- Email: [email protected]
Role of the fourth complement component (C4) in the regulation of contact sensitivity
CELLULAR
IMMUNOLOGY
119,243-25
l(l989)
Role of the Fourth Complement Component (C4) in the Regulation of Contact Sensitivity II. Qualitative Differences between C4 Molecules from High- and Low-C4 Mouse Strains FRANCESCO DIELI,
Gumo SIRECI, MARIA R. Dr PACE, AND ALFREDO SALERNO’
Institute ofGeneral Pathology, University of Palermo. 90134 Palermo, Italy Received March 16. 1988; acceptedSeptember 14, 1988 Lymph node cells collected from CBA/J mice 4 days after painting with picryl chloride induce contact sensitivity in naive recipient mice by virtue of hapten IgM immuno complexes. The immunizing capacity of these cells (“4-day” cells) is abolished after incubation of the cells with a C4dehcient guinea pig serum reconstituted with plasma or purified C4 from mice with high C4 levels (C4h), but not with plasma or purified C4 from mice with low C4 levels (C41).The inhibition of the immunizing capacity of 4day cells is due to the activation of the early components ofthe classical complement pathway which is likely to result in the solubilization of membrane-bound immunocomplexes. However, the same amounts of CBA/J and BALB/c C4 have a different effect in inhibiting the induction ofcontact sensitivity by 4day cells. In fact, by doseresponse experiments, we have found that the amount of C41 able to inhibit the induction of contact sensitivity is about threefold higher than that of C4h. Analysis of the covalent binding ability of C4h and C41reveals that C4h is able to bind to the surface of 4-day cells more efficiently than C41and this probably accounts for the difference of the two C4 molecules in inhibiting the immunizing capacity of 4-day cells. Results are discussedin terms of different reactivities of C4h and C41with the surface of 4&y cells. o 1989AC&I& RUSS, IX
INTRODUCTION Lymph node cells obtained from CBA/J mice 4 days after painting with picryl chloride (Pcl) or oxazolone (“Cday” cells) are able to induce contact sensitivity when injected into the footpads of naive recipient mice (1). This immunizing process is carried out by cell-associated IgM-hapten immuno complexes and is abolished after incubation of the cells with complement from mice with high C4 levels (C4h), but not with complement from mice with low C4 levels (C41)(2,3). The inhibitory effect of complement from mice with high C4 levels is due to the activation of the early components of the classical complement pathway (C l-3), indicating that the modification of the antigen-presenting capacity of 4-day cells might ’ To whom correspondence should be addressed at Institute of General Pathology, University of Palermo, Corso Tukory 2 it,90 134 Palermo, Italy.
243 ooo8-8749/89$3.00 Copyright 0 1989 by Academic Press,Inc. All rights of reproduction in any form reserved.
244
DIELI
ET AL.
be a consequence of the solubilization of membrane-associated immuno complexes
(4,5). In a previous paper we have reported that the presence of 4-day cells is controlled by the levels of C4; in fact, mice with high C4 levels lack 4-day cells in their lymph nodes, due to the in vivo activation of the early components of the classical complement pathway, which occurs in these mice following skin painting. In contrast, mice with low C4 levels possess4-day cells able to induce contact sensitivity and do not show complement activation following painting with Pcl(3). The aim of the present investigation is to analyze whether the inhibitory effect of high C4 serum complement is only dependent on the levels of C4 evaluated as hemolytic efficiency. In particular, we have examined whether equivalent amounts of C4 from C4h and C41mice have the same efficiency in inhibiting the antigen-presenting capacity of 4-day cells. MATERIALS AND METHODS
Mice. Male mice of the strains CBA/J, A.AL, and A/J were obtained from G. L. Bomholtgard Ltd. (Ry, Denmark). BALB/c mice were the generous gift of Dr. F. Y. Liew, the Wellcome Research Laboratories, London. The mice were bred at the Institute of General Pathology of Palermo. Eight- to twelve-week-old male mice were used in all experiments and each experimental group consisted of at least five mice. Induction of contact sensitivity by If-day cells, Donor mice were shaved on the thorax and abdomen and painted with 0.15 ml 5% ethanolic picryl chloride (Pcl, BDH, Poole, UK) on the shaved skin and forepaws. Four days later, the regional shoulder girdle and inguinal lymph nodes were removed, dissociated in medium RPM1 1640 (GIBCO, New York) supplemented with 2.5% heat-inactivated fetal calf serum (GIBCO), and washed three times; 2 X lo6 cells in 0.1 ml of the medium were injected into the footpads of each mouse; i.e., each hind footpad received 0.05 ml. Five days later, the mice were challenged by painting both sides of both ears with 1% Pcl in olive oil and the increment of ear thickness was measured at 24 hr with an engineer’s micrometer and expressedin units of 10e3cm f standard deviation (SD). Incubation of lymph nodecellswith complement.Lymph node cells ( 107)were incubated with 1 ml of normal rabbit serum diluted l/5 in veronal-buffered saline (VBS) or with 1 ml of varying dilutions of normal mouse plasma, at 37°C for 1 hr. Sera and plasmas were then collected and complement titered. Complement titrations. All reagents were prepared as described by Kabat and Mayer (6). (a) CH50. Rabbit or mouse hemolytic complement (CH50) was measured according to Cinader et al. (7). Briefly, 0. l-ml serial dilutions of serum in VBS were mixed with 0.05 ml of a suspension of sheeperythrocytes (1.25%, v/v) which had been sensitized with a large amount of rabbit IgG antibodies (EA), kindly provided by Professor A. Bellavia from this institute. The mixtures were incubated with periodic shacking for 60 min at 37°C 1 ml cold VBS was added, and the cells were collected by centrifugation. The supernatant fluids were discarded and the sedimented cells were lysed by addition of 3 ml of distilled water. (b) C4. Assays of C4 were based on the ability of samples to supply C4 to the sera of CCdeficient guinea pig (C4dGP), as described by Atkinson et al. (8) and by our-
ROLE MURINE
C4 IN CONTACT
SENSITIVITY
245
selves (2). Briefly, 0.2-ml varying dilutions of rabbit serum or mouse plasma were mixed with 0.2 ml of C4dGP serum diluted l/40 in VBS and with 0.5 ml of EA (1.5 x log/ml) for 2 hr at 37°C with frequent shacking. Three milliliters of cold VBS was then added, the mixtures were centrifuged, and the supematants were measured. In all hemolytic assays,the hemoglobin concentration in the supematants, or lysed in cells at the end of incubation period, was spectrophotometrically measured at a wavelength of 4 I2 nm in a Gilford 300 T- 1 spectrophotometer. Units per milliliter were calculated by means of the von Krogh transformation from the reciprocal of the dilution of serum giving 50% lysis. Results are expressed as percentage residual activity using the following formula: % = (E U/ml)/(C U/ml) X 100, where E represents the serum activity of the experimental group and C, the activity of serum incubated without cells. Data are expressed as mean + SD of percentage residual hemolytic activity from at least two different experiments, each carried out in triplicate. Purified complement components. A functionally pure human C 1, used in assembling complement component cellular intermediates, was prepared by a euglobulin precipitation of whole serum (9, 10). Mouse C4 was purified by affinity chromatography (11) on a column of CNBr-activated Sepharose4B (Pharmacia, Uppsala) coupled with rabbit IgG anti-human C4 (Dako, Dakopatts a/s, Denmark). This antiserum was shown to be strongly cross-reactive with mouse C4, but not with Sip, as resulted by immunoprecipitation of mouse plasma and analysis of the precipitates on SDSPAGE. Briefly, mouse plasma containing 10 mM EDTA, pH 7.4, 10 mA4 PMSF, and 5 m&f EACA was loaded onto the anti-C4 column preequilibrated with a buffer containing 50 mM Tris, 50 mM Na-phosphate, 12.5 n04 Na-tetraborate, 2.5 mM EDTA, 0.1 mM PMSF, 0.25 mM EACA, and 0.02% Na-azide, pH 7.0. The column was washed with the same buffer until the ODZgoof the eluate approached zero. C4 was eluted with buffer containing 100 mMTris, 100 mM Na-phosphate, 25 mMNatetraborate, 5 mM EDTA, 0.2 mM PMSF, 0.25 mM EACA, and 0.02% Na-azide, at pH 11.5. The eluate was tested for the presence of C4 both by hemolytic titration and by SDS-PAGE and was then concentrated to the original volume of plasma by Millipore CX- 10 membrane. The purity of the eluted material was determined by SDS-PAGE (see Fig. 1) followed by Western blotting and by this method, no contamination was detected in the affinity-purified C4 preparations. The total yield of C4 protein was 50% for BALB/c C4 and 60% for CBA/J C4. The concentration of C4 protein in plasma, as well as in purified preparations, was determined by radial immunodiffusion, using experimental conditions described by Ferreira and Nussenzweig (12) for murine C3. Specific hemolytic activities of whole plasma and affinity-purified C4 from the same donors were comparable in all cases,as shown in Table 1. Thus, the purification procedure did not affect the activity of the C4 protein. Uptake of C4 onto Cl-bearing I-day cells. This was performed according to Circolo and Borsos (13), with minor modifications. Briefly, lo7 4-day cells were incubated with 100 pg purified human Cl in 100 ~1 VBS for 30 min at 37°C. The cells were washed twice with VBS and then incubated with different concentrations of purified C4 for 30 min at 37°C. After two washes the cells were incubated with rabbit antihuman C4 at 0°C for 1 hr. The cells were washed twice and then incubated with 1 j&i ‘251-Protein A (Amersham International, UK, sp act 370 MBq/mg) for 1 hr at room temperature. After three more washes,the cells were counted for radioactivity
246
DIELI ET AL.
-116 -
97
-
66
-
45
-
31
FIG. 1. SDS-PAGE analysis of affinity-purified C4 from BALB/c (lane 1) and CBA/J (lane 2) mice. The numbers at the right side of the figure indicate the positions to which marker proteins migrated and their relative sizesin kilodaltons.
in a Packard gamma counter and the results expressed in counts per minute (cpm) f SD. Inhibition of uptake of C4 onto Cl-bearing I-day cellsby small molecules.C4 depositions were performed as described above, except that the assay buffer (VBS) contained different concentrations of fluid-phase nucleophile potentially capable of reacting with the thioester of C4 (14). Ethylacetimidate modification of I-day cells. Four-day cells’ amino groups were modified with ethylacetimidate (Fluka, Buchs, Switzerland) at a cell concentration
TABLE 1 Comparison of Hemolytic Activities of BALB/c and CBA/J C4 in Whole Plasma and Affinity-Purified Preparations Donor
Source
Hemolytic units W/ml)
C4 protein WmU
Specific activity W/Pi4
BALBfc
Plasma Purified c4 Plasma Purified c4
4329 2139 419 270
78 39 8 5
55.5 54.8 52.3 52.6
CBA/J
247
ROLE MURINE C4 IN CONTACT SENSITIVITY A Plasma Q-
---------------
C4
0 +ve control
9, -
Purified
C4
---------‘me
--- “IJtro’
--v---w--_
--
5
----
I/@
-ve
-ve ----
-
control
contr /f
-------_---____
1 OJ ,
I
h
I
50
25
12
6
p Q C4 added
FIG. 2. Inhibition of the induction of contact sensitivity by C4dGP serum reconstituted with CBA/J or BALB/c C4. Four-day lymph node cells obtained from CBA/J mice were incubated with C4dGP serum reconstituted with varying amounts of CBA/J (0) or BALB/c (0) C4. In the experiment depicted in (A), the whole mouse plasma was used as a source of C4, whereas in the experiment depicted in (B), the C4dGP serum was reconstituted with purified mouse C4.
of 109/ml in carbonate-buffered saline, pH 8.0. The suspension was made 50 m.M in ethylacetimidate and incubated at 37°C for 1 hr. The cells were then washed three times in carbonate saline and three more times in VBS (15). Statistics.The double-tailed Student’s t test was used to analyze the significance of the differences between control and experimental groups. RESULTS
Inhibition of I-Day Cells by D$erent Amounts of BALB/c and CBA/J C4 In the first experiment we investigated whether it was possible for complement from C41 mice to inhibit the induction of contact sensitivity by 4&y cells. For this purpose, the Ccdeficient guinea pig serum was reconstituted with different amounts of C4 from BALB/c or CBA/J plasma or with purified C4 from these mice. The reconstituted serum was then tested for its ability to inhibit the induction of contact sensitivity by 4-day cells. Results reported in Fig. 2A show the inhibition of contact sensitivity observed in C4-deficient serum reconstituted with varying amounts of C4 from BALB/c or CBA/J plasma. The amount of BALB/c C4 inhibiting 50% contact sensitivity was calculated to be 7.8 pg, whereas 24 pg of CBA/J C4 was necessaryto inhibit 50% contact sensitivity. Figure 2B confirms the above reported results and shows that the amount of purified C4 necessaryto inhibit 50% contact sensitivity was 6.1 pg for BALB/c and 20.2 pg for CBA/J. The ability of CBA/J C4 to inhibit the induction of contact sensitivity was accompanied by its activation; in fact, as shown in Fig. 3, incubation of 4&y cells with CCdeficient serum reconstituted with 50 and 25 pg CBA/J C4 resulted in the consumption of C4 hemolytic activity. However, reconstitution of C4-deficient serum with 12 tig of CBA/J C4, which fails to inhibit the induction of contact sensitivity, resulted in 50% consumption of C4 hemolytic
248
DIELI ET AL.
1
50
I
2.5 pg C4
1
12
6
added
FIG. 3. Consumption of C4 hemolytic activity by 4-day cells. Fourday lymph node cells were incubated with C4dGP serum reconstituted with varying amounts of purified CBA/J (0) or BALB/c (0) C4, for 1 hr at 37°C. At the end of incubation period, the serum was collected and C4 hemolytic activity titered as described under Materials and Methods. Results are expressed as mean percentage consumption of C4 hemolytic activity +- SD.
activity, thus indicating that CBA/J C4 is activated inefficiently by 4-day cells and its activation does not correlate with the inhibitory activity on 4-day cells.
Binding of CBA/Jand BAL.B/c C4 to 4-Day Cells Results reported in Fig. 2 show that the same amounts of BALB/c and CBA/J C4 differ in their ability to inhibit 4-day cell activity. The different behavior of BALB/c and CBA/J C4 might be due to a different binding of C4 to immune complexes bearing 4-day cells. To investigate this possibility 4-day cells were incubated with purified human Cl and thereafter with different amounts of BALB/c and CBA/J C4. The binding of C4 was extimated as described under Materials and Methods. Figure 4 shows that BALB/c C4 binds 4-day cells more efficiently than CBA/J C4, thus suggesting that BALB/c and CBA/J C4 molecules differ in their ability to bind immune complexes bearing 4-day cells. These results might explain the different efficiency of CBA/J and BALB/c C4 to inhibit 4-day cell activity.
Efect of Fluid-Phase Nucleophile on the Uptake of C4 onto Cl-Bearing I-Day Cells It has been previously reported that the binding of C4 to a cell surface occurs via the formation of a covalent ester bond with hydroxyl groups provided by polysaccharides ( 1l- 14). In the following experiments we investigated whether the less efficient binding of CBA/J C4 to 4-day cells might depend on its inhability to form ester bonds with hydroxyl groups. We used C 1-bearing 4-day cells in conjunction with a carbohydrate (glucose) as a hydroxyl group donor, or an amino acid (glycine) as an amino group donor, to study the relative capacities of these low-molecular-weight acceptors to inhibit the uptake of C4 to the cell membrane.
249
ROLE MURINE C4 IN CONTACT SENSITIVITY
i5 12 pg C4 added
0
5‘0
FIG. 4. Binding of CBA/J and BALB/c C4 to 4-day cells. Four-day lymph node cells were incubated with purified human Cl and then with varying amounts of CBA/J (0) or BALE/c (0) C4. The uptake of C4 onto Cl-bearing 4-day cells was estimated asdescribed under Materials and Methods. Results are expressed as cpm + SD, after subtraction of background values (C 1-bearing 4-day cells + anti-C4 antibody) of 1874 * 1192cpm.
The results of this experiment are shown in Fig. 5. Glycine (Fig. 5A) was able to inhibit the uptake of CBA/J C4 about twofold better than BALB/c C4; in contrast, glucose (Fig. 5B) was able to inhibit selectively the uptake of BALB/c C4, but failed to inhibit the uptake of CBA/J 0% These experiments indicate that CBA/J C4 reacts preferentially with amino group donors, whereas BALB/c C4 selectshydroxyl group donors, and suggestthat the less efficient binding of CBA/J C4 to 4-day cells might be due to its relative inhability to form ester bonds with hydroxyl groups present on cell membranes.
A GLYCINE
S GLUCOSE
60
60 :
I
0
1
t 0
,
30
60
30
>
60
mM inhibitor added
FIG. 5. Effect of small molecules on the binding of C4 to 4day cells. The binding of CBA/J (0) and
BALB/c (0) C4 onto Cl-bearing 4-day lymph node cells was performed as described under Materials and Methods, except that the assay buffer contained different concentrations of glycine (Fig. 5A) and glucose (Fig. 5B). Results are expressed as percentage inhibition of C4 uptake done in the absence of fluid-phase inhibitors.
250
DIELI ET AL. TABLE 2 Effect of Amino Group Modification of 4day Cells on the Uptake of BALBc and CBA/J C4 C4 from
c4 uptake on normal cells
C4 uptake on ethylacetimidate-treated cells
CBA/J BALB/c
1,489 + 423 14,275 5 1127
3,245 + 1375 (0.43) 13,133 f 1072 (0.92)
Note. Results are expressed in cpm + SD, after subtraction of background values, which were as follows: Cl-bearing 4day cells + anti-C4 = 1351 + 604 cpm; Cl-bearing ethylacetimidate-treated 4&y cells + anti-C4 = 1646 +- 792 cpm. The ratio of deposition ethylacetimidate/normal is shown in parentheses.
As an approach to further assessingthe different behavior of BALB/c and CBA/J C4, ethylacetimidate was used to chemically modify the amino groups present on the surface of 4-day cells, becausethis reagent is unreactive toward hydroxyl groups (15). As can be seen in Table 2, modification of amino groups had no appreciable effect on the deposition of BALB/c C4 on C 1-bearing 4-day cells, but resulted in the reduction in the binding of CBA/J C4. These results further support the possibility that BALB/c and CBA/J C4 display very strong preferences for hydroxyl and amino groups, respectively. DISCUSSION This investigation was prompted by our previous studies showing that lymph node cells from CBA/J mice painted with Pcl (Cday cells) lose their ability to induce contact sensitivity in naive recipient mice after incubation in serum from mice with high C4 levels but not in serum from mice with low C4 levels (2,3). In this paper we have analyzed whether the inhibitory activity of complement from high-C4 mice was due to the levels of C4, evaluated as hemolytic efficiency or other factors could play a role. We first carried out dose-response experiments in an attempt to determine the exact amount of CBA/J and BALB/c C4 required to inhibit 50% induction of contact sensitivity. Data reported in Fig. 2 show that the amount of CBA/J C4 able to inhibit 50% contact sensitivity was about threefold higher than that of BALB/c C4. Taken all together the present results indicate that the hemolytic C4 assay is evidently not sufficient to evaluate the functional activity of C4: in addition, they suggestthat the C4 molecules from high-C4 mice could be structurally or functionally different from the low-C4 mice ones. The different behavior of CBA/J and BALB/c C4 and the observation that C4 from C41mice, at least under some experimental conditions (seeFig. 3) was partially activated by 4-day cells, but failed to inhibit their immunizing capacity, prompted us to investigate whether it might be dependent on a different binding of the two C4 molecules to 4-day cells. Results reported in Fig. 4 show that BALB/c C4 binds 4day cells more efficiently than CBA/J C4, thus suggesting that the difference in the ability to inhibit the induction of contact sensitivity might be primarily due to the different binding to 4-day cells. Since Ig have lesspolysaccharides than the cell surface (14, 16) one could hypothesize that BALB/c C4, through the reactive acyl group, might form an ester bond with sugars more efficiently than CBA/J C4.
ROLE MURINE C4 IN CONTACT SENSITIVITY
251
Based on the results of the small molecule inhibition studies (Fig. 5) it appears that this is the case:in fact, BALB/c C4 reacts preferentially with hydroxyl groups of sugars, whereas CBA/J C4 fails to react with hydroxyl groups, but shows a preferential reactivity with amino groups. Furthermore, using chemical modification of amino groups to change the surface composition of 4-day cells, we have observed a selective inhibition of the covalent binding of CBA/J C4. A question which remains to be cleared is why C4 molecules from C41 mice, although hemolitically active as C4 molecules from C4h mice, are less efficient than these last ones in inhibiting 4&y cell activity. One possibility is that in the hemolytic assayfor mouse C4, IgG antibodies are used to sensitize sheep erithrocytes, whereas the antibody bound to 4-day cells is of the IgM isotype. In fact, it has been reported ( 17-l 9) that C4 binds covalently to the heavy, and at a minor extent, to the light chain of the Fab portion of IgG, and it has been speculated (18) that hemolytic activity is primarily due to C4b bound to IgG and not to that bound directly to the cell surface. Moreover, Circolo and Borsos ( 13) reported that IgM molecules on cells do not bind C4 to any measurable degreeand the only hemolytic sequenceinitiated by IgM molecules at a cell surface proceeds via cell-bound and not by &M-bound C4. We do not know at present what is the possible mechanism which determines the different reactivities of CBA/J and BALB/c C4, but the simplest explanation is that a structural difference between the two C4 molecules might be responsible. More extensive structural and biological data, however, will be needed to test this possibility. ACKNOWLEDGMENTS This work has been supported by grants from the National ResearchCouncil (CNR, Rome), the Ministry for Education (MPI 40% and MPI 60%) and the Italian Association for Cancer Research (AIRC, Milan).
REFERENCES 1. Asherson, G. L., Colizzi, V., and Watkins, M. C., Immunology48,561, 1983. 2. Salerno, A., Brai, M., Dieli, F., Colonna Romano, G., Abrignani, S., Colizzi, V., and Asherson, G. L., Cell. Immunol. 89,316, 1984. 3. Dieli, F., and Salerno, A., Cell. Immunol. 105,386, 1986. 4. Eden, A., Bianco, C., and Nussenzweig, V., Cell. Immunol. 7,459, 1973. 5. Eden, A., Bianco, C., Bogart, B., and Nussenzweig, V., Cell. Immunol. 7,474, 1973. 6. Kabat, E. A., and Mayer, M. M., “Experimental Immunochemistry.” Thomas, Springfield, IL, I96 1. I. Cinader, B., Dubinsky, S., and Wardlaw, A. C., J. Exp. Med. 120,897,1964. 8. Atkinson, J. P., MC Ginnis, R., and Shreffler, D. C., J. Immunol. Methods 33,35 1, 1980. 9. Cooper, N. R., and Muller-Eberhard, H. J., Immunochemistty5, 155, 1968. 10. Lachman, P. J., and Hobart, M. J., “Handbook of Experimental Immunology” (D. M. Weir, Ed.), Vol. 3. Blackwell, Oxford, 1978. 11. Dodds, A. W., Law, S. K. A., and Porter, R. R., EMBO J. 4,1322, 1985. 12. Ferreira, A., and Nussenzweig, V., J. Exp. Med. 141, 5 13, 1975. 13. Circolo, A., and Borsos, T., J. Immunol. 129, 1485, 1982. 14. Isenman, D. E., andYoung, J. R., J. Immunol. 132,3019,1984. 15. Isenman, D. E., and Young, J. R., J. Zmmunol. 136,2542,1986. 16. Law, S. K. A., Dodds, A. W., and Porter, R. R., EMBO J. 3,1819,1984. 17. Goers, J. W. F., and Porter, R. R., Biochem. J. 175,675,1978. 18. Campbell, R. C., Dodds, A. W., and Porter, R. R., Biochem. J. 189,67,1980. 19. Alcolea, J. M., Anton, L. C., Marques, G., Sanchez-Cotral, P., and Vivanco, F., Complement 4, 2 1, 1987.
IMMUNOLOGY
119,243-25
l(l989)
Role of the Fourth Complement Component (C4) in the Regulation of Contact Sensitivity II. Qualitative Differences between C4 Molecules from High- and Low-C4 Mouse Strains FRANCESCO DIELI,
Gumo SIRECI, MARIA R. Dr PACE, AND ALFREDO SALERNO’
Institute ofGeneral Pathology, University of Palermo. 90134 Palermo, Italy Received March 16. 1988; acceptedSeptember 14, 1988 Lymph node cells collected from CBA/J mice 4 days after painting with picryl chloride induce contact sensitivity in naive recipient mice by virtue of hapten IgM immuno complexes. The immunizing capacity of these cells (“4-day” cells) is abolished after incubation of the cells with a C4dehcient guinea pig serum reconstituted with plasma or purified C4 from mice with high C4 levels (C4h), but not with plasma or purified C4 from mice with low C4 levels (C41).The inhibition of the immunizing capacity of 4day cells is due to the activation of the early components ofthe classical complement pathway which is likely to result in the solubilization of membrane-bound immunocomplexes. However, the same amounts of CBA/J and BALB/c C4 have a different effect in inhibiting the induction ofcontact sensitivity by 4day cells. In fact, by doseresponse experiments, we have found that the amount of C41 able to inhibit the induction of contact sensitivity is about threefold higher than that of C4h. Analysis of the covalent binding ability of C4h and C41reveals that C4h is able to bind to the surface of 4-day cells more efficiently than C41and this probably accounts for the difference of the two C4 molecules in inhibiting the immunizing capacity of 4-day cells. Results are discussedin terms of different reactivities of C4h and C41with the surface of 4&y cells. o 1989AC&I& RUSS, IX
INTRODUCTION Lymph node cells obtained from CBA/J mice 4 days after painting with picryl chloride (Pcl) or oxazolone (“Cday” cells) are able to induce contact sensitivity when injected into the footpads of naive recipient mice (1). This immunizing process is carried out by cell-associated IgM-hapten immuno complexes and is abolished after incubation of the cells with complement from mice with high C4 levels (C4h), but not with complement from mice with low C4 levels (C41)(2,3). The inhibitory effect of complement from mice with high C4 levels is due to the activation of the early components of the classical complement pathway (C l-3), indicating that the modification of the antigen-presenting capacity of 4-day cells might ’ To whom correspondence should be addressed at Institute of General Pathology, University of Palermo, Corso Tukory 2 it,90 134 Palermo, Italy.
243 ooo8-8749/89$3.00 Copyright 0 1989 by Academic Press,Inc. All rights of reproduction in any form reserved.
244
DIELI
ET AL.
be a consequence of the solubilization of membrane-associated immuno complexes
(4,5). In a previous paper we have reported that the presence of 4-day cells is controlled by the levels of C4; in fact, mice with high C4 levels lack 4-day cells in their lymph nodes, due to the in vivo activation of the early components of the classical complement pathway, which occurs in these mice following skin painting. In contrast, mice with low C4 levels possess4-day cells able to induce contact sensitivity and do not show complement activation following painting with Pcl(3). The aim of the present investigation is to analyze whether the inhibitory effect of high C4 serum complement is only dependent on the levels of C4 evaluated as hemolytic efficiency. In particular, we have examined whether equivalent amounts of C4 from C4h and C41mice have the same efficiency in inhibiting the antigen-presenting capacity of 4-day cells. MATERIALS AND METHODS
Mice. Male mice of the strains CBA/J, A.AL, and A/J were obtained from G. L. Bomholtgard Ltd. (Ry, Denmark). BALB/c mice were the generous gift of Dr. F. Y. Liew, the Wellcome Research Laboratories, London. The mice were bred at the Institute of General Pathology of Palermo. Eight- to twelve-week-old male mice were used in all experiments and each experimental group consisted of at least five mice. Induction of contact sensitivity by If-day cells, Donor mice were shaved on the thorax and abdomen and painted with 0.15 ml 5% ethanolic picryl chloride (Pcl, BDH, Poole, UK) on the shaved skin and forepaws. Four days later, the regional shoulder girdle and inguinal lymph nodes were removed, dissociated in medium RPM1 1640 (GIBCO, New York) supplemented with 2.5% heat-inactivated fetal calf serum (GIBCO), and washed three times; 2 X lo6 cells in 0.1 ml of the medium were injected into the footpads of each mouse; i.e., each hind footpad received 0.05 ml. Five days later, the mice were challenged by painting both sides of both ears with 1% Pcl in olive oil and the increment of ear thickness was measured at 24 hr with an engineer’s micrometer and expressedin units of 10e3cm f standard deviation (SD). Incubation of lymph nodecellswith complement.Lymph node cells ( 107)were incubated with 1 ml of normal rabbit serum diluted l/5 in veronal-buffered saline (VBS) or with 1 ml of varying dilutions of normal mouse plasma, at 37°C for 1 hr. Sera and plasmas were then collected and complement titered. Complement titrations. All reagents were prepared as described by Kabat and Mayer (6). (a) CH50. Rabbit or mouse hemolytic complement (CH50) was measured according to Cinader et al. (7). Briefly, 0. l-ml serial dilutions of serum in VBS were mixed with 0.05 ml of a suspension of sheeperythrocytes (1.25%, v/v) which had been sensitized with a large amount of rabbit IgG antibodies (EA), kindly provided by Professor A. Bellavia from this institute. The mixtures were incubated with periodic shacking for 60 min at 37°C 1 ml cold VBS was added, and the cells were collected by centrifugation. The supernatant fluids were discarded and the sedimented cells were lysed by addition of 3 ml of distilled water. (b) C4. Assays of C4 were based on the ability of samples to supply C4 to the sera of CCdeficient guinea pig (C4dGP), as described by Atkinson et al. (8) and by our-
ROLE MURINE
C4 IN CONTACT
SENSITIVITY
245
selves (2). Briefly, 0.2-ml varying dilutions of rabbit serum or mouse plasma were mixed with 0.2 ml of C4dGP serum diluted l/40 in VBS and with 0.5 ml of EA (1.5 x log/ml) for 2 hr at 37°C with frequent shacking. Three milliliters of cold VBS was then added, the mixtures were centrifuged, and the supematants were measured. In all hemolytic assays,the hemoglobin concentration in the supematants, or lysed in cells at the end of incubation period, was spectrophotometrically measured at a wavelength of 4 I2 nm in a Gilford 300 T- 1 spectrophotometer. Units per milliliter were calculated by means of the von Krogh transformation from the reciprocal of the dilution of serum giving 50% lysis. Results are expressed as percentage residual activity using the following formula: % = (E U/ml)/(C U/ml) X 100, where E represents the serum activity of the experimental group and C, the activity of serum incubated without cells. Data are expressed as mean + SD of percentage residual hemolytic activity from at least two different experiments, each carried out in triplicate. Purified complement components. A functionally pure human C 1, used in assembling complement component cellular intermediates, was prepared by a euglobulin precipitation of whole serum (9, 10). Mouse C4 was purified by affinity chromatography (11) on a column of CNBr-activated Sepharose4B (Pharmacia, Uppsala) coupled with rabbit IgG anti-human C4 (Dako, Dakopatts a/s, Denmark). This antiserum was shown to be strongly cross-reactive with mouse C4, but not with Sip, as resulted by immunoprecipitation of mouse plasma and analysis of the precipitates on SDSPAGE. Briefly, mouse plasma containing 10 mM EDTA, pH 7.4, 10 mA4 PMSF, and 5 m&f EACA was loaded onto the anti-C4 column preequilibrated with a buffer containing 50 mM Tris, 50 mM Na-phosphate, 12.5 n04 Na-tetraborate, 2.5 mM EDTA, 0.1 mM PMSF, 0.25 mM EACA, and 0.02% Na-azide, pH 7.0. The column was washed with the same buffer until the ODZgoof the eluate approached zero. C4 was eluted with buffer containing 100 mMTris, 100 mM Na-phosphate, 25 mMNatetraborate, 5 mM EDTA, 0.2 mM PMSF, 0.25 mM EACA, and 0.02% Na-azide, at pH 11.5. The eluate was tested for the presence of C4 both by hemolytic titration and by SDS-PAGE and was then concentrated to the original volume of plasma by Millipore CX- 10 membrane. The purity of the eluted material was determined by SDS-PAGE (see Fig. 1) followed by Western blotting and by this method, no contamination was detected in the affinity-purified C4 preparations. The total yield of C4 protein was 50% for BALB/c C4 and 60% for CBA/J C4. The concentration of C4 protein in plasma, as well as in purified preparations, was determined by radial immunodiffusion, using experimental conditions described by Ferreira and Nussenzweig (12) for murine C3. Specific hemolytic activities of whole plasma and affinity-purified C4 from the same donors were comparable in all cases,as shown in Table 1. Thus, the purification procedure did not affect the activity of the C4 protein. Uptake of C4 onto Cl-bearing I-day cells. This was performed according to Circolo and Borsos (13), with minor modifications. Briefly, lo7 4-day cells were incubated with 100 pg purified human Cl in 100 ~1 VBS for 30 min at 37°C. The cells were washed twice with VBS and then incubated with different concentrations of purified C4 for 30 min at 37°C. After two washes the cells were incubated with rabbit antihuman C4 at 0°C for 1 hr. The cells were washed twice and then incubated with 1 j&i ‘251-Protein A (Amersham International, UK, sp act 370 MBq/mg) for 1 hr at room temperature. After three more washes,the cells were counted for radioactivity
246
DIELI ET AL.
-116 -
97
-
66
-
45
-
31
FIG. 1. SDS-PAGE analysis of affinity-purified C4 from BALB/c (lane 1) and CBA/J (lane 2) mice. The numbers at the right side of the figure indicate the positions to which marker proteins migrated and their relative sizesin kilodaltons.
in a Packard gamma counter and the results expressed in counts per minute (cpm) f SD. Inhibition of uptake of C4 onto Cl-bearing I-day cellsby small molecules.C4 depositions were performed as described above, except that the assay buffer (VBS) contained different concentrations of fluid-phase nucleophile potentially capable of reacting with the thioester of C4 (14). Ethylacetimidate modification of I-day cells. Four-day cells’ amino groups were modified with ethylacetimidate (Fluka, Buchs, Switzerland) at a cell concentration
TABLE 1 Comparison of Hemolytic Activities of BALB/c and CBA/J C4 in Whole Plasma and Affinity-Purified Preparations Donor
Source
Hemolytic units W/ml)
C4 protein WmU
Specific activity W/Pi4
BALBfc
Plasma Purified c4 Plasma Purified c4
4329 2139 419 270
78 39 8 5
55.5 54.8 52.3 52.6
CBA/J
247
ROLE MURINE C4 IN CONTACT SENSITIVITY A Plasma Q-
---------------
C4
0 +ve control
9, -
Purified
C4
---------‘me
--- “IJtro’
--v---w--_
--
5
----
I/@
-ve
-ve ----
-
control
contr /f
-------_---____
1 OJ ,
I
h
I
50
25
12
6
p Q C4 added
FIG. 2. Inhibition of the induction of contact sensitivity by C4dGP serum reconstituted with CBA/J or BALB/c C4. Four-day lymph node cells obtained from CBA/J mice were incubated with C4dGP serum reconstituted with varying amounts of CBA/J (0) or BALB/c (0) C4. In the experiment depicted in (A), the whole mouse plasma was used as a source of C4, whereas in the experiment depicted in (B), the C4dGP serum was reconstituted with purified mouse C4.
of 109/ml in carbonate-buffered saline, pH 8.0. The suspension was made 50 m.M in ethylacetimidate and incubated at 37°C for 1 hr. The cells were then washed three times in carbonate saline and three more times in VBS (15). Statistics.The double-tailed Student’s t test was used to analyze the significance of the differences between control and experimental groups. RESULTS
Inhibition of I-Day Cells by D$erent Amounts of BALB/c and CBA/J C4 In the first experiment we investigated whether it was possible for complement from C41 mice to inhibit the induction of contact sensitivity by 4&y cells. For this purpose, the Ccdeficient guinea pig serum was reconstituted with different amounts of C4 from BALB/c or CBA/J plasma or with purified C4 from these mice. The reconstituted serum was then tested for its ability to inhibit the induction of contact sensitivity by 4-day cells. Results reported in Fig. 2A show the inhibition of contact sensitivity observed in C4-deficient serum reconstituted with varying amounts of C4 from BALB/c or CBA/J plasma. The amount of BALB/c C4 inhibiting 50% contact sensitivity was calculated to be 7.8 pg, whereas 24 pg of CBA/J C4 was necessaryto inhibit 50% contact sensitivity. Figure 2B confirms the above reported results and shows that the amount of purified C4 necessaryto inhibit 50% contact sensitivity was 6.1 pg for BALB/c and 20.2 pg for CBA/J. The ability of CBA/J C4 to inhibit the induction of contact sensitivity was accompanied by its activation; in fact, as shown in Fig. 3, incubation of 4&y cells with CCdeficient serum reconstituted with 50 and 25 pg CBA/J C4 resulted in the consumption of C4 hemolytic activity. However, reconstitution of C4-deficient serum with 12 tig of CBA/J C4, which fails to inhibit the induction of contact sensitivity, resulted in 50% consumption of C4 hemolytic
248
DIELI ET AL.
1
50
I
2.5 pg C4
1
12
6
added
FIG. 3. Consumption of C4 hemolytic activity by 4-day cells. Fourday lymph node cells were incubated with C4dGP serum reconstituted with varying amounts of purified CBA/J (0) or BALB/c (0) C4, for 1 hr at 37°C. At the end of incubation period, the serum was collected and C4 hemolytic activity titered as described under Materials and Methods. Results are expressed as mean percentage consumption of C4 hemolytic activity +- SD.
activity, thus indicating that CBA/J C4 is activated inefficiently by 4-day cells and its activation does not correlate with the inhibitory activity on 4-day cells.
Binding of CBA/Jand BAL.B/c C4 to 4-Day Cells Results reported in Fig. 2 show that the same amounts of BALB/c and CBA/J C4 differ in their ability to inhibit 4-day cell activity. The different behavior of BALB/c and CBA/J C4 might be due to a different binding of C4 to immune complexes bearing 4-day cells. To investigate this possibility 4-day cells were incubated with purified human Cl and thereafter with different amounts of BALB/c and CBA/J C4. The binding of C4 was extimated as described under Materials and Methods. Figure 4 shows that BALB/c C4 binds 4-day cells more efficiently than CBA/J C4, thus suggesting that BALB/c and CBA/J C4 molecules differ in their ability to bind immune complexes bearing 4-day cells. These results might explain the different efficiency of CBA/J and BALB/c C4 to inhibit 4-day cell activity.
Efect of Fluid-Phase Nucleophile on the Uptake of C4 onto Cl-Bearing I-Day Cells It has been previously reported that the binding of C4 to a cell surface occurs via the formation of a covalent ester bond with hydroxyl groups provided by polysaccharides ( 1l- 14). In the following experiments we investigated whether the less efficient binding of CBA/J C4 to 4-day cells might depend on its inhability to form ester bonds with hydroxyl groups. We used C 1-bearing 4-day cells in conjunction with a carbohydrate (glucose) as a hydroxyl group donor, or an amino acid (glycine) as an amino group donor, to study the relative capacities of these low-molecular-weight acceptors to inhibit the uptake of C4 to the cell membrane.
249
ROLE MURINE C4 IN CONTACT SENSITIVITY
i5 12 pg C4 added
0
5‘0
FIG. 4. Binding of CBA/J and BALB/c C4 to 4-day cells. Four-day lymph node cells were incubated with purified human Cl and then with varying amounts of CBA/J (0) or BALE/c (0) C4. The uptake of C4 onto Cl-bearing 4-day cells was estimated asdescribed under Materials and Methods. Results are expressed as cpm + SD, after subtraction of background values (C 1-bearing 4-day cells + anti-C4 antibody) of 1874 * 1192cpm.
The results of this experiment are shown in Fig. 5. Glycine (Fig. 5A) was able to inhibit the uptake of CBA/J C4 about twofold better than BALB/c C4; in contrast, glucose (Fig. 5B) was able to inhibit selectively the uptake of BALB/c C4, but failed to inhibit the uptake of CBA/J 0% These experiments indicate that CBA/J C4 reacts preferentially with amino group donors, whereas BALB/c C4 selectshydroxyl group donors, and suggestthat the less efficient binding of CBA/J C4 to 4-day cells might be due to its relative inhability to form ester bonds with hydroxyl groups present on cell membranes.
A GLYCINE
S GLUCOSE
60
60 :
I
0
1
t 0
,
30
60
30
>
60
mM inhibitor added
FIG. 5. Effect of small molecules on the binding of C4 to 4day cells. The binding of CBA/J (0) and
BALB/c (0) C4 onto Cl-bearing 4-day lymph node cells was performed as described under Materials and Methods, except that the assay buffer contained different concentrations of glycine (Fig. 5A) and glucose (Fig. 5B). Results are expressed as percentage inhibition of C4 uptake done in the absence of fluid-phase inhibitors.
250
DIELI ET AL. TABLE 2 Effect of Amino Group Modification of 4day Cells on the Uptake of BALBc and CBA/J C4 C4 from
c4 uptake on normal cells
C4 uptake on ethylacetimidate-treated cells
CBA/J BALB/c
1,489 + 423 14,275 5 1127
3,245 + 1375 (0.43) 13,133 f 1072 (0.92)
Note. Results are expressed in cpm + SD, after subtraction of background values, which were as follows: Cl-bearing 4day cells + anti-C4 = 1351 + 604 cpm; Cl-bearing ethylacetimidate-treated 4&y cells + anti-C4 = 1646 +- 792 cpm. The ratio of deposition ethylacetimidate/normal is shown in parentheses.
As an approach to further assessingthe different behavior of BALB/c and CBA/J C4, ethylacetimidate was used to chemically modify the amino groups present on the surface of 4-day cells, becausethis reagent is unreactive toward hydroxyl groups (15). As can be seen in Table 2, modification of amino groups had no appreciable effect on the deposition of BALB/c C4 on C 1-bearing 4-day cells, but resulted in the reduction in the binding of CBA/J C4. These results further support the possibility that BALB/c and CBA/J C4 display very strong preferences for hydroxyl and amino groups, respectively. DISCUSSION This investigation was prompted by our previous studies showing that lymph node cells from CBA/J mice painted with Pcl (Cday cells) lose their ability to induce contact sensitivity in naive recipient mice after incubation in serum from mice with high C4 levels but not in serum from mice with low C4 levels (2,3). In this paper we have analyzed whether the inhibitory activity of complement from high-C4 mice was due to the levels of C4, evaluated as hemolytic efficiency or other factors could play a role. We first carried out dose-response experiments in an attempt to determine the exact amount of CBA/J and BALB/c C4 required to inhibit 50% induction of contact sensitivity. Data reported in Fig. 2 show that the amount of CBA/J C4 able to inhibit 50% contact sensitivity was about threefold higher than that of BALB/c C4. Taken all together the present results indicate that the hemolytic C4 assay is evidently not sufficient to evaluate the functional activity of C4: in addition, they suggestthat the C4 molecules from high-C4 mice could be structurally or functionally different from the low-C4 mice ones. The different behavior of CBA/J and BALB/c C4 and the observation that C4 from C41mice, at least under some experimental conditions (seeFig. 3) was partially activated by 4-day cells, but failed to inhibit their immunizing capacity, prompted us to investigate whether it might be dependent on a different binding of the two C4 molecules to 4-day cells. Results reported in Fig. 4 show that BALB/c C4 binds 4day cells more efficiently than CBA/J C4, thus suggesting that the difference in the ability to inhibit the induction of contact sensitivity might be primarily due to the different binding to 4-day cells. Since Ig have lesspolysaccharides than the cell surface (14, 16) one could hypothesize that BALB/c C4, through the reactive acyl group, might form an ester bond with sugars more efficiently than CBA/J C4.
ROLE MURINE C4 IN CONTACT SENSITIVITY
251
Based on the results of the small molecule inhibition studies (Fig. 5) it appears that this is the case:in fact, BALB/c C4 reacts preferentially with hydroxyl groups of sugars, whereas CBA/J C4 fails to react with hydroxyl groups, but shows a preferential reactivity with amino groups. Furthermore, using chemical modification of amino groups to change the surface composition of 4-day cells, we have observed a selective inhibition of the covalent binding of CBA/J C4. A question which remains to be cleared is why C4 molecules from C41 mice, although hemolitically active as C4 molecules from C4h mice, are less efficient than these last ones in inhibiting 4&y cell activity. One possibility is that in the hemolytic assayfor mouse C4, IgG antibodies are used to sensitize sheep erithrocytes, whereas the antibody bound to 4-day cells is of the IgM isotype. In fact, it has been reported ( 17-l 9) that C4 binds covalently to the heavy, and at a minor extent, to the light chain of the Fab portion of IgG, and it has been speculated (18) that hemolytic activity is primarily due to C4b bound to IgG and not to that bound directly to the cell surface. Moreover, Circolo and Borsos ( 13) reported that IgM molecules on cells do not bind C4 to any measurable degreeand the only hemolytic sequenceinitiated by IgM molecules at a cell surface proceeds via cell-bound and not by &M-bound C4. We do not know at present what is the possible mechanism which determines the different reactivities of CBA/J and BALB/c C4, but the simplest explanation is that a structural difference between the two C4 molecules might be responsible. More extensive structural and biological data, however, will be needed to test this possibility. ACKNOWLEDGMENTS This work has been supported by grants from the National ResearchCouncil (CNR, Rome), the Ministry for Education (MPI 40% and MPI 60%) and the Italian Association for Cancer Research (AIRC, Milan).
REFERENCES 1. Asherson, G. L., Colizzi, V., and Watkins, M. C., Immunology48,561, 1983. 2. Salerno, A., Brai, M., Dieli, F., Colonna Romano, G., Abrignani, S., Colizzi, V., and Asherson, G. L., Cell. Immunol. 89,316, 1984. 3. Dieli, F., and Salerno, A., Cell. Immunol. 105,386, 1986. 4. Eden, A., Bianco, C., and Nussenzweig, V., Cell. Immunol. 7,459, 1973. 5. Eden, A., Bianco, C., Bogart, B., and Nussenzweig, V., Cell. Immunol. 7,474, 1973. 6. Kabat, E. A., and Mayer, M. M., “Experimental Immunochemistry.” Thomas, Springfield, IL, I96 1. I. Cinader, B., Dubinsky, S., and Wardlaw, A. C., J. Exp. Med. 120,897,1964. 8. Atkinson, J. P., MC Ginnis, R., and Shreffler, D. C., J. Immunol. Methods 33,35 1, 1980. 9. Cooper, N. R., and Muller-Eberhard, H. J., Immunochemistty5, 155, 1968. 10. Lachman, P. J., and Hobart, M. J., “Handbook of Experimental Immunology” (D. M. Weir, Ed.), Vol. 3. Blackwell, Oxford, 1978. 11. Dodds, A. W., Law, S. K. A., and Porter, R. R., EMBO J. 4,1322, 1985. 12. Ferreira, A., and Nussenzweig, V., J. Exp. Med. 141, 5 13, 1975. 13. Circolo, A., and Borsos, T., J. Immunol. 129, 1485, 1982. 14. Isenman, D. E., andYoung, J. R., J. Immunol. 132,3019,1984. 15. Isenman, D. E., and Young, J. R., J. Zmmunol. 136,2542,1986. 16. Law, S. K. A., Dodds, A. W., and Porter, R. R., EMBO J. 3,1819,1984. 17. Goers, J. W. F., and Porter, R. R., Biochem. J. 175,675,1978. 18. Campbell, R. C., Dodds, A. W., and Porter, R. R., Biochem. J. 189,67,1980. 19. Alcolea, J. M., Anton, L. C., Marques, G., Sanchez-Cotral, P., and Vivanco, F., Complement 4, 2 1, 1987.