- Email: [email protected]
The role of the immune system in central nervous system regeneration (theoretical considerations)
Medical Hypotheses 0 Imgman Group
(19%) 26, 13- 15 UK Ltd 1988
The Role of the Immune System in Central Nervous System Regeneration (Theoretical Considerations) THOMAS Labour&
A. KELLY College,
2120 Dorchester
Ave. Boston,
MA. 02724. U.S.A.
Steroids Abstract - The nervous and immune systems are shown to be interrelated. inhibit both regeneration and the immune response. Crushing injuries enhance regeneration. A synthesis of these apparently unrelated phenomena is formulated into an immune hypothesis of regeneration. Steroids, in suppressing the immune response, decrease lymphokine production and the subsequent stimulation of nerve cell regeneration. Crushing injuries to nerve fibers cause nerve protein to become antigenic, thereby eliciting the production of “non-toxic” antibodies, which paradoxically protect the nerve fibers from immune attack and cause an augmented lymphokine production and subsequent nerve cell regGneration.
Introduction
The enhancement of regeneration of new axonal processes of nerve fibers has long been sought. Various hypotheses have arisen in that quest. Since myelinated nerves can undergo Wallerian degeneration and regeneration, the neurolemma of the Schwann cell has been thought to play a role in regeneration. This role is dual in nature: 1. The myelin sheath provides structural support and alignment of the axon core as it regrows, and 2. The Schwann cells secrete molecules that stimulate nerve growth. (1) Steroids used in the treatment of inflammation in spinal cord injury have actually been shown to impede regeneration (5, 15). Some might object to steroids’ role as immunosuppressants by saying that there are
other immunosuppressant drugs such as cyclophosphamide and azathioprine that do not inhibit and may enhance regeneration. Yet, when one considers the molecular biology of these drugs which work on the DNA of target cells versus steroids which do not, one might concede that a different mode of action might possibly translate to a different result. The nervous and immune systems have been shown to be interrelated, physically, by nerve fibers of the vagus nerve wired directly from the brain to the thymus, necessary for normal thymic function (3), and psychologically, by the influence of mood states on immune performance (11). Studies by Geschwind (7) revealed that a central nervous system left-handedness,
14 phenomenon, may be associated with increased risk of developing autoimmune disease and that this link may depend on levels of testosterone. Recently, Bluestein (2) showed that nerve cells and while blood cells share very late activation antigens (VLAs) in common. He suggests a possible connection between VLAs and systemic lupus erythematosus: initially, in lupus, antibodies (ABs) are made against VLAs on suppressor T cells. These destroy the immune system’s recognition of nerves and they are attacked as foreign tissue. There is a key parallel between the nervous and immune systems - communication. All cells intercommunicate by mutual activation or inhibition; e.g., “contact inhibition” in tissue culture. However, in the nervous and immune systems this communication is more precisely refined. Nerves communicate by electrical impulses generated by membrane ion changes, stimulated by chemical neurotransmitters across physical synapses. The immune system by protein “hormone-like” communicates substances called . lymphokines. These include platelet-derived growth factor (PDGF), which Rozengurt (13) showed to be an early mitogenic signal; gamma interferon, (y-IF), which stimulates natural killer cells (NK cells) to attack virally infected and cancer cells; and the interleukins (IL-l, IL-2, IL-3), which regulate immune cell proliferation and activation. Aside from similar parallel relationships, the nervous and immune systems share certain mediators in common, although their role may be different in the two systems. For instance, leukotrienes, known to be involved with prostaglandins and bradykinins in mediating pain, have been shown by Samuelson (14) to be mediators of immediate hypersensitivity reactions. More recently, Gurney et al (8) have shown that neuroleukin, a neuronal growth factor, is a lymphokine modulator of B-cell function. Since the relationship between the nervous and immune systems has been established, and since a growth-inducing molecule has been suggested in central nervous tissue (CNS) regeneration, I propose that lymphokines be tested as growth-inducing substances in spinal nerve regeneration experiments. Theoretical considerations
Aside from growth and cell proliferation, the immune system might be involved in regenera-
MEDICAL HYPOTHESES
tion in another way. The role of antibody attack of nerve fibers as a mechanism inhibiting regeneration was proposed by Feringa (5). However, the immune system, in attacking the nerves, may help to promote regeneration in an almost paradoxical way. A similar situation was shown to be true of the fetus by Mizel (12). The fetus, as the product of a genetic recombined cell - the fertilized egg - contains different genes, and hence antigens (Ags) that are not those of the mother, and, therefore, are attacked by her immune system. However, this attack is to the fetus’s benefit since the antibodies formed are “non-toxic,” or “enhancing” antibodies (ABs), which coat the fetal antigens on the trophoblast and actually protect them from recognition by maternal immune cells, thereby preventing their destruction. This type of scenario could operate in regeneration. If the normal proteins of exposed nerve endings in transected nerves are antigenic because they have never interfaced with immune cells. then it is possible, too, that some nerve proteins could be altered during the process of transection and the events of inflammation that follow it. Altered proteins in nerves after transection could be attacked by “non-toxic” antibodies. These “non-toxic” antibodies could act as decoys, drawing the immune attack away from normal nerve protein antigens, thereby protecting them from destruction. In addition, immune activation would result in more lymphokines which would thus stimulate regeneration. A similar scenario has been proposed by Mizel (12) to explain how tumors avoid destruction by the immune system. It is interesting to note the early work of Levi-Montalcini et al (10) who showed the groivth stimulating effects of mouse sarcoma on sympathetic nerves. If tumors decoy and tumors stimulate regeneration, couldn’t the actual mechanism of regeneration in nerves be the very same immune decoying I propose it is? Recently, Herlyn et al (9) showed that gamma interferon. (y-IF) induces or augments the expression of cell surface major histocompatibility complex (MHC) class II antigens. Antigen presentation would be necessary to stimulate the immune system to make antibodies against altered nerve antigens. It is interesting in this scenario that y-IF would play a dual role: antibody stimulation and regeneration. But why would the immune system attack its own nerves? Perhaps in the process of transection and subsequent inflammation axon proteins
Till:
IMMIINt:
SYSTEM
IN (‘ENTKAL
NERVOUS
SYSTEM
REGENERATION
would be modified and become antigenic. Yao et al (16) showed alterations in endoneurial lipids in Wallerian degeneration and regeneration. Similar changes in protein would not be surprising. Remember, too. that macrophages are involved in inflammation, and macrophages are antigen presenting cells (APCs). They could consume protein nerve debris after transection and. in digesting it, modify it and present it to the immune system as new antigen to be subsequently reacted against with “non-toxic” antibody. Evidence already cited (5. 15) indicates that steroids inhibit regeneration. Also, Gaster et al (6) showed that crushing stimulated regeneration. Consider these facts in the light of my immunologic hypothesis. Steroids inhibit regeneration by inhibiting immune cells. including antigen processing macrophages. Hence, lymphokine stimulation of regeneration decreases. Crushing stimulates regeneration by increasing inflammation. The increased inflammatory response. with its increased APCs, converts nerve protein to antigenic protein. Thus. the protective role of “non-toxic” antibody is elicited, as well as the regenerating activity of lymphokines. But why wouldn’t this scenario operate normally in CNS transected nerves, resulting in spontaneously occurring regeneration? The answer is myelin. In myelinated nerves the Schwann cells secrete neuronal growth factor, in unmyelinated nerves these must be supplied exogenously by lymphokine treatment, as I propose. In myelinated nerves, the myelin itself blocks immune recognition of antigenic nerve protein. In unmyelinated nerves, where antigenic nerve proteins exposed by transection are subject to attack by the immune system’, we would want to stimulate the production of “nontoxic” decoy antibodies to detract from the attack on normal nerve protein.
the decoying mechanism preferentially. Because of the above, I believe lymphokines should be tested in nerve regeneration models. To extend the lymphokine-regeneration hypothesis further, since lymphocytes are controlled by ion channels (4), it is possible that regulatory actions of both the immune and nervous systems could be modulated in the future by alteration of intracellular K+ or Ca++ ion ccncentrations by substances known to block ion channels. References I. Aguayo.
2. 3. 4.
5 _
6.
7. 8.
9.
10.
11.
Conclusion Why would increased stimulation of the immune response selectively stimulate only “non-toxic” antibodies? One could not ascertain this without experimentation. But since certain lymphokines (gamma-y-interferon) stimulate antigen presentation (9), it seems plausible that specifically modulated stimulation (of the antigen presenting cells) could lead to an increase in different AB idiotypes (including “non-toxic” AB to altered nerve proteins) rather than just a single idiotype to normal nerve proteins. This would enhance
15
A. How can nerve cells be encouraged to reestablish functional connections. Science News IZY: 2114. 1986. Bluestein, G. Nerve. immune link found on membranes. Science News 130: 118. 1986. Bullock, K. in R. Guillemin et al (eds.): Neural Modulation of Immunity. New York: Raven. p. I1 I. IYX5. Chandy. K. G., DeCoursey. T. E.. Cahalan. M. I>.. McLaughlin, C.. and Gupta. S. Voltage-gated potassium channels are required for human T lymphocyte activation. J. Exp. med. 160: 36Y. 1984. Feringa, E. R.. Johnson. R. D.. and Wendt. S. J. Spinal cord regeneration in rats after immunosuppressive treatment. Arch. Neural. 32: 676, 1975. Caster. R. N.. Davidson, T. M.. Rand, R. W., and Fonkalsrud, E. W. Comparison of nerve regeneration rates following controlled freezing and crushing. Arch. Surg. 103: 37X. 1971. Geschwind. N. Autoimmunity in left-handerc. Science 217: 141. 1982. Gurney. M. E.. Apatoff. B. R., Spear. G. T.. Baumel. M. J.. Antel. J. P., Brown, M.. Bania, M., Reder. A. T. Neuroleukin: a lymphokine product of lectin-stimulated T cells. Science 234: 574. lYS6. Herlyn. M., Guerry. D.. and Koprowskr. H. Rccombinant gamma-interferon induces changes in expression and shedding of antigens associated with normal human melanocytes. nevus cells. and primary and metastatic melanoma cells. J. of Immunology 134: 4226. IYX5. Levi-Montalcini. R.. Meyer, H.. and Hamburger. V. Irr vitro experiments on the effects of mouse sarcoma lX(I and 37 on the spinal and sympathetic ganglia of the chick embryo. Cancer Res. 14: 49. lYS4. Locke, S. and Colligan, D. The Healer Within. New York: E. P. Dutton. 1986.
12. Mizel, S. B. and Jaret. P. In Self-Defense. New York: Harcourt. Brace. Jovanovich, pp. 145, IYX5. 13. Rozengurt. E. Early signals in the mitogenic response. Science 234: 161. 19X6. 14. Samuelson. B. Leukotrienes: mediators of immediate hypcrscnsitivity reactions and inflammation. Science 210: 568. 1983. IS. Scheff. S. W.. Hoff, S. F.. and Anderson, K. J. Altered regulation of lesion-induced synaptogenesis by adrcnalectomy and corticosterone in young adult rats. Exp. Neurol. Y3: 456, 1986. 16. Yao. J. K.. Natarajan. V. and Dyck. P. J. The sequential alterations of endoneurial cholestereol and fatty acid in Wallerian degeneration and regeneration. J. Neurothem. 35: 933. 1980.
(19%) 26, 13- 15 UK Ltd 1988
The Role of the Immune System in Central Nervous System Regeneration (Theoretical Considerations) THOMAS Labour&
A. KELLY College,
2120 Dorchester
Ave. Boston,
MA. 02724. U.S.A.
Steroids Abstract - The nervous and immune systems are shown to be interrelated. inhibit both regeneration and the immune response. Crushing injuries enhance regeneration. A synthesis of these apparently unrelated phenomena is formulated into an immune hypothesis of regeneration. Steroids, in suppressing the immune response, decrease lymphokine production and the subsequent stimulation of nerve cell regeneration. Crushing injuries to nerve fibers cause nerve protein to become antigenic, thereby eliciting the production of “non-toxic” antibodies, which paradoxically protect the nerve fibers from immune attack and cause an augmented lymphokine production and subsequent nerve cell regGneration.
Introduction
The enhancement of regeneration of new axonal processes of nerve fibers has long been sought. Various hypotheses have arisen in that quest. Since myelinated nerves can undergo Wallerian degeneration and regeneration, the neurolemma of the Schwann cell has been thought to play a role in regeneration. This role is dual in nature: 1. The myelin sheath provides structural support and alignment of the axon core as it regrows, and 2. The Schwann cells secrete molecules that stimulate nerve growth. (1) Steroids used in the treatment of inflammation in spinal cord injury have actually been shown to impede regeneration (5, 15). Some might object to steroids’ role as immunosuppressants by saying that there are
other immunosuppressant drugs such as cyclophosphamide and azathioprine that do not inhibit and may enhance regeneration. Yet, when one considers the molecular biology of these drugs which work on the DNA of target cells versus steroids which do not, one might concede that a different mode of action might possibly translate to a different result. The nervous and immune systems have been shown to be interrelated, physically, by nerve fibers of the vagus nerve wired directly from the brain to the thymus, necessary for normal thymic function (3), and psychologically, by the influence of mood states on immune performance (11). Studies by Geschwind (7) revealed that a central nervous system left-handedness,
14 phenomenon, may be associated with increased risk of developing autoimmune disease and that this link may depend on levels of testosterone. Recently, Bluestein (2) showed that nerve cells and while blood cells share very late activation antigens (VLAs) in common. He suggests a possible connection between VLAs and systemic lupus erythematosus: initially, in lupus, antibodies (ABs) are made against VLAs on suppressor T cells. These destroy the immune system’s recognition of nerves and they are attacked as foreign tissue. There is a key parallel between the nervous and immune systems - communication. All cells intercommunicate by mutual activation or inhibition; e.g., “contact inhibition” in tissue culture. However, in the nervous and immune systems this communication is more precisely refined. Nerves communicate by electrical impulses generated by membrane ion changes, stimulated by chemical neurotransmitters across physical synapses. The immune system by protein “hormone-like” communicates substances called . lymphokines. These include platelet-derived growth factor (PDGF), which Rozengurt (13) showed to be an early mitogenic signal; gamma interferon, (y-IF), which stimulates natural killer cells (NK cells) to attack virally infected and cancer cells; and the interleukins (IL-l, IL-2, IL-3), which regulate immune cell proliferation and activation. Aside from similar parallel relationships, the nervous and immune systems share certain mediators in common, although their role may be different in the two systems. For instance, leukotrienes, known to be involved with prostaglandins and bradykinins in mediating pain, have been shown by Samuelson (14) to be mediators of immediate hypersensitivity reactions. More recently, Gurney et al (8) have shown that neuroleukin, a neuronal growth factor, is a lymphokine modulator of B-cell function. Since the relationship between the nervous and immune systems has been established, and since a growth-inducing molecule has been suggested in central nervous tissue (CNS) regeneration, I propose that lymphokines be tested as growth-inducing substances in spinal nerve regeneration experiments. Theoretical considerations
Aside from growth and cell proliferation, the immune system might be involved in regenera-
MEDICAL HYPOTHESES
tion in another way. The role of antibody attack of nerve fibers as a mechanism inhibiting regeneration was proposed by Feringa (5). However, the immune system, in attacking the nerves, may help to promote regeneration in an almost paradoxical way. A similar situation was shown to be true of the fetus by Mizel (12). The fetus, as the product of a genetic recombined cell - the fertilized egg - contains different genes, and hence antigens (Ags) that are not those of the mother, and, therefore, are attacked by her immune system. However, this attack is to the fetus’s benefit since the antibodies formed are “non-toxic,” or “enhancing” antibodies (ABs), which coat the fetal antigens on the trophoblast and actually protect them from recognition by maternal immune cells, thereby preventing their destruction. This type of scenario could operate in regeneration. If the normal proteins of exposed nerve endings in transected nerves are antigenic because they have never interfaced with immune cells. then it is possible, too, that some nerve proteins could be altered during the process of transection and the events of inflammation that follow it. Altered proteins in nerves after transection could be attacked by “non-toxic” antibodies. These “non-toxic” antibodies could act as decoys, drawing the immune attack away from normal nerve protein antigens, thereby protecting them from destruction. In addition, immune activation would result in more lymphokines which would thus stimulate regeneration. A similar scenario has been proposed by Mizel (12) to explain how tumors avoid destruction by the immune system. It is interesting to note the early work of Levi-Montalcini et al (10) who showed the groivth stimulating effects of mouse sarcoma on sympathetic nerves. If tumors decoy and tumors stimulate regeneration, couldn’t the actual mechanism of regeneration in nerves be the very same immune decoying I propose it is? Recently, Herlyn et al (9) showed that gamma interferon. (y-IF) induces or augments the expression of cell surface major histocompatibility complex (MHC) class II antigens. Antigen presentation would be necessary to stimulate the immune system to make antibodies against altered nerve antigens. It is interesting in this scenario that y-IF would play a dual role: antibody stimulation and regeneration. But why would the immune system attack its own nerves? Perhaps in the process of transection and subsequent inflammation axon proteins
Till:
IMMIINt:
SYSTEM
IN (‘ENTKAL
NERVOUS
SYSTEM
REGENERATION
would be modified and become antigenic. Yao et al (16) showed alterations in endoneurial lipids in Wallerian degeneration and regeneration. Similar changes in protein would not be surprising. Remember, too. that macrophages are involved in inflammation, and macrophages are antigen presenting cells (APCs). They could consume protein nerve debris after transection and. in digesting it, modify it and present it to the immune system as new antigen to be subsequently reacted against with “non-toxic” antibody. Evidence already cited (5. 15) indicates that steroids inhibit regeneration. Also, Gaster et al (6) showed that crushing stimulated regeneration. Consider these facts in the light of my immunologic hypothesis. Steroids inhibit regeneration by inhibiting immune cells. including antigen processing macrophages. Hence, lymphokine stimulation of regeneration decreases. Crushing stimulates regeneration by increasing inflammation. The increased inflammatory response. with its increased APCs, converts nerve protein to antigenic protein. Thus. the protective role of “non-toxic” antibody is elicited, as well as the regenerating activity of lymphokines. But why wouldn’t this scenario operate normally in CNS transected nerves, resulting in spontaneously occurring regeneration? The answer is myelin. In myelinated nerves the Schwann cells secrete neuronal growth factor, in unmyelinated nerves these must be supplied exogenously by lymphokine treatment, as I propose. In myelinated nerves, the myelin itself blocks immune recognition of antigenic nerve protein. In unmyelinated nerves, where antigenic nerve proteins exposed by transection are subject to attack by the immune system’, we would want to stimulate the production of “nontoxic” decoy antibodies to detract from the attack on normal nerve protein.
the decoying mechanism preferentially. Because of the above, I believe lymphokines should be tested in nerve regeneration models. To extend the lymphokine-regeneration hypothesis further, since lymphocytes are controlled by ion channels (4), it is possible that regulatory actions of both the immune and nervous systems could be modulated in the future by alteration of intracellular K+ or Ca++ ion ccncentrations by substances known to block ion channels. References I. Aguayo.
2. 3. 4.
5 _
6.
7. 8.
9.
10.
11.
Conclusion Why would increased stimulation of the immune response selectively stimulate only “non-toxic” antibodies? One could not ascertain this without experimentation. But since certain lymphokines (gamma-y-interferon) stimulate antigen presentation (9), it seems plausible that specifically modulated stimulation (of the antigen presenting cells) could lead to an increase in different AB idiotypes (including “non-toxic” AB to altered nerve proteins) rather than just a single idiotype to normal nerve proteins. This would enhance
15
A. How can nerve cells be encouraged to reestablish functional connections. Science News IZY: 2114. 1986. Bluestein, G. Nerve. immune link found on membranes. Science News 130: 118. 1986. Bullock, K. in R. Guillemin et al (eds.): Neural Modulation of Immunity. New York: Raven. p. I1 I. IYX5. Chandy. K. G., DeCoursey. T. E.. Cahalan. M. I>.. McLaughlin, C.. and Gupta. S. Voltage-gated potassium channels are required for human T lymphocyte activation. J. Exp. med. 160: 36Y. 1984. Feringa, E. R.. Johnson. R. D.. and Wendt. S. J. Spinal cord regeneration in rats after immunosuppressive treatment. Arch. Neural. 32: 676, 1975. Caster. R. N.. Davidson, T. M.. Rand, R. W., and Fonkalsrud, E. W. Comparison of nerve regeneration rates following controlled freezing and crushing. Arch. Surg. 103: 37X. 1971. Geschwind. N. Autoimmunity in left-handerc. Science 217: 141. 1982. Gurney. M. E.. Apatoff. B. R., Spear. G. T.. Baumel. M. J.. Antel. J. P., Brown, M.. Bania, M., Reder. A. T. Neuroleukin: a lymphokine product of lectin-stimulated T cells. Science 234: 574. lYS6. Herlyn. M., Guerry. D.. and Koprowskr. H. Rccombinant gamma-interferon induces changes in expression and shedding of antigens associated with normal human melanocytes. nevus cells. and primary and metastatic melanoma cells. J. of Immunology 134: 4226. IYX5. Levi-Montalcini. R.. Meyer, H.. and Hamburger. V. Irr vitro experiments on the effects of mouse sarcoma lX(I and 37 on the spinal and sympathetic ganglia of the chick embryo. Cancer Res. 14: 49. lYS4. Locke, S. and Colligan, D. The Healer Within. New York: E. P. Dutton. 1986.
12. Mizel, S. B. and Jaret. P. In Self-Defense. New York: Harcourt. Brace. Jovanovich, pp. 145, IYX5. 13. Rozengurt. E. Early signals in the mitogenic response. Science 234: 161. 19X6. 14. Samuelson. B. Leukotrienes: mediators of immediate hypcrscnsitivity reactions and inflammation. Science 210: 568. 1983. IS. Scheff. S. W.. Hoff, S. F.. and Anderson, K. J. Altered regulation of lesion-induced synaptogenesis by adrcnalectomy and corticosterone in young adult rats. Exp. Neurol. Y3: 456, 1986. 16. Yao. J. K.. Natarajan. V. and Dyck. P. J. The sequential alterations of endoneurial cholestereol and fatty acid in Wallerian degeneration and regeneration. J. Neurothem. 35: 933. 1980.