As compared to other assays, the IVCCA has three main advantages. (1) It provides a direct readout that is proportional to in vivo cytokine secretion (Finkelman and Morris, 1999). (2) It detects total-body cytokine secretion and is efficacious at detecting cytokine production wherever IgG antibodies penetrate, including the lungs (Finkelman and Morris, 1999). It would probably not work well to detect cytokines produced in areas in which IgG antibodies do not normally penetrate well, such as the central nervous system. (3) It does not require sacrificing the animal to obtain tissue samples and, because it samples cytokine production without neutralizing most secreted cytokine, it does not inhibit cytokine-dependent processes (Finkelman and Morris, 1999). In addition, because complexes formed between cytokine and biotin-anti-cytokine mAb cannot contain more than a single mAb molecule, they do not fix complement or bind more avidly than circulating self-IgG to Fcy receptors.
The IVCCA has been useful for studies of regulation of cytokine production (Finkelman et al., 2000), cytokine cross-talk (Thornton et al., 2001), cytokine-mediated host protection against infection (Urban et al., 1998, 2000), cytokine regulation of inflammatory conditions (e.g., allergy and immune-mediated arthritis; Thornton et al., 2000), and pathogenesis of inflammatory conditions (e.g., acute and chronic graft versus host disease; Via et al., 2001).
Presently, the IVCCA does not work for every cytokine. This may reflect the lack of high affinity antibodies to some cytokines or the presence of serum molecules, such as soluble receptors, that compete for cytokine with injected anti-cytokine mAbs. In addition, some cytokines which are homodimers, such as IL-5, may form large complexes with injected di-meric anti-cytokine mAbs that are rapidly removed from circulation by the reticuloen-dothelial system; however, this is not a problem for some other cytokines that are homodimers, including IL-10 and IFN-y.
Even for cytokines that are detected well by the IVCCA (Table 6.28.1), the quantity detected is influenced by competition between injected anti-cytokine mAb and intrinsic cytok-ine-binding molecules. For example, more IL-4 is detected by IVCCA in IL-4 receptor-deficient mice than in wild-type mice (Finkelman et al., 2000).
The IVCCA cannot distinguish the cell type or organ that is producing a cytokine. Other assays that can be used to provide such information include fluorescence staining for in-tracytoplasmic cytokines (unit 6.24; Openshaw et al., 1995), determination of cytokine mRNA content of organs or purified cell types (Chapter 6; Svetic et al., 1991), and ELISpot assays (unit 6.19; Fujihashi et al., 1993) performed with purified cells or cells from a specific organ.
The IVCCA can be less sensitive than other assays, such as ex vivo cytokine mRNA determination (Chapter 6), at evaluating increases in cytokine expression when such increases are highly localized and generate amounts of cytokine that are small compared to those produced in other sites in unstimulated mice.
The biotin-labeled mAbs injected into mice as part of the IVCCA are potentially immuno-genic. This has been an issue predominantly when repeated IVCCAs, and thus repeated mAb injections, have been performed in the same mice.
The readout of the IVCCA is relative, rather than absolute. It can be used to compare cytok-ine production between mice, but does not reveal the absolute quantity of cytokine secreted or the number of cytokine secreting cells.
The sensitivity of the IVCCA depends in large measure on the care with which the ELISA is performed (see Chapter 2 for detailed discussion). Because concentrations in the range of 10 to 20 pg/ml are being detected, considerable care must be taken to avoid contamination or spillage between wells during reagent addition or washing steps. Occasional problems with this assay have been traced to the presence of bubbles in pipetted reagents, contamination of ELISA plate washers, contaminated water supplies, defective multichannel pipettors or pipet tips, reagents that have gone through multiple freeze-thaw cycles, and reagents that were allowed to sit unrefrigerated for hours at a time. In most cases, these problems become apparent when poor standard curves are obtained and/or duplicate measurements show excessive variability.
A second problem in making cytokine measurements can occur as a result of inflammatory responses or infection. The large increases in cytokine production that occur following viral, bacterial, protozoan, fungal, and helminth infections can obscure smaller changes in cytokine production that occur during a purposeful immu nization. Consequently, an increase in production of a specific cytokine that is easily detected by IVCCA in mice in an SPF colony may be completely masked by increased background levels of the same cytokine in a "dirty" colony.
Not all anti-cytokine Ab pairs that work well for standard in vitro ELISAs work well for IVCCAs. For example, it was necessary to test various combinations of six anti-TNF-a mAbs to develop an effective IVCCA for TNF-a.
The cytokine:biotin-anti-cytokine mAb complexes that form in vivo vary in stability. For example, IFN-y:biotin-R46A2 complexes are considerably more stable than IL-4:biotin-BVD4-1D11 complexes. For this reason, increasing the time between injection of biotin-anti-IFN-y mAb and bleeding can substantially increase the amount of IFN-y detected, while increasing the time between injection of biotin-anti-IL-4 mAb and bleeding to >4 hr usually has little effect on the amount of IL-4 detected. For the same reason, the IL-4 IVCCA can be repeated every 3 days using the same mice (which are injected with additional 10 |g doses of biotin-anti-IL-4 mAb), while the IFN-y IVCCA can only be performed one time per mouse. It is not currently known whether an individual mouse can be used for repeated determinations of IL-2, IL-6, IL-10, or TNF-a by IVCCA.
Similar mean values for serum cytokine levels are obtained if biotin-labeled anti-cytokine mAbs are injected i.v. or i.p.; however, repro-ducibility has been somewhat better with i.v. injection. Subcutaneous injection of biotin-labeled anti-cytokine mAbs have given lower values in the authors' laboratory than were obtained with i.v. or i.p. injection.
The number of different cytokines that can be assayed simultaneously is limited by the amount of serum that can be obtained from a mouse and the amount of serum that must be used in the IVCCA. If assays are performed in duplicate using 25 |l of a 1:2 dilution for each cytokine, 25 | l serum is required per cytokine. In practice, the authors prefer twice this amount to allow assays to be repeated in case technical problems cause suboptimal results during the initial assay.
The authors use Pierce SuperSignal ELISA Femto substrate at one-tenth the concentration recommended by the manufacturer. This reduces the magnitude of the readout, but not the sensitivity of the assay, which, in the hands of the authors, is limited by signal/noise ratio rather than the absolute value of the readout. In addition, the use of a reduced amount of sub strate also decreases assay cost. However, increasing substrate concentration may increase assay sensitivity if modifications are first made to the ELISA to increase signal/noise ratio.
The sensitivity of most of the ELISAs in the authors' laboratory is usually 15 to 30 pg/ml. With great care, sensitivity can be increased to ~5 pg/ml by lowering background readings. The IVCCA generally detects IL-2, IFN-y, and TNF-a levels above background even in mice raised under SPF conditions. These levels rise considerably with immunization or infection. IL-4 and -6 levels are frequently undetectable in healthy, unimmunized mice, but rise rapidly with the proper immunization or infection (particularly worm infections for IL-4). IL-10 has been undetectable in unimmunized mice and has often been undetectable even after infection or immunization, but is easily detected in mice injected with LPS or anti-CD3 mAb. The authors suspect that the IL-10 IVCCA is less sensitive than the IVCCAs for the other five cytokines.
Preparation of microtiter plates for the IVCCA (coating with antibody) requires an overnight incubation. Once microtiter plates are prepared, the entire assay, including dilution of serum samples, takes 1.5 to 2 hr if using the luminescence-based and at least 2 to 2.5 hr using the absorbance-based approach. The ab-sorbance-based approach can take considerably longer if prolonged times are used for development after adding substrate.
Anderson, U. and Anderson, J. 1994. Immunola-belling of cytokine producing cells in tissues and suspension. In Cytokine Producing Cells (D. Fradelizie and D. Emelie, eds.) p. 32-49. IN-SERM, Paris.
Carter, L.L. and Swain, S.L. 1997. Single cell analyses of cytokine production. Curr. Opin. Immunol. 9:177-182.
Finkelman, F.D. and Morris, S.C. 1999. Development of an assay to measure in vivo cytokine production in the mouse. Int. Immunology 11:1811-1818.
Finkelman, F.D., Madden, K.B., Morris, S.C., Holmes, J.M., Boiani, N., Katona, I.M., and Mal-iszewski, C.R. 1993. Anti-cytokine antibodies as carrier proteins: Prolongation of in vivo effects of exogenous cytokines by injection of cytokine—anti-cytokine antibody complexes. J. Immunol. 151:1235-1244.
Cytokines and Their Cellular Receptors
Was this article helpful?