Modulation of the secondary antibody response of murine lymphocytes to sheep red blood cells in vitro by neuraminidase and exoglycosidases

Immunology Letters, 29 (1991) 235 - 240 Elsevier

IMLET 01637

Modulation of the secondary antibody response of murine lymphocytes to sheep red blood cells in vitro by neuraminidase

and exoglycosidases

M a k o t o N a k a m u r a , Tomoak i Yoshida, Ken-ichi Isobe, Takashi Iwamoto , S. M. J a m s h e d u r R a hm a n , Yue hua Zhang , Tadao Hasegawa, Masa tosh i Ich ihara and Izumi N a k a s h i m a

Department of Immunology, Nagoya University School of Medicine, Nagoya, Japan

(Received 26 March 1991; accepted 12 April 1991)

1. Summary

The effect of digestion of lymphocytes with neuraminidase and exoglycosidases on the secondary antibody response in vitro to sheep red blood cell (SRBC) antigen was tested. Treatment of spleen ceils from SRBC-primed mice with 3/~g/ml of neuraminidase slightly but significantly augmented their plaque-forming cell response to SRBC, whereas treatment with 100 tzg/ml of a mixture of exoglycosidases did not. Rather unexpectedly, however, treatment of the spleen cells with the mixture of both neuraminidase and exoglycosidases greatly augmented the response. This enzyme action was substrate specific inasmuch it was ablated by addition of mucin as a neuraminidase inhibitor to the enzyme mixture. The target of the enzyme activity was not glass-adherent macrophages, but was glass-non- adherent suppressor cells in the antigen-primed cell population. Evidence was provided that the pheno- type of suppressor cells whose activity was ablated by the enzyme treatment was Thy-I +. It is suggested from these results that sialilated complex type oligosaccharides on antigen-primed T cells play a critical role in their suppressor activity.

Key words: Neuraminidase; Exoglycosidase; Cell surface carbohydrate; Secondary antibody response; Suppressor T cell

Correspondence to." Dr. lzumi Nakashima, Department of Im- munology, Nagoya University School of Medicine, 65 q-surumai- cho, Showa-ku, Nagoya, Aichi 466, Japan.

2. Introduction

The surface of lymphocytes is covered with carbohydrates in the form of glycoproteins or glycolipids, but the functions of these carbohydrates largely re- main to be understood. Potentially important roles of cell surface carbohydrates have long been suggested in cell-cell interactions, and direct evidence of the contribution of carbohydrate-receptor interaction to the immunological events has recently been provided by the discovery of a family of lectin adhesion molecules [1 - 3]. However, we still have little knowl- edge about the roles of such carbohydrate-lectin interactions in cell-to-cell communication for lymphocyte functions. Addition of monosaccharides [4-9] , disaccharides [10] or glycopeptides [11] to cultures of the cells in the immune system or treatment of the cells before culture with tunicamycin [12, 13], an inhibitor of N-linked oligosaccharide synthesis, has been shown to reduce the action of cytotoxic T lymphocytes [I1, 13], NK cells [81 and suppressor T cells [6, 9], and proliferative T cell [12, 4] and antibody [5] responses. These treatments were supposed to inhibit the interaction of carbohydrates with putative receptors in the lectin family. However, addition of sugars may inhibit the cell metabolism nonspecifically [7, 8], and tunicamycin may be cytotoxic. The use of enzymes for digesting the cell surface carbohydrates is an alternative method to test the role of each of different oligosaccharides in the immunological events [14-17]. Treatment of B cells or T cells with neuraminidase augmented the response to Ia al- loantigens on B cells [15, 17], suggesting a role for

0165-2478 / 91 / $ 3.50 © 1991 Elsevier Science Publishers B.V. 235

sialic acid or sialic acid-hidden structures in cell interaction. Using this technique we have recently demonstrated that a unique immunogenicity of murine Thy-1 directly stimulating allospecific B cells for anti-Thy-1 IgM antibody response [18] is under strict control by cell surface carbohydrates; we suggest from the results that complex type and high- mannose type N-linked oliogosaccharides play differential roles [19]. Here an attempt was made to see whether digestion of lymphocytes with sialidases and exoglycosidases would modify their antibody response in vitro to sheep red blood cells (SRBC). The results show that such treatment does affect the secondary antibody response of previously antigen- primed spleen cells, primarily reducing suppressor Thy-I ÷ lymphocyte activity.

3. Materials and Methods

3.1. Mice

C 3 H / H e N strain of mice, 8 -12 weeks old, bred at the Institute of Animal Research, Nagoya Univer- sity School of Medicine, were used.

3.2. Cell preparation and culture

Mice were given an intraperitoneal injection of 108 sheep red blood cells (SRBC) for priming. The spleen was extracted 7 - 10 days later, and spleen cell suspension in Eagle's minimal essential medium (MEM) was prepared. The SRBC-immune spleen cells suspended finally in RPMI-1640 medium sup- plemented with glutamic acid, Hepes, 5 × 10 -5 2ME and 1007o FCS (10 6 cells/200/zl/well) were cultured in 96-well plastic dishes [20] under the conditions of 5070 CO2 in air at 37°C, in the presence or absence of SRBC (106/well except where otherwise noted) as antigen, or LPS (10/~g/ml) prepared from Klebsiella pneumoniae 03 strain Kasuya as a potent polyclonal B cell activator [21]. PFC responses to SRBC and LPS were assayed on day 3 in case of SRBC and on day 2 in case of LPS when the responses peaked. For some experiments, the spleen cell suspension in 1007o FCS/RPMI was incubated on the tissue culture plate for 1 h, and non-adherent ceils were removed. The plate was washed 3 times with medium to prepare the adherent cell monolayer sheet. The spleen cell suspension in 1007o FCS/RPMI (106 cells/ml) was also

236

incubated on glass petri dishes for 1 h to eliminate glass-adherent cells; to collected non-adherent cells were again incubated on fresh dishes for another 30 min to prepare non-adherent cells. Contamina- tion of phagocytic macrophages in non-adherent cell fraction was <0.1070. Peritoneal exudate ceils were collected from mice that had received 0.5 ml of Bayol 10 days before, and washed twice with medium before use. Some spleen cells were treated with anti- Thy-1 mAb and complement to eliminate Thy-1 ÷ T lymphocytes as described [22]. This treatment spe- cifically killed > 9007o Thy-1 ÷ thymocytes.

3.4. Treatment with enzymes

Except where otherwise noted, spleen cells for culture (107 cells/ml) of peritoneal macrophage sheets were treated with Clostridium perfringens neuraminidase (3/~g/ml; Sigma Chem. Co., St. Louis, MO) or a mixture of exoglycosidases from Charonia lampas and Turbo cornutus (Seikagaku Kogyo, Co. Ltd., Tokyo; 100 ~g/ml) in MEM, pH 7.0, for 60 min. The latter mixture contained (x-N-acetyl- galactosaminidase (C.L., 0.4 units/mg), ~-N- acetylhexosaminidase ( C i . , 2.8 units/mg), ~N- acetylhexosaminidase (T.C., 50 units/mg), ~L- fucosidase ( C i . , 4.5 units/mg), ~galactosidase ( C i . , 1.1 units/mg) and c~-mannosidase (T.C., 4.3 units/mg). Preliminary experiments showed that these doses of enzymes were not cytotoxic to spleen cells under examination by trypan blue dye exclusion test. Mucin (1 mg/ml; Sigma) was used as an inhibitor of neuraminidase.

3.5. Plaque-forming-cell assay

Anti-SRBC antibody response was monitored by PFC assay as described [20]. Briefly, suspension of cultured spleen cells (100/zl) was mixed with 50°70 SRBC suspension (15 /~1) and guinea pig complement (15/zl), and the mixture was incubated in Cun- ningham's chambers. Total PFC numbers in control spleen cell cultures were variable from experiment to experiment because of unavoidable fluctation of culture conditions. But repeated experiments in which total PFC numbers varied much showed that the ratio of values of different experimental groups were stable. All experiments were done at least 3 times with reproducible results, but only the result

of the representative was presented.

3.6. Statistics

The geometric mean o f quadrupl icate assays with s tandard deviation was presented for each experimental group. Significance o f difference be- tween the two values were statistically tested by Stu- dent 's t-test.

4. Results

Fig. 1 shows the representive o f the results o f experiments repeated over 10 times, in which the effect o f t reatment o f immune spleen cells with various concentrat ions o f neuraminidase and exoglycosidases on their secondary an t ibody responses to SRBC were tested. Treatment with exoglycosidases alone failed to affect the an t ibody responses significantly, whereas the an t ibody responses were augmented significantly by treatment with neuramini-

dase alone. The latter result may correspond to the earlier observat ion that t reatment o f B cells or T cells with neuraminidase augmented the T cell-B cell interaction [16, 17]. Rather unexpectedly, however, the treatment with both neuraminidase and exoglycosidases extensively promoted the responses. All the responses to 104 to 3 × 106 SRBC were greatly augmented by this treatment, a l though the reduced response to an excessively high dose such a s 10 7 SRBC was not restored by the enzyme treatment (Fig. 2). Augmenta t ion was observed during the whole peri- od (from day 2 to day 5) o f the an t ibody responses, and the synergistic actions were observed with various concentrat ions o f neuraminidase and exoglycosidases, a l though the op t imum concentra- tions were 3 #g /ml for neuraminidase and 100/zg/ml for exoglycosidases (data not shown). Viable cell numbers determined before or after cell culture were not changed significantly by any o f these treatments. Substrate specificity o f the neuraminidase action was conf i rmed by the successful

Treatment of responder spleen cells PFC/Culture 100 200 300 400 500

, i , i J i i i

None

Exoglycosidases ~ ~."~ ~'~ '~: . ~ P< 0.001 j P ' < 0.001

Neuraminidase ~ ......... ,>,~,,~ P< o.001 ~ .~ ~,.'.~'~. •

Neuraminidase + Exoglycosidases ~ , ' . " ~ i ~ g ~'i:!ii-~!?'~?-~!i!i~7~'i~,~

Fig. 1. Effect of the prior treatment of SRBC-immune spleen cells with neuraminidase and exoglycosidases on their secondary PFC responses to SRBC in vitro. The enzyme-treated and control untreated spleen cells were stimulated with SRBC in vitro for secondary

anti-SRBC PFC response. Sheep

red blood cells

(No, / well)

0

10 4

10 5

10 6

3X10 6

10 7

Treatment (Neuraminidase

+ Exoglycosidases) 0

(-) (+) (-) (+) (-) (+)

(-) (+)

(-) I (+) (-) (+)

PFC / Culture 50 100 150 200 250

i i i

P< 0.02

• } - P< 0.002

~ : ~ < 0.005

Fig. 2. Effect of the enzyme treatment of spleen ceils on the PFC responses to various amounts of antigen. The enzyme-treated and untreated control SRBC-immune spleen cells were stimulated with different numbers of SRBC in vitro for secondary anti-SRBC PFC response.

237

blocking of the augmenting effect with mucin (data not shown).

Study was then conducted to identify the target cell of the enzyme treatment. Although data are not included in the figures, the B cell response to LPS for generation of anti-SRBC PFC was not augmented significantly by the enzyme treatment, suggesting that the B cells were not the direct target. As shown in Fig. 3A, the treatment of non-adherent spleen cells but not adherent spleen cells with the two enzymes was effective for augmenting the response. Addition of a small amount of Bayol-stimulated peritoneal macrophages into the culture augmented the response of spleen cells, whereas addition of a large amount suppressed it. As also shown in Fig. 3B, the enzyme treatment neither augmented the response-promoting activity of macrophages nor eliminated their activity to suppress the response. These results showed that the target of the enzyme treatment was not adherent accessory ceils.

Fig. 4 shows that untreated primed cells but not untreated unprimed cells suppress the response of

the enzyme-treated primed cells. This suggested that the enzyme treatment eliminated some suppressor activity of the primed spleen cells which had been increased after priming. We therefore tested whether the supressor cells in our system carry Thy-1 as a marker o f T cells. Since both helper T cells and putative suppressor T cells should be Thy-I +, the treatments of cells with enzymes and anti-Thy-1 mAb were expected to provide a mixture of helper and suppressor T cells in different ratios and a constant amount of memory B lymphocytes. As shown in Fig. 5, only the groups from which Thy-1 + suppressor cells were eliminated showed high antibody response, and the levels of antibody responses of in- dividual groups were just proportional to those expected from the ratios of Thy-I + helper and Thy- l + suppressor T cells. Repeated experiments showed basically the same result.

5. Discussion

This study showed that the treatment of previous-

A) Fraction of spleen cells treated with

Neuraminidase + Exoglycosidases

None

Adherent

Non-adherent

Adherent + Non-adherent

B) No, of peritoneal Treatment

macrophages added (Neuraminidase +

into the culture of Exoglycosidases) spleen cells

( ) 10 4

(+)

( - ) 10 5

(+)

PFC / Culture 0 40 80 120 160 I " '

I ~.: .~.~;} i~ i i~!~~;:~~!~i i~: : . i:: i .;~::::l ,

0 50 100

P F C / C u l t u r e

150 200

Fig. 3. The enzyme-treatment of non-adherent but not adherent spleen or peritoneal cells is effective for modification of PFC response. (A) The adherent and non-adherent fractions of SRBC-immune spleen cells were treated with enzymes prior to reconstitution of both fractions for in vitro stimulation with SRBC. (B) SRBC-immune spleen cells were mixed with indicated numbers of enzyme-treated or

untreated control peritoneal exudate cells, and were stimulated with SRBC in vitro for PFC response.

238

Spleen cells (x l06 /we l l ) cultured (Treated or untreated with

Neuraminidase + Exoglycesidases)

Untreated primed (1)

Treated primed (1)

Treated primed (0.5) + Untreated unprimed (0.5)

Treated primed (0.5) + Untreated primed (0.5)

Treated primed (0.5) + Treated unprimed (0.5)

PFC / Culture 20 40 60 80

~ 1 I I I

I

i~iiiiiP ~i~i~!i~i: '~ :::i:~T '~: i : 1 I } ~< o.oo5

..................... ~ J- p< 0.00,

Fig. 4. The enzyme treatment abolishes the suppressor cell activity of SRBC-primed spleen cells. The indicated numbers of enzyme-treated or untreated SRBC-immune or normal spleen cells were mixed and stimulated with SRBC for anti-SRBC PFC response.

Primed spleen cells (xlO6/well) cultured (Treated or untreated with

Neuraminidase + Exoglycosidases)

Untreated (1)

Enzyme treated (1)

Enzyme treated (0.5) + Untreated (0.5)

Enzyme treated (0.5) + Anti Thy-1 treated (0.5)

Expected levels of helper (h) and suprressor (s)

PFC / Culture T cell activities 0 20 40 60 80 100 120

h + S I, I I I . , ;

I - P < 0.001 h ~i~i!i~iiii~iiiii~iii(iiiiiiiiiiii{~{~iiii~i!!iiiiiii~iii~!~iiii~i~iii~ii~{ii{ii~i{~iii~iH 7 - P< o.oo5

h + S/2 i i i i H

~ P< 0.005 Enzyme and anti Thy-1 treated (0.5) + Untreated (0.5) h/2 + s/2

I .

Enzyme and anti Thy-1 treated (0.5) + Anti Thy-1 treated (0.5) x ~l

Fig. 5. The suppressor cells whose activity is abolished by the enzyme-treatment are Thy-1 +. The indicated numbers of enzyme-treated or untreated SRBC-immune spleen cells that had been treated or untreated with anti-Thy-I mAh and C were mixed at various combinations, and the mixtures were stimulated with SRBC for anti-SRBC PFC response. The expected levels of helper (h) and suppressor (s) T cell activities in each mixture of cells are shown by numbers.

ly SRBC-primed spleen cells with two different enzymes such as neuraminidase and glycosidase prepared an augmented secondary antibody response to SRBC. The data suggested that the phenomenon did not result from direct activation of B cells, helper T cells or accessory adherent cells (macrophages) or elimination of the suppressor adherent cells. Evi- dence was provided that Thy-1 + T cells functionally displaying suppressor activity were the target of the enzyme treatment. The result suggested that the complex type oligosaccharides whose sialic acid at the non-reducing end was eliminated by neuraminidase were primarily involved in the observed suppression. High-mannose type oligosaccharides do not seem to be active, because exoglycosidase alone that contained c~-mannosidase was not effective. Koszinowiki and Kramer [6] and Tosato [91 reported that the suppressor T cell function, which would pre-

vent the generation of alloreactive CTL in MLC [6] and of immunoglobulin-secreting cells in mitogen- stimulated culture [9], was inhibited by addition of c~-D-mannose or methyl-~-D-mannopyranoside to the culture, suggesting the involvement of high- mannose type oligosaccharides. These earlier obser- vations do not agree with our present conclusion that complex type oligosaccharides are primarily involved in the suppression. This discrepancy may be due to differences in the experimental procedure; however, addition of monosaccharides such as c~-D- mannose has also been shown to inhibit immunological events nonspecifically [7, 8].

After extensive studies on suppressor T cells [23], evidence of their cell type independence is still lack- ing. Our present results nevertheless suggest that some Thy-1 + T cells present in previously antigen- primed, but not in unprimed mice had both suppres-

239

sor activity in vitro and exceptionally high suscepti- bility to treatment with neuraminidase plus exoglycosidases. This does not necessarily mean that the T cells displaying the suppressor activity are an independent subpopulation. It may also be the case that too many helper T cells, possibly expressing special carbohydrates, caused suppression and partial deletion of these cells by the enzyme treatment increased the response. We do not know the possible role of the carbohydrates in the suppressor T cell activity. One likely, though not proven, explanation is that T cells bearing special carbohydrates interact with other T cells or B cells through lectin- carbohydrate binding, and such interaction is linked to antigen-receptor interaction for suppression.

Acknowledgements

This study was supported in part by a Grant-in- Aid from the Ministry of Education, Science and Culture of Japan, and by the Toyota Foundation.

References

[1] ~dnoc k , T. A., Butcher, E. C., Stoolman, L. M. and Rosen, S. D. (1987) J. Cell. Bio[. 104, 725.

[2] Johnson, G. I., Cook, R. G. and McEver, R. P. (1989) Cell 56, 1033.

[3] Lasky, L. A., Singer, M. S., Yednock, T. A., Dowbenko, D., Fennie, C., Rodriguez, H., Nguyen, T., Stachel, S. and Ros- en, S. D. (1989) Cell 56, 1045.

[4] Muchmore, A. V., Decker, J. M. and Blaese, R. M. (1980~ J. [mmunol. 125, 1306.

[5] Tomaska, L. D. and Parish, C. R. (1981} Eur. J. lmmunol. 11, 181.

[6] Korzinowski, U. K. and Kramer, M. (1981) Nature 289, 181. [7] Ulrich, E (1983) Cell. Immunol. 80, 241. [8] Stutman, O., Dien, P., Wisun, R. E. and Lattime, E. C.

(1980) Proc. Nat. Acad. Sci. USA 77, 2895. [9] Tosato, G., Pike, S. and Blaese, R. M. (1983) J. Exp. Med.

158, 1048. [10] Muchmore, A. V., Decker, J. M., Blaese, R. M. and Nilsson,

B. (1984) J. Exp. Med. 160, 1672. [ l l] Pilmore, N. J. G. and Miller, R. G. (1984) J. Immunol. 133,

1763. [12] Hart, G. W. (1982) J. Biol. Chem. 257, 151. [13] Black, P. L., Vitetta, E. S., Forman, J., Kang, C.-Y., May,

R. D. and Uhr, J. W. (1981) Eur. J. Immunol. 11, 48. [14] Katoh, Y., Ono, S., Yakahama, Y., Miyake, K. and Hamaoka.

T. (1986) J. lmmunol. 37, 2871. [15] Cowing, C. and Chapdelaine, J. M. (1983) Proc. Natl. Acad.

Sci. USA 80, 6000. [16] Powell, L. D., Whiteheart, S. W. and Hart, G. W. (1987) J.

lmmunol. 139, 262. [17] Taira, S., Kakiuchi, T., Minami, M. and Nariuchi, H. (1986)

J. Immunol. 139, 262. [18] Isobe, K., Nakashima, I., Nagase, E, Kato, N., Mizoguchi,

K., Kawashima, K. and Lake, P. (1984) J. Immunol. 132, ll00.

[19] Isobe, K., ~bshida, T., Iwamoto, T. and Nakashima, 1. (1989) Immunol. Lett. I, 90.

[20] Nakashima, I., Nagase, F., Yokochi, T., Kojima, T., Ohata, M. and Kato, N. (1976) Immunology 31, 649.

[21] Nakashima, I., Ohta, M., Kido, N., Fujii, Y., ~tbkochi, T., Nagase, E, Hasegawa, T., Mori, M., Isobe, K., lVlizoguchi, K., Saito, M., Kato, N. and Kasai, N. (1984) in: Bacterial En- dotoxin, Chemical, Biological and Clinical Aspects (Hom- ma, J.Y. et al., Eds.), pp. 173-185, Verlag Chemie, Weinheim.

[22] Nakashima, 1., Mizoguchi, M., Kato, N., Nagase, E, lsobe, K., Saito, M. and Suzuki, K. (1982)Eur. J. lmmunol. 12,713.

[23] Tada, T. and Okumura, K. (1979) Ad~. Immunol. 28, 1.

240

Documents

Modulation of the secondary antibody response of murine lymphocytes to sheep red blood cells in vitro by neuraminidase and exoglycosidases