- Email: [email protected]
Adoptive transfer of experimental allergic encephalomyelitis using serum-free, chemically defined in vitro culture conditions
Journal of Neuroimmunology, 40 (1992) 111-114
111
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-5728/92/$05.00 JN1 02227
Short communication
Adoptive transfer of experimental allergic encephalomyelitis using serum-free, chemically defined in vitro culture conditions J. T h e o d o r e Phillips and Kimberly A. Meyers Dit,ision of Neuroimmunology, Department of Neurology, University of Texas Southwestern Medical Center, Dallas, TX, USA (Received 4 February 1992) (Revised, received 2 April 1992) (Accepted 2 April 1992)
Key words: Experimental allergic encephalomyelitis; Adoptive transfer; Passive transfer; Serum-free medium
Summary Enhanced adoptive transfer of experimental allergic encephalomyelitis (EAE) in rats requires in vitro culture of encephalitogen-sensitized donor spleen cells with either myelin basic protein or T-cell mitogen for 18-72 h prior to transfer to unimmunized recipients. The required in vitro culture period offers an opportunity to address in detail cellular and molecular immunoregulatory processes involved in the development of EAE. Conventional culture conditions using fetal bovine serum may impose analytical limitations due to the chemical complexity of the media. To permit better definition of the chemical events associated with the development of EAE, we report the successful adoptive transfer of EAE in Lewis rats using completely chemically defined, serum-free culture conditions.
Introduction Adoptive (passive) transfer of experimental allergic encephalomyelitis (EAE) is a frequently used procedure to study immunoregulatory events in the development of experimental allergic encephalomyelitis (EAE). In a typical protocol, spleen cells taken from a donor animal 10-12 days after immunization with spinal cord homogenate or myelin basic protein (MBP) are cultured in vitro for 18-72 h with either MBP or a T
Correspondence to: J.T. Phillips, Division of Neuroimmunology, Department of Neurology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235-9036, USA.
cell mitogen. At the end of this period, the sensitized spleen cells are administered to histocompatible, but unimmunized, recipients, and within a few days typical EAE subsequently appears (reviewed in Hinrichs, 1984; Swanborg, 1988). The in vitro culture period seems to be required for donor spleen cells to acquire competence to transfer EAE, but the cellular and molecular events which underlie this transformation remain largely unknown. Recent studies indicate that distinct T cell subsets and multiple paths of T cell development may be induced during the in vitro culture period, enabling the effective transfer of EAE memory and effector T cells (Bouwer et al., 1990). Furthermore, Racke and colleagues (1991) have shown that addition of transforming growth factor-/31 during the in vitro culture period in-
112
hibits the ultimate encephalogenicity of adoptively transferred MBP-specific T cells. Studies of this type will lead to an increased understanding of the immunoregulatory events important in the development of E A E and other autoimmunc conditions. In order to analyse more completely the immunoregulatory molecules and interactions involved in the in vitro development of EAE transfer competence, a completely defined culture medium would be useful Commonly used culture formulations include 5 - 1 0 % fetal bovine serum supplementation (Hinrichs, 1984; Swanborg, 1988), which due to its overall biochemical complexity, can hinder detailed analysis of cytokine involvement in the development of E A E transfer competence. To this end, we report the successful use of a simple and completely defined, serumfree culture medium, which enables the adoptive transfer of E A E with efficacy comparable to commonly used serum-containing formulations.
Adoptit'e tran,sfer of EAE
MBP was prepared from guinea pig spinal cords (Pel-Freez, Rogers, A R ) according to the protocol of Brostoff and Mason (1986). Purity of the MBP preparation was established by SDS polyaerylamide gel electrophoresis.
12 days post-immunization, spleens were aseptically removed from donors, and single-cell suspensions were prepared. Red cells wcrc not removed from the spleen cell suspensions. Suspensions were washed three times in Dulbecco's phosphate-buffered saline (DPBS; Sigma, St. Louis, MO) containing 0.1% bovine serum albumin (Sigma), and then rcsuspended at 2 million cells/ml in either serum-containing (SCM) or serum-free (SFM) media. SCM consisted of 5~i fetal bovine serum (Gibco, Grand Island, NY, 2 mM l.-glutamine (Gibco), 20 /,M 2-mercaptoethanol (2-ME; Sigma), 50 U / m l penicillin and 51) /.tg/m[ streptomycin (Gibeo) in RPMI-164{I (Sigma). SFM consisted of 10 / , g / m l human transferrin (Gibeo), 1(t /xg/ml bovine insulin (Sigma), 0.1 mM MEM non-essential amino acids (Sigma), 2 mM i.-glutamine (Gibco), 5 0 / , M 2-ME, 50 U / m l penicillin and 50 p . g / m l streptomycin in lscove's modified Dulbecco's medium (Sigma). A similar medium formulation has been shown to support the serum-free growth-activated murine lymphocytes (Gersten and Cohn, 1987). MBP was added to each medium at a final concentration ot 2 p.g/ml. Spleen cells in either SCM or SFM were then cultured in 75 cm 2 tissue culture flasks (Falcon 3024; Becton-Dickinson, Lincoln Park, N J) for 72 h in a humidified 59~ CO_~/95(~i air incubator at 37°C. Following incubation, cells were collected, washed twice in DPBS, and viability determined by standard Trypan blue exclusion. No differences in cell viabilities or cell concentrations were detectable in either culture formulation at 72 h. Viable cell concentrations were adjusted to 50 million/ml in DPBS immediately prior to intra-peritoneal (i.p.) delivery into recipient rats on day 0. Individual panels of rats rcceived i.p. doses ranging from 10-50 million cultured cells per rat.
Sensitization of donors
Clinical evahmtion
Lewis rats were immunized with an emulsion containing equal parts of MBP in saline and complete Freund's adjuvant (Difco, Detroit, MI; Mycobacteriurn tuberculosis H37Ra 10 m g / m l ) . Each animal received 0.1 ml subcutaneously into a rear footpad for a total of 50 p.g MBP and 500 /.tg H37Ra per animal.
Recipient rats were evaluated at the same time each day in a blinded fashion over 14 days for evidence of disease. The blinded investigator rated disease severity according to the following standard scale: 0, normal; 1, flaccid tail; 2, slight hindlimb weakness; 3, moderate hindlimb weakness; and 4, total hindlimb weakness (paraplegia)
Materials and methods
Ra Is Female Lewis rats (Harlan Sprague Dawley, Indianapolis, IN) aged 8 - 1 0 weeks were used in all experiments, and received food and water without restriction. Rats were acclimated for 1 week prior to experiments. Operative procedures were performed under ketamine anesthesia. All rats were ear-coded and randomized with regard to treatment, and were housed five rats to a cage.
Preparation of MBP
113
and incontinence. An average cumulative score was derived that incorporates overall clinical severity, disease incidence and duration of clinical signs into one comparative index: average cumulative score = sum of all daily clinical scores on all rats within a treatment group between days 2 and 14 post-cell t r a n s f e r / t o t a l number of rats within that treatment group. Results shown are pooled from three separate experiments.
traperitoneal route used in our studies versus the intravenous administration route. At low transfer numbers (10 and 20 million cells), recipients of SFM-cultured cells had a higher incidence of illness, and consequently higher average cumulative scores, than did recipients of SCM-cultured cells. The average maximum clinical score and total sick days of those rats which became ill were approximately the same in both groups. Histologic analysis of spinal cord sections from recipient rats demonstrated white matter inflammatory infiltrates typical of EAE, although no differences were noted among sick animals from different treatment groups (data not shown). Figure 2 shows the characteristic appearance of an inflammatory infiltrate in spinal cord white matter from rats administered SFM-cultured cells. These results show that E A E may be adoptively transferred using a completely defined, serum-free medium during the required 72 h in vitro culture period prior to cell transfer. The efficiency of transfer on a per-cell basis is equal to or better than that achieved with commonly used methods employing standard serum-containing culture conditions. At lower transferred cell
Results and discussion
Figure 1 compares the ability of equal numbers of viable MBP-sensitized Lewis spleen cells cultured in either serum-containing medium (SCM) or serum-free medium (SFM) to transfer E A E to syngeneic, unimmunized recipients. Table 1 indicates the overall severity of disease, as measured by average cumulative score, incidence of rats with EAE, and maximum clinical severity score. The incidence of transferred illness in these experiments is somewhat lower than seen in other studies (compare Bouwer et al., 1990). This likely reflects the lower transfer efficiency of the in-
4
4
4 c
20H
1 OH e £. o u
40H
// 0
t-i M (J -,.4 c .,d
2
(J
o o~ a L o > ,<
/o \ O/e
•
0
2
0
\
_._.._#',. *,.._~o_._._, 4
6
8
10
12
14
'f
0
•
\
\\
0 -i-i--'~. 6 2 4
Days
O
8
10
12
4
0
2
4
6
8
10
12
Post-TPansfeP
Fig. I. Average EAE clinical scores of recipient Lewis rats are shown after adoptive intraperitoneal transfer of 10 (a), 20 (b), or 40 (e) million MBP-immunized donor Lewis spleen cells on day 0. Donor cells were cultured for 72 h in either serum-free (SFM; ©) or standard serum-containing (SCM; o) media prior to transfer. Average clinical score = sum of clinical scores in all rats in a treatment group on that day/total number of rats in that treatment group. Data are pooled from three separate experiments.
114 FABLE I COMPARISON OF A D O P T I V E L Y T R A N S F E R R E D EAE USING D O N O R CELLS C U L T U R E D IN S E R U M - F R E E VERSUS S E R U M - C O N T A I N I N G C O N D I T I O N S Transferred donor cells
Average cumulative score ~
Serum-free medium 10 million 4.8 20 million 4.9 40 million 3.5 50 million 3.5 Serum-containing medium l(I million 1.2 2(I million 2.5 411 million 3.5
Number of rats sick
Average maximum clinical score of sick rats 1,
7/10 7/10 3/6 5/6
2.4 3.3 3.(I 3.2
1/6 2/6 14/21
3 3.(/ 2.4
medium, containing insulin, transferrin, and MBP as its only protein or hormone constituents. Therefore, using SFM conditions, the immunoregulatory effects of various cytokines, either added to the media or elaborated in situ as a result of antigen-specific stimulation will be easier to detect and further investigate. The use of chemically defined culture conditions during the in vitro acquisition of E A E transfer competence should allow more detailed descriptions of relevant molecular and cellular interactions involved in this important phase of EAE autoimmune effector cell development.
Acknowledgements ~' Average cumulative score = sum of all daily clinical scores on all rats within a treatment group between days 2 and 14 post-cell transfer/total number of rats within that treatment group. h Average maximum clinical score of sick r a t s - sum of the maximum clinical scores reached by individual rats which become ill within a treatment g r o u p / n u m b e r of rats that become ill within that treatment group.
numbers (10-20 million), the transfer efficiency of SFM-derived cells may exceed that of SCMderived cells. SFM is a completely defined
Fig. 2. Hematoxylin and eosin-stained section of a rat spinal cord demonstrates perivascular and parenchymal mononuclear cell infiltration 6 days after transfer of 20 million SFM-cultured, MBP-activated donor spleen cells. Original magnification: × 100.
This work was supported by National Multiple Sclerosis Society Grant RG1981A and by NIH Grant R29-DK39526.
References Bouwer, H.G.A., Dietsch, G.N. and Hinrichs, D.J. (1990) Adoptive transfer of experimental allergic encephalomyelitis: conditions influencing memory' and effector cell development. Cell. lmmunol. 131,219 231. Brostoff, S.W. and Mason, D.W. (1986) The role of lymphocyte subpopulations in the transfer of rat EAE. J. Neuroimmunol. 10, 331-340. Gersten, M.J. and Cohn, M. (1987) T cell responses studied in a basal serum-free medium. Methods Enzymol. 150, 129 133. Hinrichs, D.J. (1984) Requirements for and regulation of adoptively transferred paralytic EAE. In: A.A. Vandenbark and J.C.M. Raus (Eds.), Immunoregulatory' Processes in Experimental Allergic Encephalomyelitis and Multiple Sclerosis, Research Monographs in Immunology, Vol. 7, Elsevier, Amsterdam, pp. 63-98. Racke, M.K., Dhib-Jalbut, S., Cannella, B., Albert, P.S., Raine, C.S. and McFarlin, D.E. (19911 Prevention and treatment of chronic relapsing experimental allergic encephalomyelitis by transforming growth factor-,81. J. Immunol. 146, 3012-3017. Swanborg, R.H. (1988) Experimental allergic encephalomyelitis. Methods Enzymol. 162,413 421.
111
© 1992 Elsevier Science Publishers B.V. All rights reserved 0165-5728/92/$05.00 JN1 02227
Short communication
Adoptive transfer of experimental allergic encephalomyelitis using serum-free, chemically defined in vitro culture conditions J. T h e o d o r e Phillips and Kimberly A. Meyers Dit,ision of Neuroimmunology, Department of Neurology, University of Texas Southwestern Medical Center, Dallas, TX, USA (Received 4 February 1992) (Revised, received 2 April 1992) (Accepted 2 April 1992)
Key words: Experimental allergic encephalomyelitis; Adoptive transfer; Passive transfer; Serum-free medium
Summary Enhanced adoptive transfer of experimental allergic encephalomyelitis (EAE) in rats requires in vitro culture of encephalitogen-sensitized donor spleen cells with either myelin basic protein or T-cell mitogen for 18-72 h prior to transfer to unimmunized recipients. The required in vitro culture period offers an opportunity to address in detail cellular and molecular immunoregulatory processes involved in the development of EAE. Conventional culture conditions using fetal bovine serum may impose analytical limitations due to the chemical complexity of the media. To permit better definition of the chemical events associated with the development of EAE, we report the successful adoptive transfer of EAE in Lewis rats using completely chemically defined, serum-free culture conditions.
Introduction Adoptive (passive) transfer of experimental allergic encephalomyelitis (EAE) is a frequently used procedure to study immunoregulatory events in the development of experimental allergic encephalomyelitis (EAE). In a typical protocol, spleen cells taken from a donor animal 10-12 days after immunization with spinal cord homogenate or myelin basic protein (MBP) are cultured in vitro for 18-72 h with either MBP or a T
Correspondence to: J.T. Phillips, Division of Neuroimmunology, Department of Neurology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75235-9036, USA.
cell mitogen. At the end of this period, the sensitized spleen cells are administered to histocompatible, but unimmunized, recipients, and within a few days typical EAE subsequently appears (reviewed in Hinrichs, 1984; Swanborg, 1988). The in vitro culture period seems to be required for donor spleen cells to acquire competence to transfer EAE, but the cellular and molecular events which underlie this transformation remain largely unknown. Recent studies indicate that distinct T cell subsets and multiple paths of T cell development may be induced during the in vitro culture period, enabling the effective transfer of EAE memory and effector T cells (Bouwer et al., 1990). Furthermore, Racke and colleagues (1991) have shown that addition of transforming growth factor-/31 during the in vitro culture period in-
112
hibits the ultimate encephalogenicity of adoptively transferred MBP-specific T cells. Studies of this type will lead to an increased understanding of the immunoregulatory events important in the development of E A E and other autoimmunc conditions. In order to analyse more completely the immunoregulatory molecules and interactions involved in the in vitro development of EAE transfer competence, a completely defined culture medium would be useful Commonly used culture formulations include 5 - 1 0 % fetal bovine serum supplementation (Hinrichs, 1984; Swanborg, 1988), which due to its overall biochemical complexity, can hinder detailed analysis of cytokine involvement in the development of E A E transfer competence. To this end, we report the successful use of a simple and completely defined, serumfree culture medium, which enables the adoptive transfer of E A E with efficacy comparable to commonly used serum-containing formulations.
Adoptit'e tran,sfer of EAE
MBP was prepared from guinea pig spinal cords (Pel-Freez, Rogers, A R ) according to the protocol of Brostoff and Mason (1986). Purity of the MBP preparation was established by SDS polyaerylamide gel electrophoresis.
12 days post-immunization, spleens were aseptically removed from donors, and single-cell suspensions were prepared. Red cells wcrc not removed from the spleen cell suspensions. Suspensions were washed three times in Dulbecco's phosphate-buffered saline (DPBS; Sigma, St. Louis, MO) containing 0.1% bovine serum albumin (Sigma), and then rcsuspended at 2 million cells/ml in either serum-containing (SCM) or serum-free (SFM) media. SCM consisted of 5~i fetal bovine serum (Gibco, Grand Island, NY, 2 mM l.-glutamine (Gibco), 20 /,M 2-mercaptoethanol (2-ME; Sigma), 50 U / m l penicillin and 51) /.tg/m[ streptomycin (Gibeo) in RPMI-164{I (Sigma). SFM consisted of 10 / , g / m l human transferrin (Gibeo), 1(t /xg/ml bovine insulin (Sigma), 0.1 mM MEM non-essential amino acids (Sigma), 2 mM i.-glutamine (Gibco), 5 0 / , M 2-ME, 50 U / m l penicillin and 50 p . g / m l streptomycin in lscove's modified Dulbecco's medium (Sigma). A similar medium formulation has been shown to support the serum-free growth-activated murine lymphocytes (Gersten and Cohn, 1987). MBP was added to each medium at a final concentration ot 2 p.g/ml. Spleen cells in either SCM or SFM were then cultured in 75 cm 2 tissue culture flasks (Falcon 3024; Becton-Dickinson, Lincoln Park, N J) for 72 h in a humidified 59~ CO_~/95(~i air incubator at 37°C. Following incubation, cells were collected, washed twice in DPBS, and viability determined by standard Trypan blue exclusion. No differences in cell viabilities or cell concentrations were detectable in either culture formulation at 72 h. Viable cell concentrations were adjusted to 50 million/ml in DPBS immediately prior to intra-peritoneal (i.p.) delivery into recipient rats on day 0. Individual panels of rats rcceived i.p. doses ranging from 10-50 million cultured cells per rat.
Sensitization of donors
Clinical evahmtion
Lewis rats were immunized with an emulsion containing equal parts of MBP in saline and complete Freund's adjuvant (Difco, Detroit, MI; Mycobacteriurn tuberculosis H37Ra 10 m g / m l ) . Each animal received 0.1 ml subcutaneously into a rear footpad for a total of 50 p.g MBP and 500 /.tg H37Ra per animal.
Recipient rats were evaluated at the same time each day in a blinded fashion over 14 days for evidence of disease. The blinded investigator rated disease severity according to the following standard scale: 0, normal; 1, flaccid tail; 2, slight hindlimb weakness; 3, moderate hindlimb weakness; and 4, total hindlimb weakness (paraplegia)
Materials and methods
Ra Is Female Lewis rats (Harlan Sprague Dawley, Indianapolis, IN) aged 8 - 1 0 weeks were used in all experiments, and received food and water without restriction. Rats were acclimated for 1 week prior to experiments. Operative procedures were performed under ketamine anesthesia. All rats were ear-coded and randomized with regard to treatment, and were housed five rats to a cage.
Preparation of MBP
113
and incontinence. An average cumulative score was derived that incorporates overall clinical severity, disease incidence and duration of clinical signs into one comparative index: average cumulative score = sum of all daily clinical scores on all rats within a treatment group between days 2 and 14 post-cell t r a n s f e r / t o t a l number of rats within that treatment group. Results shown are pooled from three separate experiments.
traperitoneal route used in our studies versus the intravenous administration route. At low transfer numbers (10 and 20 million cells), recipients of SFM-cultured cells had a higher incidence of illness, and consequently higher average cumulative scores, than did recipients of SCM-cultured cells. The average maximum clinical score and total sick days of those rats which became ill were approximately the same in both groups. Histologic analysis of spinal cord sections from recipient rats demonstrated white matter inflammatory infiltrates typical of EAE, although no differences were noted among sick animals from different treatment groups (data not shown). Figure 2 shows the characteristic appearance of an inflammatory infiltrate in spinal cord white matter from rats administered SFM-cultured cells. These results show that E A E may be adoptively transferred using a completely defined, serum-free medium during the required 72 h in vitro culture period prior to cell transfer. The efficiency of transfer on a per-cell basis is equal to or better than that achieved with commonly used methods employing standard serum-containing culture conditions. At lower transferred cell
Results and discussion
Figure 1 compares the ability of equal numbers of viable MBP-sensitized Lewis spleen cells cultured in either serum-containing medium (SCM) or serum-free medium (SFM) to transfer E A E to syngeneic, unimmunized recipients. Table 1 indicates the overall severity of disease, as measured by average cumulative score, incidence of rats with EAE, and maximum clinical severity score. The incidence of transferred illness in these experiments is somewhat lower than seen in other studies (compare Bouwer et al., 1990). This likely reflects the lower transfer efficiency of the in-
4
4
4 c
20H
1 OH e £. o u
40H
// 0
t-i M (J -,.4 c .,d
2
(J
o o~ a L o > ,<
/o \ O/e
•
0
2
0
\
_._.._#',. *,.._~o_._._, 4
6
8
10
12
14
'f
0
•
\
\\
0 -i-i--'~. 6 2 4
Days
O
8
10
12
4
0
2
4
6
8
10
12
Post-TPansfeP
Fig. I. Average EAE clinical scores of recipient Lewis rats are shown after adoptive intraperitoneal transfer of 10 (a), 20 (b), or 40 (e) million MBP-immunized donor Lewis spleen cells on day 0. Donor cells were cultured for 72 h in either serum-free (SFM; ©) or standard serum-containing (SCM; o) media prior to transfer. Average clinical score = sum of clinical scores in all rats in a treatment group on that day/total number of rats in that treatment group. Data are pooled from three separate experiments.
114 FABLE I COMPARISON OF A D O P T I V E L Y T R A N S F E R R E D EAE USING D O N O R CELLS C U L T U R E D IN S E R U M - F R E E VERSUS S E R U M - C O N T A I N I N G C O N D I T I O N S Transferred donor cells
Average cumulative score ~
Serum-free medium 10 million 4.8 20 million 4.9 40 million 3.5 50 million 3.5 Serum-containing medium l(I million 1.2 2(I million 2.5 411 million 3.5
Number of rats sick
Average maximum clinical score of sick rats 1,
7/10 7/10 3/6 5/6
2.4 3.3 3.(I 3.2
1/6 2/6 14/21
3 3.(/ 2.4
medium, containing insulin, transferrin, and MBP as its only protein or hormone constituents. Therefore, using SFM conditions, the immunoregulatory effects of various cytokines, either added to the media or elaborated in situ as a result of antigen-specific stimulation will be easier to detect and further investigate. The use of chemically defined culture conditions during the in vitro acquisition of E A E transfer competence should allow more detailed descriptions of relevant molecular and cellular interactions involved in this important phase of EAE autoimmune effector cell development.
Acknowledgements ~' Average cumulative score = sum of all daily clinical scores on all rats within a treatment group between days 2 and 14 post-cell transfer/total number of rats within that treatment group. h Average maximum clinical score of sick r a t s - sum of the maximum clinical scores reached by individual rats which become ill within a treatment g r o u p / n u m b e r of rats that become ill within that treatment group.
numbers (10-20 million), the transfer efficiency of SFM-derived cells may exceed that of SCMderived cells. SFM is a completely defined
Fig. 2. Hematoxylin and eosin-stained section of a rat spinal cord demonstrates perivascular and parenchymal mononuclear cell infiltration 6 days after transfer of 20 million SFM-cultured, MBP-activated donor spleen cells. Original magnification: × 100.
This work was supported by National Multiple Sclerosis Society Grant RG1981A and by NIH Grant R29-DK39526.
References Bouwer, H.G.A., Dietsch, G.N. and Hinrichs, D.J. (1990) Adoptive transfer of experimental allergic encephalomyelitis: conditions influencing memory' and effector cell development. Cell. lmmunol. 131,219 231. Brostoff, S.W. and Mason, D.W. (1986) The role of lymphocyte subpopulations in the transfer of rat EAE. J. Neuroimmunol. 10, 331-340. Gersten, M.J. and Cohn, M. (1987) T cell responses studied in a basal serum-free medium. Methods Enzymol. 150, 129 133. Hinrichs, D.J. (1984) Requirements for and regulation of adoptively transferred paralytic EAE. In: A.A. Vandenbark and J.C.M. Raus (Eds.), Immunoregulatory' Processes in Experimental Allergic Encephalomyelitis and Multiple Sclerosis, Research Monographs in Immunology, Vol. 7, Elsevier, Amsterdam, pp. 63-98. Racke, M.K., Dhib-Jalbut, S., Cannella, B., Albert, P.S., Raine, C.S. and McFarlin, D.E. (19911 Prevention and treatment of chronic relapsing experimental allergic encephalomyelitis by transforming growth factor-,81. J. Immunol. 146, 3012-3017. Swanborg, R.H. (1988) Experimental allergic encephalomyelitis. Methods Enzymol. 162,413 421.