhibitory component) still released a ‘normal’ level of IL-2 into the culture medium after stimulation with PHA (Fig 6) demonstrating that the cells had not adsorbed any inhibitory substance from the concentrate.
Modulation of PHA-induced IL-2 receptor (CD25) expression by factor VllI concentrates Using flow cytofluorimetric analysis, the induction of IL-2 receptor by a n optimal concentration of PHA, as measured by the percentage of CD25+ cells, was evaluated in PBMC obtained from four individual donors. The mean percentage of CD25+ cells following stimulation with PHA was 8 1 . 5 f8 . 8 1 . In the presence of an inhibitory concentrate (i.e. one which strongly inhibits IL-2 secretion), a significant
Factor Vlll concentrates do not effect expression of the intermediate afinity ( p 7 5 ) IL-2 receptor chain Using the TU27 antibody, the effect of factor VIII concentrates on expression of the intermediate affinity IL-2 receptor on PHA induced blasts was examined. Preliminary data indicated that PHA by itself did not effect the percentage of cells expressing the p75 kD 11,-2 receptor. None of the concentrates appeared to have any significant effect on the expression of the p75 kD IL-2 receptor (data not shown). Inhibitory effect offactor VllI concentrates in an in viva antigenic stimulation system To investigate if inhibitory effects on IL-2 secretion mediated by factor VlII concentrates could also be observed following in vivo stimulation of T cells with antigen (rather than in vitro
B.
A.
50,000
80,000
PHA
8
-PHA
rADSFACVII1 PHAIFAC Vlll
e
40,000
60,000
30,000
\
E0
40,000 20,000
20,000 10,000
n 1:2
1:4
1:8
Supernatant dilution
1:16
0
1:2
1:4
1:8
1:16
Supernatant dilution
Fig 5 . Effect of adsorbed and non-adsorbed factor VIII concentrate on PHA mediated IL-2 secretion.In A. product F (4 IU/ml) was incubated with three successive batches of stimulated cells. In B, three successive batches of unstimulated cells were used to adsorb the product F (4 IU/ml). The adsorbed concentrate was then tested for effect on IL2 secretion from fresh Jurkat cells.
Meenu Wudhwa et uI
580
80.000
0
.
--=-+=_- -_
--_
1:2
b------~
1:4
1:16
1:8
1.2
1 :4
1.8
1:16
Supernatant Dilution
Supernatantdilution Fig 6. Effect of adsorbed and non-adsorbed factor VIII concentrate on PHA mediated IL-2 secretion.In this experiment.product F (4 IU/ml)
Fig 8. Etrect of two different factor VIII concentrates on inhibition of IL-i! secretion by lymph node cells from mice undergoing a primary
was incubated overnight with unstimulated cells. The symbols indicate IL-2 secretion on stimulation with PHA from: ( a ) cells previously incubated overnight with factor VIII concentrate to adsorb the inhibitory component (open square): (b) fresh cells in the presence of supernatant containing factor VIII (previouslyincubated overnight with cells to adsorb the inhibitory component) (closed triangle);(c)fresh cells in the presence offactor VIII concentrate (i.e. not used to adsorb the inhibitory component) (closed square): and (d) fresh cells in medium alone (open triangle).
apparent. Similar experiments carried out using weakly inhibitory or non-inhibitory concentrates showed that the pattern of inhibition observed using the in vivo stimulatory system was similar to that seen with Jurkat cells in v i m (Fig 8).
activation with lectins). hind footpads of mice were injected with sheep red blood cells. When cell suspensions, obtained from popliteal lymph nodes 3 d after the antigenic challenge, were incubated overnight, IL-2 was secreted into the culture medium without any in vitro stimulus. However, if a strongly inhibitory factor VTII concentrate was incubated with the lymph node cells. an inhibition of IL-2 secretion was
immune response to sheep red blood cells.
The nature of the inhibitory component(s) Despite the apparent differences in inhibitory activity, the size exclusion chromatography profile of the various products was similar and did not show any obvious differences between the inhibitory and non-inhibitory preparations (Thorpe et al, 1989). Evaluation of the fractions from inhibitory preparations for their effect on IL-2 secretion indicated that a major peak of inhibitory activity was evident in the fractions which were eluted subsequent to the void volume (containing factor VIII). A much smaller peak was also observed at a molecular weight of approximately 60 kD. These components could not be identified by SDS PAGE since there is no obvious specific band in the inhibitory fractions which is not present in non-inhibitory portions of the column profile (data not shown). In some concentrates, a strong inhibitory component eluted a t the total column volume (the expected elution point for buffer salts) (Fig 9). As expected, none of the fractions from non-inhibitory factor VIII products exerted any inhibitory effects (data not shown).
DISCUSSION
TOO
10’
lo2 FL 1
lo3
10‘
10’
10’
103
lo4
FL 1
Log (Fluorescence Intensity) Fig 7 . Effect of factor VIII concentrates on the expression of the low affinity chain (CD25) of the 1L-2 receptor using peripheral blood mononuclear cells. la) Control. (b) PHA. Ic) PHA+factor VIII concentrate (4 IU/rnl). and (d) PHA+factor VIII concentrate (4 I U j ml) + IL-2 ( 100 1U/rnl j. The data shown has been obtained from an experiment in which product F was used.
The data on inhibition of IL-2 secretion by various types of factor VIII concentrates confirms our previous observations. The degree of inhibition with the intermediate purity concentrates differed considerably among the various products, suggesting that inhibitory action was due to some specific component(s). rather than being merely a function of protein concentration. Extensive dialysis had little effect on inhibitory activity of most intermediate purity concentrates, indicating that most of the inhibitory action is due to a large molecular weight component(s).
FVIIl Concentrate Inhibition of IL-2 Secretion
58 1
0.
2
100 I
80
11.2
1
0.8
60
,E 0
aD N
0.6 40
@
8C
a
0.4
2
9 20
0.2
0
Fraction Number
Fraction Number Fig 9. Fractionation of inhibitory activity in two different factor VIII concentrates. Figures A and B represent products F and G. Factor VIII was recovered in the void volume at fraction 3 3 in figure A and at fraction 39 in figure B. Closed squares denote the inhibitory activity in the fractions and open squares denote the absorbance of the fractions at 280 nm.
It is notable that high-purity concentrates, prepared by conventional chromatography, displayed similar inhibitory characteristics to intermediate purity products. However, with these high-purity products, a much larger proportion of the activity was removed by dialysis, suggesting that salt stabilizers such as citrate were responsible for a large part of the inhibitory activity-indeed the formulation buffer for one of the products itself showed strong inhibitory action. Nevertheless, even after extensive dialysis some inhibitory activity remained in high purity concentrates, and this indicates that the protein(s) responsible had been partially, but not completely, removed by the purification process. Results of size exclusion chromatography showed that most of the inhibitory activity was located in a protein fraction of around 200 kD. We were unable to confirm the earlier results of Lederman et al (1986), who found activity also in the void volume. This fraction contains large amounts of fibrinogen and fibronectin, which account for the bulk of the protein in intermediate purity concentrates, but these are unlikely to be responsible for the inhibitory action, as they had no effect when tested as purified proteins (unpublished observation). Further identification of the inhibitory components has not proved possible to date, though clearly the activity is removed by monoclonal antibody affinity purification, since concentrates prepared from both plasma and recombinant sources by this method showed no inhibitory action. The possibility that inhibitory factor VIII concentrates contain molecules which bind specifically to cell surface components involved in mitogen mediated triggering of IL-2 secretion was investigated by attempting to adsorb such substance(s) by incubation with resting or PHA activated Jurkat cells. No reduction in inhibitory activity was observed with factor VIII concentrates adsorbed with activated or non-
activated cells implying that the inhibitory component(s) mediate their effects by mechanism(s) which do not involve a simple ‘blocking’ interaction with cell surface components. The results obtained using three separate consecutive adsorptions with large numbers of activated Jurkat cells suggest that it is unlikely that the failure to adsorb inhibitory activity was due to a low level of expression of cell-surface molecules involved in binding of inhibitory components present in factor VIII concentrates. Non-stimulated Jurkat cells incubated with an inhibitory factor VIII concentrate completely retained the ability to secrete IL-2 when stimulated by PHA if the factor VIII concentrate was removed from the culture system, which also suggests that no significant adsorption of inhibitory molecules occurs at the cell surface, at least prior to lectin activation. Experiments involving addition of inhibitory factor VIII concentrates to cultures of Jurkat cells at various intervals after stimulation with PHA showed that inhibition of It-2 secretion occurs almost instantaneously when concentrate is added. This also suggests that the mechanism of inhibition involves process(es)other than simple blocking of binding of lectin to its receptor@),or interference with early triggering events which occur soon after lectin binding to cell surface molecules. The results of several studies clearly indicate that some factor VIII concentrates inhibit lectin induced IL-2 secretion in vitru; however. there is little direct information available relating to factor VIII concentrate effects on IL-2 secretion resulting from an in vivo challenge with antigen. Our experiments described in this report using the mouse/sheep erythrocyte challenge model for in vivo induction of IL-2 secretion (Spitz et al. 1985) show that factor VIII concentrates which inhibit IL-2 secretion in vitru are also potent inhibitors of in vivo antigen triggered IL-2 secretion by mouse
582
Meenu Wadhwa et af
lymphocytes. Inhibitory activity of concentrates in the in vivo model reflected those observed with those concentrates using in vitro experiments with Jurkat cells and PHA stimulation. It can be concluded that such concentrates are likely to inhibit T cell IL-2 secretion regardless of environment or stimulus. We have confirmed a previous report (Hay et al. 1 9 9 0 )that factor VIII concentrates can inhibit the expression of the 55 kD chain of the IL-2 receptor (CD25) on the surface of PHA stimulated T cells. The capacity of different products to mediate this effect varied and seemed to correlate with ability to inhibit IL-2 secretion by Jurkat cells. It has been suggested (Hay et a!. 1990) that concentrate inhibition of IL-2 secretion occurs because CD2 5 expression 1 upregulation) is inhibited: however, it is well established that IL-2 is a potent upregulator of the 5 5 kD chain of the receptor (Reem & Yeh. 1984) and we observed that addition of rDNA derived IL-2 to cultures could at least partially overcome concentrate inhibition of PHA induced upregulation of CD2 5. It therefore seems most likely that factor VIII concentrates which inhibit expression of CD25 do this at least partially by inhibiting secretion of IL-2. which results in considerably reduced IL-2 mediated upregulation of the cell surface bound 55 kD receptor chain. However, we could not entirely reverse concentrate inhibition of PHA induced CD2S even by addition of relatively large amounts of 11,-2 and so the phenomenon may be caused by more than one mechanism. We failed to observe any effects of factor VIII concentrates on cell surface amounts of the 7 5 kD chain of the IL-2 receptor. as would be expected for the dynamics of expression of this molecule (Thuy ~t al, 1987). The relevance of these in virro studies to immune disturbances in haemophilia A patients remains unclear. Most of the studies on haemophilic patients have been done on baseline samples, i.e. from patients not receiving concentrates at the time of sampling: it is possible that injection of concentrates results in acute depression of IL-2 secretion, such as we observe in vitro. followed by gradual normalization as the inhibitory components are cleared from the circulation. Madhok et a1 (1990) found that the ability to secrete IL-2 in response to antigen was impaired in patients both negative and positive for antibodies against HIV: in a subsequent study the same group found evidence for T cell sensitization to factor VIII concentrates (Madhok et al. 1991b). and it was suggested that repeated alloantigen stimulation could lead to a partial exhaustion of the ability of lymphocytes to secrete IL-2. From the point of view of treatment of haemophilia. the two most important questions are whether the immune suppression observed in HIV negative patients has any clinical significance with respect to exposure to infectious agents, and whether there is any difference in outcome in HIV positive patients given different types of concentrates. In one comparison of a monoclonal antibody affinity purified concentrate with two intermediate purity products there were no significant differences between patients treated with the various concentrates with respect to skin anergy and CD4 counts (Brettler et a!. 1989): however, in another study patients given a monoclonal antibody affinity purified concentrate had a slower decline in CD4 counts than a group
given a n intermediate purity product (de Biasi et a!, 1991). It should be noted, however, that the intermediate purity products in these two studies were not the same, and our studies emphasize that, at least with respect to IL-2 secretion, there are wide differences between products even in the same 'purity' group. Therefore the conclusion of this latter study, that very high purity concentrates are superior to intermediate purity concentrates for treatment of HIV positive patients, should be viewed with caution.
ACKNOWLEDGMENTS We thank Bio Products Laboratory, Elstree, U.K.: Scottish National Blood Transfusion Service, Edinburgh, U.K.: Centre Regionale de Transfusion de Sang, Lille, France: Alpha Therapeutic, Los Angeles, U.S.A.; Behringwerke, Marburg, Germany: Octapharma GmbH, Germany: Armour Pharmaceuticals. Kankakee, U.S.A.: Cutter Laboratories, Berkeley, U.S.A.: and Baxter Laboratories, Los Angeles, U.S.A., for providing the factor VIII concentrates used in this study. We thank Dr Kazuo Sugamura for his generous gift of the TC;! 7 antibody, Dr Tony Mire-Sluis for helpful discussions and Deborah Kirk for preparation of the manuscript. The sup2ort for parts of this project by the Scottish National Blood Transfusion Service is greatly appreciated.
REFERENCES Bretller. D.B.. Forsberg. A.D.. Levine. P.H.. Petillo, J., Lamon, K. & Sullivan. J.L. (1989) Factor VIII: C concentrate purified from plasma using monoclonal antibodies: human studies. Blood. 73, 1k 59-1863. Cuthbert. R.J.G..Ludlam. C.A., Steel. C.M.. Beatson, 1). & Pentheres, J.F. ( 1992)Immunologicalalterationsin HIV seronegativehaemophiliacs: relationships to blood product therapy. British Journal of Hoemotology, 80, 364- 3 69. de Biasi. R.. Rocino. A.. Miraglia. E.. Mastrullo, L. & Quirino, A.A. ( 1'39 1) The impact of a very high purity factor VIII concentrate on the immune system of human immunodeficiency virus-infected
hemophiliacs: a randomized prospective, two-year comparison wi1.h an intermediate purity concentrate. Blood, 78, 1919-1922. Finncy. D.J. (1978) Statistical Methods in Biological Assay. Charles Griffin. London. Froebel. K.S.. Madhok. R.. Forbes, C.D.. Lennie. S . , Lowe, G.D.O. & StLrrock. R.D. (1 983) Immunological abnormalities in haemophilia: are they caused by American factor VIII concentrate. British Medical ]ournal. 28 7. 1091-109 3. Gillis. S.. Fern. M.M.. Ou. W.E. & Smith, K.A. (1978) T cell growth facer: parameters of production and a quantitative microassay for activity. Journal of Immunology. 120, 2027-2032. Hay. C.R.M.. McEvoy. P. & Duggan-Keen. M. (1990) Inhibition of lymphocyte ILL-receptor expression by factor VIII concentrate: a possible cause of immunosuppression in haemophiliacs. British ]otrrnal of Haematology. 75. 278-281. kderman. M.M., Saunders. C.. Toossi, Z., Lemon, N.. Everson. B. & Rainoff. O.D. (1986) Antihemophiliac factor (factor VIII) preparations inhibit lymphocyte proliferation and production of interleukin-2. Journal of Laboratory and Clinical Mcdicine. 107, 4i 1-478.
FVIII Concentrate Inhibition of IL-2 Secretion Ludlam, C.A., Tucker, J., Steel. C.M., Tedder, R.S., CheingsongPopov, R., Weiss, R.A., McClelland,D.B.. Philip, I. & Prescott. R.J. (1985) Human T-lymphotropic virus type 111 (HTLV-111)infection in seronegative haemophiliacs after transfusion of factor VIII. Lancet, ii, 233-236. Madhok, R. (1989) The in vivo and in vitro effects of clotting factor concentrates on the immune system. Replacement Therapy in Haemophilia: The Problems and Solutions (ed. by P. M. Mannucci). pp. 55-60. Medi Fax International, U.K. Madhok, R., Gracie, J.A.,Forbes, C.D. & Lowe, G.D.O. (1991a) B cell dysfunction in haemophilia in the absence and presence of HIV-1 infection. Thrombosis and Haemostasis, 65, (i), 7-10. Madhok. R., Gracie, J.A., Lowe. G.D.O., Froebel, K., Burnett, A., Follett, E. &Forbes. C.D. (1986) Impaired cell mediated immunity in haemophilia in the absence of infection with human immunodeficiency virus. British Medical Journal. 293, 978-98 1. Madhok. R.. Gracie. J.A..Smith, J., Lowe. G.D. &Forbes. C.D. (1990) Capacity to produce interleukin-2 is impaired in haemophilia in the absence and presence of HIV I infection. British Journal of Haematology, 76, 70-74. Madhok. R., Smith, J., Jenkins, A. & Lowe, G.D.O. (1991b) T cell sensitization to factor VIII in haemophilia A? British Journal of Haematology. 79, 235-238. Ohashi. Y.. Takeshita. T.. Nagata,K.. Mori,S.&Sugamura, K. (1989) Differential expression of the interleukin-2 receptor subunits, p55 and p75 on various populations of primary peripheral blood mononuclear cells. Journal of Immunology, 143, 3548-3555.
5 83
Reem, G.H. & Yeh. N. (1984) Interleukin 2 regulates expression of its receptor and synthesis of gamma interferon by human T lymphocytes. Science, 225, 429-430. Spitz. M.. Gearing. A.. Callus. M., Spitz. L. & Thorpe. R. (1985) Interleukin-2 in vivo: production of and response to interleukin-2 in lymphoid organs undergoing a primary immune response to heterologous erythrocytes. Immunology. 54, 52 7-532. Thorpe. R.. Dilger. P.. Dawson. N.J. & Barrowcliffe. T.W. (1489) Inhibition of interleukin 2 secretion by factor VIII concentrates: a possible cause of immunosuppression in haemophiliacs. British Journal of Haematology, 71, 387-391. Thuy. L.T.B., Dukovich. M.. Peffer, N.J.. Fauci. A S . . Kehrl. J.H. & Greene. W.C. (1987) Direct activation of human resting T cells by IL-2: the role of an IL-2 receptor distinct from the Tac protein. Journal of Immunology, 139, 1550-1556. [Jchiyama, T.. Broder, S. & Waldmann. T.A. (1981) A monoclonal antibody (anti-tac) reactive with activated and functionally mature human T cells. I . Production of anti-tac monoclonal antibody and distribution of tac (+) cells. Journal of Immunologg. 126, 1393-1397. Wadhwa. M.. Bird. C., Tinker, A., Mire-Sluis, A. & Thorpe. R. ( 1 9 Y 1 ) Quantitative biological assays for individual cytokines. Cytokines: A Practical Approach (ed. by F. R. Balkwill), pp. 309-330. Oxford University Press.
Mechanisms of inhibition of T cell IL-2 secretion by factor VIII concentrates TUBBS,*TREVOR B A R R O W C L I F F EB,E~ R N A R D MAHONA N D ROBINTHORPEDivisions of lmrnunobiology and *Haernatology, NIBSC, South Mimms, P o t t m Bar, Herts. MEEN11 W A D H W A , P A U L A D I L G E R , J I L L
Received 5 February 7 992; accepted for publication 27 June 1992
Summary. We have continued our previous study of the inhibitory effects of factor VIII concentrates on IL-2 secretion by T cells. Experiments with an extended range of products confirm our previous conclusion that some but not all low, intermediate and high purity concentrates possess inhibitory activity on IL-2 secretion. The inhibition occurs almost immediately after addition of factor VIII concentrate and it was not possible to adsorb inhibitory activity with activated or non-activated cells: this suggests that the mechanism of inhibition involves interference with early T cell activation events rather than simple blocking of cell surface components by inhibitory molecules. The inhibitory components were shown to reside in different molecular weight fractions of concentrates. A strongly inhibitory component of approximately 200 kD and a minor species of approximately 60 kD were identified in strongly inhibitory concentrates. Some products contained a
dialysable inhibitory substance which is most likely a salt as it was also present in some formulation buffers. The proportions of the inhibitory components varied widely between products. We have found that the pattern of inhibition using in vitro systems reflects that observed using a mouse in vivo antigen challenge method. In addition we have shown that the previously reported concentrate mediated inhibition of lectin induced low affinity IL-2 receptor (CD25) is mainly a consequence of diminished IL-2 secretion rather than a ‘direct’ effect on CD2 5 expression. Considering the wide variation between products of the same purity group, caution should be exercised in drawing conclusions concerning the immunosuppressive effects of a particular type of concentrate in haemophilia patients from study with only one product from that group.
Since the occurrence of HIV-1 infection and AIDS in haemophiliacs, resulting from the use of HIV contaminated factor VIII concentrates, there has been intensive study of the immune system in both HIV seropositive and seronegative patients. It is now recognized that, in addition to the immunosuppressive effects of HIV infection, an apparently milder form of immunosuppression can occur in HIV negative haemophiliacs (Ludlam et al, 1985; Madhok et al, 1986; Madhok. 1989; Cuthbert et al, 1992). The clinical significance of this form of immunosuppression, manifested as deficient responses to skin antigen tests and involving both T-cell and B-cell dysfunction (Madhok et al. 1990, 1991a, b), remains uncertain. Although the immunological changes are thought to be related in some way to repeated alloantigen stimulation from high-dose
factor VIII therapy, the exact mechanism, and the agent(s) responsible for the immunosuppressive effects are unknown. In vitro studies have shown that factor VIII Concentrates can be potent inhibitors of lymphocyte proliferation and interleukin-2 (IL-2)secretion by T lymphocytes (Froebelef al. 1983; Lederman et al, 1986; Thorpe et al. 1989). In our previous studies we found no obvious relationship between concentrate inhibitory activity and protein concentration. Although very high purity factor VIII products were all noninhibitory, the degree of inhibition with intermediate-purity products varied widely (between 7% and 98%) at the same factor VTII concentration (Thorpe et al, 1989). In the present study we have carried out further investigations on the mechanism by which factor VIII concentrates inhibit IL-2 secretion in vitro, including analyses of the molecular weight of the inhibitory component(s). effect of concentrates on in vivo antigen stimulated IL-2 secretion, and studies of the effect of concentrates on IL-2 receptor expression.
Correspondence: Dr M. Wadhwa, Division of Immunobiology. NIBSC. Blanche Lane, South Mimms, Potters Bar, Herts. EN6 3QG.
575
(8mls of 2 x 106/ml)
(0.8mls of 42IU/ml) Incubated overnight at 37OC
Supernatants assayed for IL-2
MATERIALS A N D METHODS
Factor VIII concentrates. The details of the concentrates used in the study are: Intermediate-purity concentrates (spec. act. 0 , 5 1 0 IU/mg protein): Dry heated (80°C. 72 h ) A, from Bio-Products Laboratory, Elstree. U.K.. and B. from Scottish National Blood Transfusion Service, Edinburgh, U.K.: Solvent/detergent treated C. from Centre Regionale de Transfusion de Sang. Lille. France, D. from Octapharma GmbH. Germany, and E, from Alpha Therapeutic, Los Angeles. U.S.A. Heated in organic solvent F. from Alpha Therapeutic, Los Angeles. U.S.A. Pasteurized G. from Behringwerke. Marburg. Germany. High-purity concentrates prepared by chromatography (spec. act. 50-2 50 IU/mg protein): Solvent/ detergent treated H. from CRTS, Lille, and J, from Octapharma. Pasteurized K. from Behringwerke. Very high-purity concentrates, prepared by rnonoc~ona~ antibody absorption (spec. act. 2 3000 IU/mg protein before addition of an albumin stabilizer): Solvent/detergent treated I,, from BPL. Elstree. Pasteurized M, from Armour Pharmaceuticals, Kankakee. {J.S.A. Recombinant N, from Cutter Laboratories. Berkeley. IJ.S.A.. and P. from Baxter laboratories. 1.0s Angeles. U.S.A. Stimulation of IL-2 secretion using Irrrkat cells. The Jurkat human T cell lymphoblastoid line was maintained at a density of 10i-lOh/ml in RPMI 1640 medium containing 10% heat inactivated fetal calf serum (culture medium). For stimulation of IL-2 secretion, Jurkat cells were suspended at a density of 2 x lohcells/ml in culture medium and cultured in the presence of 1% PHA (Wellcome)with or without factor VIII concentrates in flat-sided tubes at 3 7°C in a humidified atmosphere containing 5% C02. In each assay. control cultures without PHA were included simultaneously. The supernatants were harvested after 18-24 h of culture and assayed for 1L-2 as described below. Measurement of IG2. IL-2 was measured using a bioassay based on proliferation of the IL-2 dependent mouse cytotoxic lymphocyte cell line, CTLL-2 (Gilliset al. 1 9 78: Wadhwa et al. 1991). This line was unaffected by factor VIII concentrates (Thorpe et al. 1989). For this, 50 pl of cell suspension ( 5 x lo3cells) was added to 50 pI of test samples or control medium and cultured for 18 h in 96-well microtitre plates at 3 7°C in a humidified CO2 incubator. A titration of an 11,-2 working standard calibrated
Fig 1 . Outline of adsorption experiment using unstirnulated Jurkatcells.
directly against the WHO 1st international standard for IL-2 ( 8 6 /504) was included in each assay. The cultures were then pulsed for 4 h with 3H-thymidine (18.5 kBq/well), harvested onto filter mats and the radioactivity incorporated into DNA estimated by scintillation counting. The percentage inhibition was calculated using parallel line analysis (Finney, 19 78). Dialysis of factor VlIl concentrates. Products were dialysed u s n g tubing size 1-8/32 (Medicell International Ltd) for 4 h at 4°C with three changes of phosphate buffered saline. The dialysed and non-dialysed products (at 5 IU/ml) were tested for inhibition of IL-2 secretion using the Jurkat cell-line bioassay. 4dsorption experiments. These experiments were performed in two different ways to determine if the inhibitory component could be adsorbed or internalized by Jurkat cells. ( 1 ) Unstimulated cells were used to establish if these could adsorb the inhibitory component using the protocol outlined in Fig 1. In some experiments, factor VIII concentrate was incubated with three successive batches of fresh unstimulated cells (6 ml of 2 x 10b/ml)over 3 h at 3 7°C instead of an extended overnight incubation with only one batch of cells. The supernatant (containing factor VIII concentrate) from the final incubation step was added to fresh Jurkat cells with PHA to test for remaining inhibitory effects on IL-2 secretion. (2) PHA stimulated Jurkat cells were used in an attempt to adsorb out the inhibitory component of the factor VIII concentrates. Factor VIII concentrate was added to three successive batches of stimulated cells (5 ml of 2 x lOh/rnl) as in Fig 2 . Appropriate controls were included in all the experiments. €:$feet of fuctor VIII concentrates on expression of low affinity TL-.? receptor (CD25). To further elucidate the possible mechanism of inhibition of K-2 secretion and T cell proliferation, the effect of factor VIII concentrates on IL-2 receptor expression was examined using peripheral blood mononuclear cells (PBMC). PBMCs were separated from blood by centrifugation over Ficol/Hypaque at 400 g for 30 min (Lymphoprep. Nycomed). Mononuclear cells were removed from the interface. washed three times in RPMI 1640 and resuspended to a density of 2 x lob cells/ml in culture medium. The cells were cultured with PHA in the presence or absence of factor VIII concentrates in flat-sided tubes for 72 h
FVIII Concentrate Inhibition of IL-2 Secretion
5 77
factor VIII concentrate ( 0 . 5 ~ 1 1 of~ 42IU/ml) -I-
PHA stimulated washed Jurkat cells (51111sof 2 x
I
l o 6 cells/ml)
Incubated 2hr at 37OC
Centrifugation
Discard
cells
PHA stimulated Jurkat cells (6 x
lo6)
Supernatant assayed for IL-2 activity Fig 2. Schematic diagram of adsorption experiment using stimulated Jurkat cells.
at 37OC in a C 0 2 incubator. Appropriate controls (medium or factor VIII concentrate alone) were also included in the assay. Following this period, the expression of the cell surface CD2 5 antigen was detected by incubation with an anti CD25 monoclonal antibody, anti-tac (Uchiyama et al, 1981) and FITC conjugated goat anti-mouse Ig and subsequent analysis using a FACScan flow cytometer (Becton-Dickinson).In some experiments, cells were cultured with exogenous IL-2 (100 IU/ml) and PHA in the presence or absence of factor VIII concentrates for 72 h and the expression of CD25 monitored as described above. Appropriate controls (treated cells and cells incubated with an irrelevant antibody of the same subclass as the anti-tac antibody) were included in all experiments. Effect of factor VlIl concentrates on the intermediate affinity IL-2 receptor (75 kD).This was assessed by flow cytometry using the protocol described above but with a different antibody, TU2 7 which recognizes the p7 5 IL-2 receptor chain (Ohashi et al, 1989). Fractionation of factor VIII concentrates. Four factor VIII concentrates of different purity were fractionated by gel filtration using a Sepharose CL-4B column (Pharmacia). Concentrates (total factor VIII content xl000 IU) were applied to the column and eluted overnight with TBS (50 mM Tris/l50 mM sodium chloride buffer pH 7.4) at 30 ml/h collecting 5 ml fractions. The fractions were monitored for protein by measuring their absorbance at 280 nm and any inhibitory effects on IL-2 secretion by Jurkat cells. Some fractions were further analysed by SDS-PAGE to determine the protein composition more precisely.
Stimulation of IL-2 secretion by in vivo antigenir challenge using sheep red blood cells (SRBC). Balb/c mice were injected in their hind footpads with 10 pl of a 10%suspension of sheep red blood cells. After 3 d the popliteal lymph nodes were aseptically removed, washed and the cells suspended in RPMI 1640 supplemented with 3% FCS. Viability was assessed by trypan blue exclusion, and the cells cultured at 5 x 10h/ml with or without factor VIII concentrate in flat-sided tubes at 37°C in a humidified C02 incubator. Supernatants from lymphocyte suspensions were harvested at 24 h by centrifugation and stored at - 20°C until assayed for IL-2 activity as described above.
RESULTS Inhibition of IL-2 secretion The intermediate-purity products tested in the present study represent most of the types of concentrate available currently in the U.K. and Europe and include those prepared using solvent/detergent methods of viral inactivation. Results with this group of products showed a similar diversity of inhibitory activity as in the previous study; at 1 IU/ml the inhibitory activity ranged from 30% to S5%, and at 2.5 IU/ml from 2 5% to 99% (Table I). Three of the products from the current group which had also been included in the previous study showed almost identical inhibitory activities in the two studies, demonstrating consistency of both production methods and assay methodology. There was little batch to batch variation in inhibitory activity.
Meenu Wadhwa et a1
5 78
Table 1. Inhibition of IL-2 secretion by concentrates.
( 5 -20%) was removed by dialysis. However, for product G, and for the conventional high-purity products H and J, the proportion was higher, 40-60%. These products, as well as product K, contain stabilizers such as citrate which are added to prevent factor VIII degradation during viral inactivation procedures. The formulation buffer, Y, which is used for products G and K was also tested: as shown in Fig 3 it also inhibited IL-2 secretion and most of this inhibitory activity was removed by dialysis. A s can be seen from Fig 3 , there were no concentrates in which the inhibitory activity was totally removed by dialysis, and in all cases the activity remained above 50% at the concentration tested.
Inhibition of I L 2 secretion (%I Concentrate type
2.5 IU/ml
Code ~~
Intermediate purity
C
25 82 99
u
84
A R
1 ILl!rnl
~
85 30
52 70
E
F Conventional high-purity
G
99 99
H
79
i3
J
Very-high-purity
K
90
I,
0 0
bl
Kiwtics of inhibition by factor VlIl concentrates To determine the kinetics of inhibition of IL-2 secretion, factor VIlI concentrate was added to Jurkat cells a t various time ini-ervals after stimulation with PHA. Maximum inhibition WBS produced only when factor VIII was added simultaneously with the mitogen. Delayed exposure to factor VIII for 2 h or more resulted in a progressive attenuation of the inhibitory effect on IL-2 secretion induced by PHA (Fig 4). This suggests that the inhibitory effect of factor VIII concentrate is exerted almost immediately on addition of the concentrate, i.e. IL-2 levels detected reflect the IL-2 which is secreted and accumulates in the culture medium prior to addition of inhibitory factor VIII concentrate.
41 43 32
N
0
P
0
The very-high-purity products. including the two rDNA derived concentrates. gave no inhibitory activity at any of the concentrations tested, up to 5 ll'/ml. However, somewhat surprisingly, the high-purity products prepared by conventional technology, with specific activities over 100 IC'/mg. had substantial inhibitory activity: 30-45'j& at 1 IU/ml and SO-SO% at 2 . 5 IlJ/ml (Table I). As shown in a subsequent section (see below). a considerable portion of this activity is eliminated by dialysis. Eflec'ts of dialysis on inhibition of 11,-2 secretion Because some factor VIII concentrates are stabilized with various buffers, high salt, etc., it is possible that these could influence IL-2 secretion. Accordingly we carried out a careful and thorough dialysis of several of the inhibitory products. after first optimizing the method. As shown in Fig 3 . the effects of dialysis differed considerably among the products. For most of the intermediate-purity concentrates. only a relatively small proportion of the total inhibitory activity
Adsorption of fmtor Vlll concentrates with resting or activated cr,'ls does riot eliminate the inhibitory component(s) from factor I.'M concentrates Resting or activated Jurkat cells were used in an attempt to adsorb the inhibitory components from the concentrates. In both cases the inhibitory component(s) from the concentrate could not be removed by adsorption onto cell surfaces after a n 01ernight incubation even when three successive batches of cells were used. The supernatants (containing factor VIII concentrate) obtained from both experiments were still capable of exerting a n unreduced inhibitory effect on IL-2 secretion when added to fresh Jurkat cells in the presence of PlIA (Fig 5A. B). In addition, unstimulated Jurkat cells previously incubated with the concentrate (in a n attempt to
C__i Before 2:'
Dialysis
A J FaCtorVII concenlrate Y .Albumtn C wale
"
A
B
C
O
E
F
G
H
J
Y
Fig 3. The effect of dialysis on inhibition of interleukin-2 secretion by factor VIII concentrates.
FVIll Concentrate lnhibition of IL-2 Secretion
5 79
reduction in the percentage of cells expressing the CD25 antigen was seen. This effect was dose-dependent with maximal inhibition occurring when factor VIII was added at a concentration of 4 IU/ml; the mean percentage of CD2 5 cells was 44.9f8.5 as compared to 81.5 k8.8 in the absence of the concentrate. A representative experiment is shown in Fig 7. When a less inhibitory concentrate was used, the expression of IL-2 receptor was not significantly diminished. As the decreased expression of the IL-2 receptor may be due to inhibition of IL-2 secretion, experiments were performed to test whether addition of exogenous IL-2 could reverse the reduction in IL-2 receptor expression caused by inhibitory concentrates. Addition of IL-2 (100 IU/ml) could partially reverse the inhibitory effect of factor VIII concentrate (4 IU/ ml) on IL-2 receptor expression: the mean percentage of CD25+ cells in these experiments was 66.2&2.2% (Fig 7). +
0‘ 0
5
I
I
I
10
15
25
20
Time of F.VIII Addition (Hours)
Fig 4. Time course of inhibition of IL-2 secretion from Jurkat cells by factor VIII concentrates. The data shown is from an experiment in which product F was used at a concentration of 4 IU/ml.
adsorb the inhibitory component) still released a ‘normal’ level of IL-2 into the culture medium after stimulation with PHA (Fig 6) demonstrating that the cells had not adsorbed any inhibitory substance from the concentrate.
Modulation of PHA-induced IL-2 receptor (CD25) expression by factor VllI concentrates Using flow cytofluorimetric analysis, the induction of IL-2 receptor by a n optimal concentration of PHA, as measured by the percentage of CD25+ cells, was evaluated in PBMC obtained from four individual donors. The mean percentage of CD25+ cells following stimulation with PHA was 8 1 . 5 f8 . 8 1 . In the presence of an inhibitory concentrate (i.e. one which strongly inhibits IL-2 secretion), a significant
Factor Vlll concentrates do not effect expression of the intermediate afinity ( p 7 5 ) IL-2 receptor chain Using the TU27 antibody, the effect of factor VIII concentrates on expression of the intermediate affinity IL-2 receptor on PHA induced blasts was examined. Preliminary data indicated that PHA by itself did not effect the percentage of cells expressing the p75 kD 11,-2 receptor. None of the concentrates appeared to have any significant effect on the expression of the p75 kD IL-2 receptor (data not shown). Inhibitory effect offactor VllI concentrates in an in viva antigenic stimulation system To investigate if inhibitory effects on IL-2 secretion mediated by factor VlII concentrates could also be observed following in vivo stimulation of T cells with antigen (rather than in vitro
B.
A.
50,000
80,000
PHA
8
-PHA
rADSFACVII1 PHAIFAC Vlll
e
40,000
60,000
30,000
\
E0
40,000 20,000
20,000 10,000
n 1:2
1:4
1:8
Supernatant dilution
1:16
0
1:2
1:4
1:8
1:16
Supernatant dilution
Fig 5 . Effect of adsorbed and non-adsorbed factor VIII concentrate on PHA mediated IL-2 secretion.In A. product F (4 IU/ml) was incubated with three successive batches of stimulated cells. In B, three successive batches of unstimulated cells were used to adsorb the product F (4 IU/ml). The adsorbed concentrate was then tested for effect on IL2 secretion from fresh Jurkat cells.
Meenu Wudhwa et uI
580
80.000
0
.
--=-+=_- -_
--_
1:2
b------~
1:4
1:16
1:8
1.2
1 :4
1.8
1:16
Supernatant Dilution
Supernatantdilution Fig 6. Effect of adsorbed and non-adsorbed factor VIII concentrate on PHA mediated IL-2 secretion.In this experiment.product F (4 IU/ml)
Fig 8. Etrect of two different factor VIII concentrates on inhibition of IL-i! secretion by lymph node cells from mice undergoing a primary
was incubated overnight with unstimulated cells. The symbols indicate IL-2 secretion on stimulation with PHA from: ( a ) cells previously incubated overnight with factor VIII concentrate to adsorb the inhibitory component (open square): (b) fresh cells in the presence of supernatant containing factor VIII (previouslyincubated overnight with cells to adsorb the inhibitory component) (closed triangle);(c)fresh cells in the presence offactor VIII concentrate (i.e. not used to adsorb the inhibitory component) (closed square): and (d) fresh cells in medium alone (open triangle).
apparent. Similar experiments carried out using weakly inhibitory or non-inhibitory concentrates showed that the pattern of inhibition observed using the in vivo stimulatory system was similar to that seen with Jurkat cells in v i m (Fig 8).
activation with lectins). hind footpads of mice were injected with sheep red blood cells. When cell suspensions, obtained from popliteal lymph nodes 3 d after the antigenic challenge, were incubated overnight, IL-2 was secreted into the culture medium without any in vitro stimulus. However, if a strongly inhibitory factor VTII concentrate was incubated with the lymph node cells. an inhibition of IL-2 secretion was
immune response to sheep red blood cells.
The nature of the inhibitory component(s) Despite the apparent differences in inhibitory activity, the size exclusion chromatography profile of the various products was similar and did not show any obvious differences between the inhibitory and non-inhibitory preparations (Thorpe et al, 1989). Evaluation of the fractions from inhibitory preparations for their effect on IL-2 secretion indicated that a major peak of inhibitory activity was evident in the fractions which were eluted subsequent to the void volume (containing factor VIII). A much smaller peak was also observed at a molecular weight of approximately 60 kD. These components could not be identified by SDS PAGE since there is no obvious specific band in the inhibitory fractions which is not present in non-inhibitory portions of the column profile (data not shown). In some concentrates, a strong inhibitory component eluted a t the total column volume (the expected elution point for buffer salts) (Fig 9). As expected, none of the fractions from non-inhibitory factor VIII products exerted any inhibitory effects (data not shown).
DISCUSSION
TOO
10’
lo2 FL 1
lo3
10‘
10’
10’
103
lo4
FL 1
Log (Fluorescence Intensity) Fig 7 . Effect of factor VIII concentrates on the expression of the low affinity chain (CD25) of the 1L-2 receptor using peripheral blood mononuclear cells. la) Control. (b) PHA. Ic) PHA+factor VIII concentrate (4 IU/rnl). and (d) PHA+factor VIII concentrate (4 I U j ml) + IL-2 ( 100 1U/rnl j. The data shown has been obtained from an experiment in which product F was used.
The data on inhibition of IL-2 secretion by various types of factor VIII concentrates confirms our previous observations. The degree of inhibition with the intermediate purity concentrates differed considerably among the various products, suggesting that inhibitory action was due to some specific component(s). rather than being merely a function of protein concentration. Extensive dialysis had little effect on inhibitory activity of most intermediate purity concentrates, indicating that most of the inhibitory action is due to a large molecular weight component(s).
FVIIl Concentrate Inhibition of IL-2 Secretion
58 1
0.
2
100 I
80
11.2
1
0.8
60
,E 0
aD N
0.6 40
@
8C
a
0.4
2
9 20
0.2
0
Fraction Number
Fraction Number Fig 9. Fractionation of inhibitory activity in two different factor VIII concentrates. Figures A and B represent products F and G. Factor VIII was recovered in the void volume at fraction 3 3 in figure A and at fraction 39 in figure B. Closed squares denote the inhibitory activity in the fractions and open squares denote the absorbance of the fractions at 280 nm.
It is notable that high-purity concentrates, prepared by conventional chromatography, displayed similar inhibitory characteristics to intermediate purity products. However, with these high-purity products, a much larger proportion of the activity was removed by dialysis, suggesting that salt stabilizers such as citrate were responsible for a large part of the inhibitory activity-indeed the formulation buffer for one of the products itself showed strong inhibitory action. Nevertheless, even after extensive dialysis some inhibitory activity remained in high purity concentrates, and this indicates that the protein(s) responsible had been partially, but not completely, removed by the purification process. Results of size exclusion chromatography showed that most of the inhibitory activity was located in a protein fraction of around 200 kD. We were unable to confirm the earlier results of Lederman et al (1986), who found activity also in the void volume. This fraction contains large amounts of fibrinogen and fibronectin, which account for the bulk of the protein in intermediate purity concentrates, but these are unlikely to be responsible for the inhibitory action, as they had no effect when tested as purified proteins (unpublished observation). Further identification of the inhibitory components has not proved possible to date, though clearly the activity is removed by monoclonal antibody affinity purification, since concentrates prepared from both plasma and recombinant sources by this method showed no inhibitory action. The possibility that inhibitory factor VIII concentrates contain molecules which bind specifically to cell surface components involved in mitogen mediated triggering of IL-2 secretion was investigated by attempting to adsorb such substance(s) by incubation with resting or PHA activated Jurkat cells. No reduction in inhibitory activity was observed with factor VIII concentrates adsorbed with activated or non-
activated cells implying that the inhibitory component(s) mediate their effects by mechanism(s) which do not involve a simple ‘blocking’ interaction with cell surface components. The results obtained using three separate consecutive adsorptions with large numbers of activated Jurkat cells suggest that it is unlikely that the failure to adsorb inhibitory activity was due to a low level of expression of cell-surface molecules involved in binding of inhibitory components present in factor VIII concentrates. Non-stimulated Jurkat cells incubated with an inhibitory factor VIII concentrate completely retained the ability to secrete IL-2 when stimulated by PHA if the factor VIII concentrate was removed from the culture system, which also suggests that no significant adsorption of inhibitory molecules occurs at the cell surface, at least prior to lectin activation. Experiments involving addition of inhibitory factor VIII concentrates to cultures of Jurkat cells at various intervals after stimulation with PHA showed that inhibition of It-2 secretion occurs almost instantaneously when concentrate is added. This also suggests that the mechanism of inhibition involves process(es)other than simple blocking of binding of lectin to its receptor@),or interference with early triggering events which occur soon after lectin binding to cell surface molecules. The results of several studies clearly indicate that some factor VIII concentrates inhibit lectin induced IL-2 secretion in vitru; however. there is little direct information available relating to factor VIII concentrate effects on IL-2 secretion resulting from an in vivo challenge with antigen. Our experiments described in this report using the mouse/sheep erythrocyte challenge model for in vivo induction of IL-2 secretion (Spitz et al. 1985) show that factor VIII concentrates which inhibit IL-2 secretion in vitru are also potent inhibitors of in vivo antigen triggered IL-2 secretion by mouse
582
Meenu Wadhwa et af
lymphocytes. Inhibitory activity of concentrates in the in vivo model reflected those observed with those concentrates using in vitro experiments with Jurkat cells and PHA stimulation. It can be concluded that such concentrates are likely to inhibit T cell IL-2 secretion regardless of environment or stimulus. We have confirmed a previous report (Hay et al. 1 9 9 0 )that factor VIII concentrates can inhibit the expression of the 55 kD chain of the IL-2 receptor (CD25) on the surface of PHA stimulated T cells. The capacity of different products to mediate this effect varied and seemed to correlate with ability to inhibit IL-2 secretion by Jurkat cells. It has been suggested (Hay et a!. 1990) that concentrate inhibition of IL-2 secretion occurs because CD2 5 expression 1 upregulation) is inhibited: however, it is well established that IL-2 is a potent upregulator of the 5 5 kD chain of the receptor (Reem & Yeh. 1984) and we observed that addition of rDNA derived IL-2 to cultures could at least partially overcome concentrate inhibition of PHA induced upregulation of CD2 5. It therefore seems most likely that factor VIII concentrates which inhibit expression of CD25 do this at least partially by inhibiting secretion of IL-2. which results in considerably reduced IL-2 mediated upregulation of the cell surface bound 55 kD receptor chain. However, we could not entirely reverse concentrate inhibition of PHA induced CD2S even by addition of relatively large amounts of 11,-2 and so the phenomenon may be caused by more than one mechanism. We failed to observe any effects of factor VIII concentrates on cell surface amounts of the 7 5 kD chain of the IL-2 receptor. as would be expected for the dynamics of expression of this molecule (Thuy ~t al, 1987). The relevance of these in virro studies to immune disturbances in haemophilia A patients remains unclear. Most of the studies on haemophilic patients have been done on baseline samples, i.e. from patients not receiving concentrates at the time of sampling: it is possible that injection of concentrates results in acute depression of IL-2 secretion, such as we observe in vitro. followed by gradual normalization as the inhibitory components are cleared from the circulation. Madhok et a1 (1990) found that the ability to secrete IL-2 in response to antigen was impaired in patients both negative and positive for antibodies against HIV: in a subsequent study the same group found evidence for T cell sensitization to factor VIII concentrates (Madhok et al. 1991b). and it was suggested that repeated alloantigen stimulation could lead to a partial exhaustion of the ability of lymphocytes to secrete IL-2. From the point of view of treatment of haemophilia. the two most important questions are whether the immune suppression observed in HIV negative patients has any clinical significance with respect to exposure to infectious agents, and whether there is any difference in outcome in HIV positive patients given different types of concentrates. In one comparison of a monoclonal antibody affinity purified concentrate with two intermediate purity products there were no significant differences between patients treated with the various concentrates with respect to skin anergy and CD4 counts (Brettler et a!. 1989): however, in another study patients given a monoclonal antibody affinity purified concentrate had a slower decline in CD4 counts than a group
given a n intermediate purity product (de Biasi et a!, 1991). It should be noted, however, that the intermediate purity products in these two studies were not the same, and our studies emphasize that, at least with respect to IL-2 secretion, there are wide differences between products even in the same 'purity' group. Therefore the conclusion of this latter study, that very high purity concentrates are superior to intermediate purity concentrates for treatment of HIV positive patients, should be viewed with caution.
ACKNOWLEDGMENTS We thank Bio Products Laboratory, Elstree, U.K.: Scottish National Blood Transfusion Service, Edinburgh, U.K.: Centre Regionale de Transfusion de Sang, Lille, France: Alpha Therapeutic, Los Angeles, U.S.A.; Behringwerke, Marburg, Germany: Octapharma GmbH, Germany: Armour Pharmaceuticals. Kankakee, U.S.A.: Cutter Laboratories, Berkeley, U.S.A.: and Baxter Laboratories, Los Angeles, U.S.A., for providing the factor VIII concentrates used in this study. We thank Dr Kazuo Sugamura for his generous gift of the TC;! 7 antibody, Dr Tony Mire-Sluis for helpful discussions and Deborah Kirk for preparation of the manuscript. The sup2ort for parts of this project by the Scottish National Blood Transfusion Service is greatly appreciated.
REFERENCES Bretller. D.B.. Forsberg. A.D.. Levine. P.H.. Petillo, J., Lamon, K. & Sullivan. J.L. (1989) Factor VIII: C concentrate purified from plasma using monoclonal antibodies: human studies. Blood. 73, 1k 59-1863. Cuthbert. R.J.G..Ludlam. C.A., Steel. C.M.. Beatson, 1). & Pentheres, J.F. ( 1992)Immunologicalalterationsin HIV seronegativehaemophiliacs: relationships to blood product therapy. British Journal of Hoemotology, 80, 364- 3 69. de Biasi. R.. Rocino. A.. Miraglia. E.. Mastrullo, L. & Quirino, A.A. ( 1'39 1) The impact of a very high purity factor VIII concentrate on the immune system of human immunodeficiency virus-infected
hemophiliacs: a randomized prospective, two-year comparison wi1.h an intermediate purity concentrate. Blood, 78, 1919-1922. Finncy. D.J. (1978) Statistical Methods in Biological Assay. Charles Griffin. London. Froebel. K.S.. Madhok. R.. Forbes, C.D.. Lennie. S . , Lowe, G.D.O. & StLrrock. R.D. (1 983) Immunological abnormalities in haemophilia: are they caused by American factor VIII concentrate. British Medical ]ournal. 28 7. 1091-109 3. Gillis. S.. Fern. M.M.. Ou. W.E. & Smith, K.A. (1978) T cell growth facer: parameters of production and a quantitative microassay for activity. Journal of Immunology. 120, 2027-2032. Hay. C.R.M.. McEvoy. P. & Duggan-Keen. M. (1990) Inhibition of lymphocyte ILL-receptor expression by factor VIII concentrate: a possible cause of immunosuppression in haemophiliacs. British ]otrrnal of Haematology. 75. 278-281. kderman. M.M., Saunders. C.. Toossi, Z., Lemon, N.. Everson. B. & Rainoff. O.D. (1986) Antihemophiliac factor (factor VIII) preparations inhibit lymphocyte proliferation and production of interleukin-2. Journal of Laboratory and Clinical Mcdicine. 107, 4i 1-478.
FVIII Concentrate Inhibition of IL-2 Secretion Ludlam, C.A., Tucker, J., Steel. C.M., Tedder, R.S., CheingsongPopov, R., Weiss, R.A., McClelland,D.B.. Philip, I. & Prescott. R.J. (1985) Human T-lymphotropic virus type 111 (HTLV-111)infection in seronegative haemophiliacs after transfusion of factor VIII. Lancet, ii, 233-236. Madhok, R. (1989) The in vivo and in vitro effects of clotting factor concentrates on the immune system. Replacement Therapy in Haemophilia: The Problems and Solutions (ed. by P. M. Mannucci). pp. 55-60. Medi Fax International, U.K. Madhok, R., Gracie, J.A.,Forbes, C.D. & Lowe, G.D.O. (1991a) B cell dysfunction in haemophilia in the absence and presence of HIV-1 infection. Thrombosis and Haemostasis, 65, (i), 7-10. Madhok. R., Gracie, J.A., Lowe. G.D.O., Froebel, K., Burnett, A., Follett, E. &Forbes. C.D. (1986) Impaired cell mediated immunity in haemophilia in the absence of infection with human immunodeficiency virus. British Medical Journal. 293, 978-98 1. Madhok. R.. Gracie. J.A..Smith, J., Lowe. G.D. &Forbes. C.D. (1990) Capacity to produce interleukin-2 is impaired in haemophilia in the absence and presence of HIV I infection. British Journal of Haematology, 76, 70-74. Madhok. R., Smith, J., Jenkins, A. & Lowe, G.D.O. (1991b) T cell sensitization to factor VIII in haemophilia A? British Journal of Haematology. 79, 235-238. Ohashi. Y.. Takeshita. T.. Nagata,K.. Mori,S.&Sugamura, K. (1989) Differential expression of the interleukin-2 receptor subunits, p55 and p75 on various populations of primary peripheral blood mononuclear cells. Journal of Immunology, 143, 3548-3555.
5 83
Reem, G.H. & Yeh. N. (1984) Interleukin 2 regulates expression of its receptor and synthesis of gamma interferon by human T lymphocytes. Science, 225, 429-430. Spitz. M.. Gearing. A.. Callus. M., Spitz. L. & Thorpe. R. (1985) Interleukin-2 in vivo: production of and response to interleukin-2 in lymphoid organs undergoing a primary immune response to heterologous erythrocytes. Immunology. 54, 52 7-532. Thorpe. R.. Dilger. P.. Dawson. N.J. & Barrowcliffe. T.W. (1489) Inhibition of interleukin 2 secretion by factor VIII concentrates: a possible cause of immunosuppression in haemophiliacs. British Journal of Haematology, 71, 387-391. Thuy. L.T.B., Dukovich. M.. Peffer, N.J.. Fauci. A S . . Kehrl. J.H. & Greene. W.C. (1987) Direct activation of human resting T cells by IL-2: the role of an IL-2 receptor distinct from the Tac protein. Journal of Immunology, 139, 1550-1556. [Jchiyama, T.. Broder, S. & Waldmann. T.A. (1981) A monoclonal antibody (anti-tac) reactive with activated and functionally mature human T cells. I . Production of anti-tac monoclonal antibody and distribution of tac (+) cells. Journal of Immunologg. 126, 1393-1397. Wadhwa. M.. Bird. C., Tinker, A., Mire-Sluis, A. & Thorpe. R. ( 1 9 Y 1 ) Quantitative biological assays for individual cytokines. Cytokines: A Practical Approach (ed. by F. R. Balkwill), pp. 309-330. Oxford University Press.
