o
-FgVfg
Immunology Today, Vol. 11, No. 1 1990
$
eta/. (1986) Nature 324, 163-166 Todd. J.A., Belt. J.I. and McDevitt, HO. (1987) Nature 329, 599-604 91 Horn, G.T.0 Bugawan, T.L., Long, C.M. and Erlich, H.A. (1988) Proc Natl Acad. Sci. USA 85, 6012-6016 Auffray, C., LJiiie, J.W., Arnot, D. et aL (1984) Nature 308, 327-333 Chang, HC., Monuchi. T. and Silver, J. (1983) Nature 305, 813-815 94 Monuchi, J., Moriuchi, T. and Silver, J. (1985) Proc. Natl Acad. Sci. USA 82, 3420-3424 Auftray, C.. Korman. AJ., RouxDesseto, M. et at. (1982) Proc. Natl Acad. ScL USA 79, 6337-6341 Hurley, C.K, Gregersen, P.K., Steiner. N. et al. (1988) J. Imrnunol. 140, 885-892 ~!7 Schenning, L., Larhammar, D., Bill, P. et at. (1984) EMBO J. 3, 447-452 Bell, J.I., Estess, P., St John. T. et at. ( i 985) Proc. Nati Acad. Sci. USA 82,
3405-3409 Tonnelie, C., DeMars, R. and Long, E.O. (1985)EMBO J. 4, 2839-2847 100 Lee, B.S.M., Bell, J.l., Rust, N.A. and McDevitt, H.O (1987) Irnmunogenetics 26, 85-91 11)1 Tsukamoto, K. Yasunami, M., Kimura, A. et al. (1987)Immunogenetics 25, 343-346 102 Boss, J.M. and Strominger, J.L (1984) Proc. Natl Acad. Sci. USA 81, 5199-5203 103 Michelsen, B. and Lernmark, ~. (1987)1 Clin. Invest. 79, 1144-1152 104 Larhammar, D., Hyldig-Nielsen, J.J. and Servenius, B. (1983) Proc. Natl Acad. Sci. USA 80, 7313-7317 105 Trowsdale, J., Young, J.A.T., Kelly, A.P. et ai. (1985) ImmunoL Rev. 85, 5-43 106 Bugawan, T.L., Horn, G.T., Long, C.M. et al. (1988)J Immunol. 141, 4024-4030
intrasplenic ~ere are two ways to raise antibodies to minute amounts of immunogen. The first is in-vitro immunization I in which the immunogen is presented to a spleen cell culture and about a week later cell fusion for hybridoma production is attempted. The second, and the subject of this review, is intrasplenic immunization2, in which the immunogen is deposited into the ~ , ~' , , ,,oou~ a, lu ~u~e --'~ . . .d,,nd~ . . . . . . . .~etr . . . . ~aKes care 0 f g ro wing the spleen cells. Both of these techniques are appropriate when only small amounts of immunogen are available. Intrasplenic immunization, however, requires less laboratory work and there is a decreased risk of contamination, o#en a problem with hybridoma cukures 3. The experience of intrasplenic immunization shows that it is the method of choice for immunization with nanogram am~un~ of ~mmunogen. A successful outcc.,~e, however, requires that the immunogen is immobJized on a carrier. This rewew by Ove Nilsson and Anders Larsson will focus on the various types of matrix which can be used as cc,rriers and on the procedures for transferring these carriers into the spleen tissue. Intrasplenic immunization was introduced by Nilsson et al. ~. Estimation of the sensitivity of the intrasplenic immunizatien technique si~owed that a total amount of about 80 ng of bovine serum albumin (BSA), dispersed in fo,:r depots, resulted in serum antibodies, provided that the antigen was immobilized on activated Sepharose beads 2. It was clearly easier to obtain an antibody response using minute amounts of immunogen by the intrasplenic route
"Department of Human Anatomy, and 2Department of Medical and Physiological Chemistn/" Uppsala University. The Biomedical Centre, 5-751 23 UpDsala,Sweden. 10
107 Gusta~sson, K., Widmark, E., Jonsson, A-K. et aL (1987) J. Biol. Che,n. 262, 8778-8786 108 Roux-Dosseto, M., Auffray, C., Lillie, J.W. et al. (1983)Proc. Nati Acad. Sci. USA 80, 6036-6040 109 Kappes, D.J., Arnot, D., Okada, K. and Strominger, J.L. (!984) EMBOJ. 3, 2985-2993 111) Tonnelle, C, DeMars, R. and Long, E. (1985)EMBOJ. 4, 2839-2847 111 Kelly, A. and Trowsdale, J. (1985) Nucleic Acids Res. 13, 1607-1621 11:~ Gustafsson, K., Emmoth, E., Widmark, E. et at. (1984) Nature 309, 76-78 113 Gorski, J., Rolhni, P., Long, E. and Mach, B. (1984)Proc. Natl Acad. Sci. USA 81, 3934-3938 114 Lair, B., Alber C., Yu, W-Y. et al. (1988)J. ImmunoL 141, 1353-1357 115 Bugawan, T.L., Angelini, G., Larrick, J. et at. (1989) Nature 339, 470-473
immunizationwith minute amountsof antigen B. Ove NilssonI and Anders Larssonz
than by intraperitoneal or intravenous routes s,6. Spitz et aL 7 independently reported production of antibodies after a single intrasplenic injection of either 20 IJg of soluble antigen or of 2.5 x 105 allogeneic cells. These findings were extended in subsequent publications 8,9. An antibody response was also obtained when the immunogen was immobilized on nitrocellulose paper, which then served as the carrier for immunization 2.s. This technique thus provides an easy way to immunize with electroblo~ed antigens after, for instance, electrophoretic purification of proteins, and less than 1 p.g protein is sufficient for immunization by this method. The success of intrasplenic immunization seems to be due to the deposition of the immunogen directly into the spleen, thus avoiding the losses associated with systemic injections, and to the immobilizing of the immunogen within the spleen, thus prolonging its exposition to the antigen-presenting cells. A requirement of the technique is that no toxic or irritating additive is used within the spleen since this would affect spleen cell function. Immunogen carriers Any type of small, inert and non-toxic material that can keep the immunogen in place while permitting its slow release is suitable. The immobilization of the immunogen on a matrix can be achieved nonspecifically or specifically: (~ 1990. Elsevier Sc ence Pubhshers Ltd. UK 0167 4919/89/$02.00
Immunology Tc,day, Vol. 11, No. 1 1990
that is, tee matrix used either attaches many different types of molecules or binds only specific ones determined by the active group (ligand) on the matrix. Appropriate matrices include small beads which can be glass, agarose or d~xtran based. A special form of bead matrix is offered by small groups of adherent cells (cell spherules, metastases, Langerhans' islets, blastocysts) that will expose immunogens at their peripheral cell surfaces and/or by their secretory activity. A similar arrangement is attained by §rowing cells on microcarriers. It should be noted, however, that in this case intracellular immunogens can also be exposed if the transferred cells decompose within the spleen. An alternative group of matrices are membranes, especially nitrocellulose membranes, usually intended for blotting. Beads Affini~.beads.A suitable carrier bead should be inert, have a high affinity for the immunogen without changing its structure, and have a size of 200-500 IJm for easy transfer into the spleen. Many types of bead made of crosslinked agarose or dextran, which are normally used for affinity chromatography, are excellent carriers (we use CNBr-activated Sepharose 4B (Pharmacia, Uppsala) to bind soluble immunogen) 2. Affinity beads have a porous matrix and pore size can be varied to match the size of the immunogen of interest. Beads can also covalently bind to specific ligands, with the immobilized ligand retaining specific binding affinity for an immunogen of interest. Thus with an enzyme as iigand, its substrate analogue or an inhibitor can be captured, a hormone will attach to its receptor, and a vitamin to its carrier protein. Derivatized beads with different ligand groups are available - for instance ligands with exposed amino, carboxyl, hydroxyl or thiol groups. Thus a matrix-ligandantigen complex can be specifically tailored for each immunogen. It is a!so possible to exploit ready-made matrix ligand absorbants, which are specific for groups of immunogen. For instance, a matrix containinq protein A absorbs the Fc region of IgG molecules, one containing a lectin absorbs its corresponding carbohydrate residue, and with polyE or polyA as ligands, reverse transcriptase, interferon, or mRNA is captured. The mode of application of immunogen to the beads can be adapted to the circumstances. For example, if the aim is to raise antibodies against some product of cultured cells, the beads can be swollen using spent medium from the microenvironment of the cells so that the product of interest is adsorbed. Products secreted in vivo can also be imbibed, if the beads can be deposited and recovered without risk of contamination. Direct application onto the beads can be achieved if the immunogen is available in solution. The immunogen can then be stabilized, either by using derivatized beads or by cross-linking with, for instance, glutaraldehyde. The capacity to use beads both for capturing an immunogen and for carrying it into the spleen greatly facilitates the work with intrasplenic immunization. Liposomes. Liposomes are phospholipid vesicles which can be made species compatible and therefore nontoxic and nonantigenic 1°. Various immunogens can be included in these membranes and thus the liposomes will serve as carriers for intrasplenic immunization. Since the lipo-
somes carl be prepared as multilametlar vesicles, they can be heavily laden with a membrane-inserted protein. Iscomes. An iscome is a cage-like structure with a diameter of approximately 35 nm designed for increasing the immunogenicity of proteins without using Freund's adjuvants 1;. It is produced from Quil A, a protein extracted from the bark ot~ Quillaja sapc~naria. Iscomes have been used to improve vaccines, and to enhance the antibody response to peptides 12. Since iscomes increase the efficiency of the antibody response, they are candidates for presenting immunogen in intrasplenic immunization. Cellspherules. Some ceils grow under normal conditions in small groups known as spherules, for exan-,plethe islets of the pancreas, early metastases ~3 and cleavage stages of fertilized eggs (morul!ae and blastocysts)4. These etl~ities are autonomous and simple to prepare for growth in vitro. They are, therefore, a suitable vehicle for transfer into the spleen and have the advantage that they grow while exposing various membrane-bound or secreted immunogens to the immune system. If the spherules consist of rapidly dividing cells these will p~oliferate in the spleen and ultimately destroy it. This happens, for instance, with both metastases and blastocysts, which produce an invasive growth of cells. In these cases the spherules must be irradiated to block their growth capacity before transfer into the spleen. Spherules can be transferred into the spleen, by pipette, suspended in buffer to which serum from the recipient has been added (see below). A more convenient way is to deposit the spherules on small pieces of nitrocellulose paper, let them dry, and then store the pieces of paper until a sufficient number has been collected. However, it must be remembered that drying the spherules might induce conformati.Jnal changes in the immunogens and expose intracellular components. Microcarrier. A cell complex similar to spherules can be prepared by growing cells on the surface of microcarriers, that is, beads with a diameter of about 200 I~m and a surface that can bind living cellst4 If microcarriers are left in a culture medium with suspended cells, the beads wil! eventually be covered by a layer of growing cells. Under optimal conditions these cells will be functionally normal, for example with respect to their phenotype and their secretory products. By depositing irradiated microcarriers in the spleen, a steady supply of immunogens from immobilized living cells can be obtained. Since the cells on the microcarriers within the spleen will be under the influence of hormones, growth factors and other signal substances of the host, a physiologically normal response by the cells may be expectEJ. Membranes The membranes used for blotting after electrophoresis are designed to keep substances immobilized and they are thus ideal for use as carriers in intrasplenic immunization. A suitable type of membrane should be selected, based on the properties of the antigen. Thus nitrocellulose membranes are the conventional material for binding of hydrophilic proteins, while hydrophobic species are attracted by nylon membranes. Cationic nylon membranes bind oligonucleotides, DNA and RNA.
11
Immunology Toclo~; "~ol. ; 1, No. I 1990
The membranes should be either triangular (base 2 mm, sides 3 ram) or circular (diameter of 2-3 mm). The immunogen in solution can be deposited o~to the membranes by soaking or by allowing it to drip from a pipette. Transfer by b!otting after electrophcresis can also be applied: the blotted membranes from one- or be-odimensional electrophoresis are prepared for immunization by punching discs from areas where a band or a spot of the immunogen is located. Staining the proteins does not seem ~o affect their antigenicity: no loss of antigenioty was observed when immunizing with 800 ng of Amido Black stained or Commassie Brilliant Blue stained BSA bound to NC-paper, compared with an unstained control s. This technique therefore, allows antibodies to be raised against specific regiens of an electrophorograrn.
Preparation procedures The choice of matrix for intrasplenlc immunization depends on the type of antigen and on the way in which the antigen can be recovered. It must be remembered that some of the techniques menticned involve the risk of conformational changes to the antigen during, for instance, procedures involving drying or cross-linking with formaldehyde. Tnis can result in the production of antibodies against artificial and functionally inactive immunogens. As a ru!e, the screening of the antibodies obtained should be performed on specimens of the same tyl~e as ~hose used for immunizatiorJ unless other routines a r e necessary, if it is not possible to purify the immunogen in an amount sufficient for immunization~ then it might be possible to initiate immunization conventionally, say intraperitoneally, with a crude material containing the native immunogen in question, then to make the last boost intrasplenically with the re~ined native substance, at might be anticipated that the last boost, although minute, would activate only those clones "hat i~roduce antibodies against the refined native substance. The transfer of the carriers into the spleen is made by pipette, in the case ~f a suspension of beads, and by watchmaker's forceps in the case of insertion ef membranes. The beads are best suspended in phosphatebuffered saline (PBS) containing 20-30% of the recipient's serum to avoid transferring additional, irrelevant immunogens. If the beads or the membranes have to be stored without drying, they should be kept in recipient serum diluted with PBS thus blocking any remaining sites with autologous proteins. The recipient can be a mouse or rat of an appropriate strain for monocionai antibody production, or a rabbit if a polyclonal antiserum is requireo, or a hen for obtaining egg antibodies. In the latter case, the antigen has to be deposited at or into the cloacal bursa ~5. We are currently exploring the feasibility of this approach. To expose the spleen, mice are given a basal anaesthetic dose of Mebumal ® (0.075 mg g-~ subcutaneously) supported by ether 2. Atropine (0.0150.025 mg)is given subcutaneously before surgery. After removal of the fur by shaving and dampening with ethanol, a cutaneous incision, about 10 mm long, is made along the left midscapular line followed by incision of the peritoneum. The tail of the spleen on its fat pad is carefully disserted out from the peritoneal cavity with the bloocl suDply intact. After the intraspi~nic deposition of antigen (see below), the spleen is replaced and the abdominal wall ard t:he skin .',ulured ~eparately. Rabbits are anaesthetized
12
with about 2.5 ml kg -1 of Mebumal ® (12 mg m1-1 saline) through an ear-vein catheter. The operation is performed essentially as for mice. The transfer of beads is made after puncturing the splenic capsule with a 27-gauge needle in the distal end of the spleen. A blunt glass micropipette containing beads in PBS with recipient serum is then inserted through the puncture hole. The beads are slowly expelled, and the pipette carefully retracted from the instillation site. The transfer of antigen absorbed by nitrocellulose paper is made through a small slit in the splenic capsule, made with the edge of an injection needle. One niece of nitrocellulose paper, handled with forceps, is inserted into the spleen through the slit and moved distally until it becomes completely embedded in the splenic tissue. The immunization schedule used is the conventional one. With beads as carriers, repeated transfers are possible, whereas the use of nitrocellulose strips precludes more than three to four boosts, since damage to the spleen may occur. The titre of serum antibodies can be checked from retroorbital or tail-vein puncture although the minute amounts of immunogen used usually do not result in the appearance of serum antibodies. Thus, fusion of spleen cells is often necessary without having obtained a positive serum response.
References 1 Reading, C.L. (1982)J. Immunol. Methods 53, 261-291 2 Nilsson, B.O., Svalander, P.C. and Larsson, A. (1987) J. Immunol. Methods 99, 67-75 3 Hirschberg, L., B~lske, G. and Holme, T. (1989) Hybridoma 8, 249-257 4 Nilsson, B.O.. Gr~nvik, K-O. and Svalander, P.C. (1983) Uppsai-~J. Med. Sci. 88, 151-153 5 Larsson,A. and Nilsson, B.O. (1988) Scand. J. Immunol. 27, 305-309 6 Ambrosius, H. and Schenderlein, C. (1986)~llecq. Immuno_l.. 32,243-252 7 Spitz, M., Spitz, L., Thorpe, R. and Eugui, E. (1984) J. Imm~,no.~. Methods 70, 39-43 8 Thorpe, R., Bird, C.~,. and Spitz, M. (1984) J. ImmunoL Methods 73, 259-265 9 Gearing, A.J.H., Thorpe, R., Spitz, L. and Spitz, M. (1985) J. Immunol. Methods 76, 337-343 10 Torchilin, V.P. and Klibanov, A.L. (1981) Enzyme Microb. -;~chnol. 3, 297-304 11 Morein, B., Sundqvist, B., Hoglund, S., Dalsgaard, K. and Osterhaus, A. (1984)Nature 308, 457-460 12 L6vgren, K., Lindmark, J., Pipkorn, R. and Morein, B. (1987) J. Immunot. Methods 98, 137-143 13 Sutherland, R.M (1988) Science 240, 177-184 14 Mered, B., Albrecht, P. and Hopps, H.E. (i980)In Vitro 16, 859-865 15 Hughes, C.L. and Henderson, D.C. (1977) ImmunoL Commun. 6, 195-206
In the next issue of IT * Dissection of an inflammatoryprocess induced by CD8 ÷ T cells P. Doher~y and colleagues * The costimulatory function of antigen presenting cells C.T. Weaver and E.R. Unar~e * An tiphospholipid antibodies, more than just a disease marker? C. Mackworth-Young