- Email: [email protected]
Effect of insulin, proinsulin and pancreatic extract on myelination and remyelination in organotypic nerve tissue in culture
Journal of the Neurological Sciences, 1985, 71:339-350 Elsevier
339
JNS 2590
Effect of Insulin, Proinsulin and Pancreatic Extract on Myelination and Remyelination in Organotypic Nerve Tissue in Culture German A. Roth, Vincent H. Jorgensen and Murray B. Bornstein The Saul R. Korey Department of Neurology and The Rose F. Kennedy Center for Research in Mental Retardation and Human Development, Albert Einstein College of Medicine of Yeshiva University, Bronx, NY
(U.S.A.) (Received 31 December, 1984) (Revised, received 5 August, 1985) (Accepted 7 August, 1985)
SUMMARY
The effect of insulin, proinsulin and crude pancreatic extract was studied in organotypic nerve tissue cultures, principally in relation to the development of myelin. Cultures were exposed to media supplemented with these substances beginning on the first day of explantation. By 4 days in vitro, there was a good neuritic outgrowth from all the fragments. That from the insulin and pancreatic extract-fed were more profuse and extended further than from the control group. By 8-12 days in vitro it was also possible to observe more myelinated axons in these treated groups. The pattern of changes in the myelin associated enzyme activity, 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNPase) paralleled the differential increase in myelination. Insulin-fed cultures showed a more rapid increase in CNPase activity, which, after 21 days in vitro reached a plateau about 3 0 - 5 0 ~ over that of the controls. Cultures treated with pancreatic extract showed a similar pattern of increased activity, while in proinsulintreated explants the activity was only significantly higher after 21 days in vitro. To study the effect of these substances on remyelination, well myelinated cultures were completely demyelinated by exposure to anti-white matter antiserum and were subsequently exposed to the same normal control or supplemented media. The amount of myelin and
This work was supported by Grant NS 11920 from N.I.H.G.A.R. was a recipient of a Research Fellowship from the Consejo Nacional de Investigaciones Cientificas y Tecnicas, Argentina. Address correspondence and reprint request to: Dr. Murray B. Bornstein, Department of Neurology (K-401), Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, U.S.A., Tel. (212) 430-2509. 0022-510X/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
340 concomitantly the CNPase activity increased rapidly and in the same proportion between the various groups as was observed previously during primary myelination. Insulin as well as crude pancreatic extract and, to some extent, proinsulin demonstrated a marked effect on the time of onset and principally on the total amount of myelin developed by treated cultures as compared to those maintained in normal nutrient medium.
Key words:
Insulin
-
Myelination
-
Nerve
tissue -
Pancreatic
extract
-
Proinsulin
-
Remyelination
INTRODUCTION
The development and maintenance of organotypic cultures of mammalian nerve tissues in the Maximow slide assembly are enhanced by the presence of serum, embryonic extract and other additives to the medium (Herschman 1974; Waymouth 1977). These active components provide greater nutritional support as well as other factors which may be required for the attachment, survival, proliferation and differentiation of the cultured tissue. Several approaches have been taken to eliminate serum requirements by adapting cultures to media without serum, or by the addition of active components of serum to the medium (Waymouth 1977; Obata 1981 ; Huck 1983). Among these components, insulin and insulin-related substances have been extensively examined. It is well established that insulin exerts a marked effect on carbohydrate utilization, protein, RNA and DNA synthesis as well as on lipogenesis (reviewed by Pilkis and Park 1974). The results obtained in CNS cultures exposed to insulin support the idea that insulin may be a growth stimulant and survival factor for primary cultures of cells dissociated from embryonic brain (Snyder and Kim 1980; Sotelo et al. 1980; Obata 1981; Huck 1983; Bhat et al. 1983). In this regard, to study the effect of insulin and insulin-related substances on the CNS development of organotypic cultures of nerve tissue, particularly on myelination, insulin, proinsulin and a preparation of crude pancreatic extract were added to the usual culture nutrient medium. Cultures were exposed to these media from the fit'st day of explantation in order to study their effect on primary myelination or on remyelination in cultures which had previously been demyelinated by exposure to anti-white matter antiserum. The development of myelin was determined by light-microscopic examination of the living explants and biochemically by measuring the activity of the myelin associated enzyme 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNPase, 3.1.4.37). EXPERIMENTAL PROCEDURES Tissue cultures
Cultures of 13-14-day-old embryo mouse spinal cord without meninges were prepared and maintained in Maximow slide assemblies (Bornstein 1973). The normal
341 nutrient medium (NNM) consisted of 33 ~ human placental serum (heat-inactivated at 56 °C for 30 min), 50~0 Eagle's minimum essential medium plus L-glutamine, 17~o Simms' balanced salt solution and 40 mM glucose. Chick embryo extract was excluded.
Myelination On the In'st day in vitro (DIV) cultures were randomly distributed into 4 groups. The f'trst received the NNM; the second, NNM supplemented with 10-6g/ml crystalline bovine insulin*; the third, NNM supplemented with 10-6g/ml bovine proinsulin*; the fourth, NNM supplemented with 10 - 6 g/ml crude pancreatic extract*. Thereafter, all cultures were maintained on their particular nutrient solutions. The medium was changed twice weekly. Samples were taken at 7, 14, 21, 28, and 35 DIV. The degree of myelination was determined in a double blind fashion by lightmicroscopic observation of the living explants and scored from 0, meaning no visible myelinated fibers on the surface of the explant to 5 for cultures completely myelinated showing a dense network of myelinated fibers in all the exposed areas; 2, 3, and 4 scored the increasing intermediate amount of myelinated fibers.
Remyelination Cultures were maintained in NNM. After 18 DIV, when cultures were well myelinated, they were exposed for 2 days to NNM supplemented with rabbit anti-white matter antiserum and guinea pig serum as a source of complement at a final concentration in the medium of 25 ~o and 10 ~o, respectively. This produced total demyelination of all cultures. Cultures were then washed free of the antiserum and distributed into 4 groups which were exposed to the same media as was indicated previously for the myelination studies.
Spinal cords Spinal cords were obtained from 13-day-old fetuses and 1, 8, 15, 22 and 29-day-old-mice, which corresponded approximately in age to the tissue culture samples. Adult spinal cords were taken from 3-4-month-old female mice.
Biochemical analysis At the time of sampling, 8-10 cultured fragments in each group were peeled from the collagen-coated coverslips, pooled, and ultrasonically homogenized in 0.9~ w/v sodium chloride (Roth et al. 1983). Aliquots of the homogenates were removed for protein determination (Lowry et al. 1951) and CNPase activity by the method of Prohaska et al. (1973) as previously described (Roth et al. 1983). For the determination of CNPase activity, the homogenates were first treated with sodium deoxycholate. One unit of enzyme activity was defined as the amount that produced 1 #tool of 2'-AMP from 2',3'-cyclic AMP/min under the experimental conditions.
* Insulin, proinsulin and crude pancreatic extract were supplied through the courtesy of Eli Lilly Laboratories, Indiana,U.S.A.
342 RESULTS
Myelination Cultures were divided into 4 groups and maintained in NNM or in NNM supplemented with insulin (INS), proinsulin (PROINS) or pancreatic extract (PEXT). All cultures were observed daily by light microscopy at 600 diameters magnification under ordinary bright-field illumination. By 4 days in vitro, there was a good outgrowth, principally neuritic, from all the fragments. That from the INS and PEXT-fed (Fig. 1B) was more profuse and extended significantly further than in the NNM group (Fig. 1A). On the basis of the length of the neuritic outgrowth, there would be no difficulty at that timepoint in distinguishing the NNM from the INS and PEXT groups. Cultures treated with PROINS had a pattern variable and intermediate between the INS and NNM groups. After 8 DIV, it was possible to observe some myelinated axons in the NNM-fed cultures whereas the other, treated cultures had appreciably more myelin (Fig. 2B). By 12 DIV, PEXT cultures had a profuse outgrowth offme neuritic processes, less in INS and PROINS groups, and practically none in the NNM group. All cultures were myelinating. The number of myelinated axons appeared greater in the treated groups principally in INS and PEXT-fed cultures (Fig. 2B). At 19-20 DIV some central necrosis was clearly distinguishable in most fragments and the outgrowth of neurites had decreased markedly. The differences among the groups in myelin and outgrowth seemed to narrow in the following days and between 30 and 35 DIV all fragments contained about the same amount of myelinated axons as observed by light microscopy. The explants also looked more granular and some myelin breakdown was observed. Determination of total protein per explant (Fig. 2A) showed that the amount of protein increased during the first week in vitro coincident with the major development and maturation ofthe tissue. No significant difference was observed among the different groups, although in some experiments protein decreased in all after the third week in vitro. In experiments where this decrease was observed, concomitant flattening and faster degeneration of the tissue was also evident in the light microscope. Determination of CNPase activity, previously demonstrated to be a reliable index of myelination and demyelination of CNS tissue (Roth et al. 1983; Roth and Bornstein 1984), yielded a more accurate and quantitative evaluation of the effect of insulin and the related substances. The pattern of changes in the CNPase activity developed in parallel with the increase in the myelination (Fig. 2C). INS-fed cultures at 14 DIV showed an increase of about 90~o (P < 0.01) in CNPase activity. After 21 DIV the CNPase activity reached a plateau, about 30-50~o over that of the NNM-fed cultures. Cultures treated with PEXT showed a similar pattern of increased activity, while in PROINS-fed explants the activity was more variable and significant difference with respect to NNM were not seen until 28 DIV ( + 20~o, P < 0.05). To rule out the possibility that the effect induced by insulin and pancreatic extract might be merely a direct activation of the CNPase activity rather than on the total process of myelination, some experiments were carried on in vitro (Table 1). For this purpose, determinations of CNPase activity in NNM-maintained cultures (19 DIV) or in adult mouse spinal cord were performed in the presence of either a detergent, such
343
Fig. 1. Brightfield photomicrographs of living, unstained spinal cord cultures ( × 40). A: 4 DIV culture maintained in NNM; B: 4 DIV INS-fed explant.
as deoxycholate (regular condition), or insulin, or both. N o activation or similarity to the activation induced by deoxycholate was observed following the addition o f insulin to the cultured or spinal cord tissues. In addition, the C N P a s e activity was measured in 1 : 1 mixtures of N N M - m a i n t a i n e d explants with those previously maintained in
14
A
[ ~NNM / [] , ~ , ~2i- •,,,o,,~ lo
II1 In
In
i
|
0
1.6
1.2
| 1.0
.¢,
0.8
~ 0.6 o. 0.4
0.2
0
7
, 14
28 DIV
Fig. 2. Myelination. The degree of myelination in each point was determined in a group of 8-10 explants. After homogenization, total protein and CNPase activity were determined on the same explants. The amount of myelin was scored by light-microscopic examination from 0, meaning no visible myelinated fibers in the surface of the explant, to 5 for cultures heavily myelinated. Each point is the mean of 6 individual determinations and bars are the SE of these means. P was obtained by one-way ANOVA analysis of the variance. The probability of significance with respect to the NNM group was: * P < 0.001; ** P < 0.01; • ** P < 0.02; **** P < 0.05; ***** P < 0.10.
345 TABLE 1 CNPase ACTIVITYIN CNS MOUSE SPINAL CORD TISSUES IN DIFFERENT CONDITIONS For determination of CNPase activity,the different homogenates were subjected to previous treatment as indicated. Each value represents the mean of 3 separated experiments. Group
DIV
Preincubate medium
CNPase activity (units/mg protein)
NNM NNM NNM NNM
19 19 19 19
Water Deoxycholate Insulin Deoxycholate + insulin
0.45 0.72 0.25 0.75
Adult spinal cord Adult spinal cord Adult spinal cord Adult spinal cord
-----
Water Deoxycholate Insulin Deoxycholate + insulin
4.82 10.22 4.27 10.46
NNM INS NNM + INS (1 : 1)
14 14 14
Deoxycholate Deoxycholate Deoxycholate
0.40 0.75 0.60
NNM INS NNM + INS (1 : 1)
21 21 21
Deoxycholate Deoxycholate Deoxycholate
0.90 1.16 1.05
N N M plus insulin. The values obtained in these mixtures coincided with the expected activity as related to the differing amounts of myelin, but without any further direct activation of the enzyme. Similar results were obtained with mixtures of tissues maintained 14 DIV in their respective media, when the effect of insulin is beginning to occur or with tissues at 21 DIV, when the plateau of myelination had been reached. In other words, the values obtained in these mixtures demonstrated that the addition of insulin or insulin-treated tissue to normal control NNM-fed cultures did not affect the latters' CNPase activity.
Remyelination Cultures maintained for 18-19 DIV in N N M were completely demyelinated by exposure to anti-white matter antiserum in the presence of complement. The loss of myelin was accompanied by a decrease of about 70-75% in the CNPase activity, without any significant change in the amount of total protein per explant (Fig. 3). After demyelination, the explants were exposed to N N M , INS, P R O I N S or PEXT media. After 2 - 3 days' exposure, some myelinated fibers became visible, principally in the INS-treated group. At the same time, CNPase activity increased about 50~o (P < 0.01) as compared to NNM-fed cultures. The amount of myelin and concomitantly the CNPase activity increased rapidly and in the same proportion among the various groups as had been observed previously during primary myelination. By 8 - 9 days after demyelination, the different groups reached the same degree of CNPase activity as was
14
A ~ EAE/NNM EAE/INS ~ EAE/IROINS EAE/PEXT ~1 NM EAE
12
lO
i
Z
8
4
2
0
1.8
i
i
i
i
i
i
6~7
8:9
10~11
B
1.4
1.2
i
1.0
0,8
~ g.
~
0.6
0.4
0.2
/
0
0
2-'3
4-5
13~14
DIV after demyelinatlon Fig. 3. Remyelination. Cultures maintained in normal nutrient medium (NNM) ([S]) were exposed to anti-white matter antiserum. After 48 h exposure (Day 0), the demyelinate cultures (m), were fed with N N M alone or supplemented with insulin (INS), proinsulin (PROINS) and crude pancreatic extract (PEXT). Each point is the mean of 5 determinations and bars are the SE of the means. P was calculated by one-way A N O V A analysis of the variance. The probability of significance with respect to the E A E / N N M group was: * P < 0.001; ** P < 0.01; *** P < 0.02; **** P < 0.05; ***** P < 0.10.
347 present in sister cultures (28 DIV) that had not previously been demyelinated. After this time, all explants, independent of the medium, began to show some degeneration. Loss of ceils, some myelin breakdown and decrease in total protein content and CNPase activity were common features. DISCUSSION
Neural tissue culture would seem to be well suited for exploring the nutritional and humoral needs of an intact developing nervous system. The difficulty, however, in maintaining organized tissue in vitro in the absence of complex and poorly characterized biological fluids; e.g., serum, has precluded the realization of such potential. One of the challenges is to identify growth factors and their possible effects on myelin formation and maintenance. Although organotypic nerve tissue cultures possess all the CNS parenchymal cellular components, the amount of myelin produced is significantly less than in comparable tissues in vivo. Spinal cord in vivo has a higher level of total lipids ( + 313 ~o) than spinal cord in vitro. Nevertheless, although the total lipid content is less, individual lipids in tissue culture are represented in the same ratio distribution as that found in vivo, except for changes in the contents of a few lipids, principally in the myelin-related galactolipids which are relatively decreased (-42~o) (Roth et al. 1983). Also, in organ cultures of rat cerebellum, the phase of relative galactolipid enrichment is delayed for at least 6 weeks, whereas during normal myelination in vivo myelin galactolipids increase from 14 days post-partum (Bradbury and Lumsden 1979). Correlative quantification of the CNPase activity in spinal cord tissue in vivo and in vitro (Fig. 4) showed that, although the onset and development of myelination is similar in both tissues, adult spinal cord in vivo reached a total activity about 10-fold higher than the corresponding tissue maintained in vitro. This underdevelopment in tissue culture principally in the amount of myelin might be related to the absence in the nutrient medium of active factors which might contribute to the maturation and maintenance of the cultured tissue. One group of substances widely being studied for trophic influences is insulin and insulin-related compounds. The effects exerted by insulin have been examined principally in muscle but also in mammary gland, liver, adipose tissue and on cultured cells (reviewed by Pilkis and Park 1974). Insulin appears to stimulate synthesis of RNA and proteins, has a marked effect on carbohydrate transport and utilization, inhibits lipolysis, and on cAMP-mediated processes in general. Insulin also has a dramatic stimulatory effect on the growth of cultured fibroblasts (Jimenez de Asua et al. 1973), neurons (Snyder and Kim 1980; Sotelo et al. 1980) and for the general survival and proliferation of brain cells (Obata 1981; Bhat et al. 1983; Huck 1983). Recently, synthesis of insulin by brain cells has been suggested (Rosenzweig et al. 1980; Birch et al. 1984) and specific insulin binding sites have been demonstrated in brain tissue (Havrankova et al. 1978) as well as in cultures of fetal brain cells (Raizada et al. 1980). However, data concerning the effect of insulin on the complex intercellular activities needed for myelinogenesis have not appeared. Although determinations of additional myelin-related parameters; e.g., myelin
348 14
12
10
I
i
j6 4
0
7
14
21
28
35 Days in vitro
Fig. 4. CNPase determination. CNPase activity was quantified in nerve tissue cultures (broken line) and in spinal cords from animals of correspondent age (solid line).
basic protein, might be desirable to certify the increase in myelination, as determined in these experiments, insulin as well as crude pancreatic extract seem to have a marked effect on the time of onset and principally on the total amount of myelin produced in these cultures as compared to those maintained in a medium without insulin (see Fig. 2). Proinsulin had a similar but smaller effect. Although experiments to elucidate the mechanism of insulin action in these cultures were not performed, it seems that insulin had no direct effect on the CNPase activity in vitro nor did it induce a secondary compound, which might alter CNPase activity in normal control tissue (see Table 1). This effect does not seem to be exclusive to INS considering that PEXT that has little insulin was able to induce a similar effect. Conversely, PROINS had a lower or no effect, which could be due principally to variable endogenous production of insulin from proinsulin. The effect of these compounds was evident on primary myelinogenesis as on remyelination although they
349 could not prevent some deterioration of the tissue similar to that observed in the NNM-fed group (see Fig. 3). It is possible that the lower values for CNPase activity in cultures as compared to tissues in vivo might result in part from a lack of some factors such as insulin. Therefore, the addition of these factors might produce a maturation pattern more closely resembling those found in vivo. Similar effects could also explain the lower lipid amounts observed in CNS cultures by Latovitzki and Silberberg (1973) and Satomi and Kishimoto (1981) whose cultures were grown in the absence of insulin in comparison to Giesing and Zilliken (1976) which supplemented their medium with insulin. Current studies are extending these experiments to demonstrate the effect of insulin and related substances on CNS tissue in vivo, particularly for their possible effects on remyelination in experimental allergic encephalomyelitis and its implications for immunologically-mediated human demyelinating diseases of the central and peripheral nervous systems. REFERENCES Bhat, N.R., G. Shanker and R.A. Pieringer (1983) Cell proliferation in growing cultures of dissociated embryonic mouse brain - - Macromolecule and ornithine decarboxylase synthesis and regulation by hormones and drugs, J. Neurosci. Res., 10: 221-230. Birch, N.P., D.L. Christie and A. G. C. Renwick (1984) Immunoreactive insulin from mouse brain cells in culture and whole rat brain, Biochem. J., 218: 19-27. Bornstein, M.B. (1973) Organotypic mammalian central and peripheral nerve tissue. In: P.F. Kruse and M.K. Patterson (Eds.), Tissue Culture - - Methods and Applications, Academic Press, New York, pp. 86-92. Bradbury, K. and C.E. Lumsden (1979) The chemical composition of myelin in organ cultures of rat cerebellum, J. Neurochem., 32: 145-154. Giesing, M. and F. Zilliken (1976) Analysis of hpid components in organotypic cultures of cerebellum during development, Brain Res., 111: 212-219. Havrankova, J., J. Roth and M. Brownstein (1978) Insulin receptors are widely distributed in the central nervous system of the rat, Nature (Lond.), 272: 827-829. Hershman, H.R. (1974) Culture of neural tissue and cells. In: N. Marks and R. Rodnight (Eds.), Research Methods in Neurochemistry, Vol. 2, Plenum Press, New York, pp. 101-160. Huck, S. (1983) Serum-free medium for cultures of the postnatal mouse cerebellum - - Only insulin is essential, Brain Res. Bull., 10: 667-674. Jimenez de Asua, L., E. S. Sudan, M. M. Flawia and H.N. Torres (1973) Effect of insulin on the growth pattern and adenylate cyclase activity of BHK fibroblasts, Proc. Nat. Acad. Sci. (U.S.A.), 70: 1388-1392. Latovitzki, N. and D.H. Silberberg (1973) Quantification of galactolipids in myelinating cultures of rat cerebellum, J. Neurochem., 20: 1771-1776. Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall (1951) Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193: 265-275. Obata, K. (I 981) Requirements for growth of neurites from embryonic chick cerebellum in culture, Develop. Brain Res., 1: 444-449. Pilkis, S.J. and C.R. Park (1974) Mechanism of action of insulin, Ann. Rev. Pharmacol., 14: 365-388. Prohaska, J. R., D.A. Clark and W.W. Wells (1973) Improved rapidity and precision in the determination of brain 2' ,Y-cyclic nucleotide Y-phosphohydrolase, Anal. Biochem., 56: 275-282. Raizada, M.K., J.W. Yang and R.E. Fellows (1980) Binding of [125I]insulin to specific receptors and stimulation of nucleotide incorporation in cells cultures from rat brain, Brain Res., 200: 389-400. Rosenzweig, J. L., J. Havrankova, M.A. Lesniak, M. Bronstein and J. Roth (1980) Insulin is ubiquitous in extrapancreatic tissues of rats and humans, Proc. Nat. Acad. Sci. (U.S.A.), 77: 572-576. Roth, G. A., R. K. Yu and M. B. Bornstein (1983) Chemical analysis of organotypic cultures of mouse spinal cord in normal, demyelinative and nondemyelinative conditions, J. Neurochem., 41: 1710-1717.
350 Roth, G. A. and M. B. Bornstein (1984) Quantitative chemical analysis of demyclination m cultured mouse spinal cord, Develop. Brain Res., 15: 105-111. S atomi, D. and Y. Kishimoto (198 I) Change ofgalactolipids and metabolism of fatty acids in the organotypic culture of myelinating mouse brain, Biochim. Biophys. Aeta, 666: 446-454. Snyder, E.Y. and S.U. Kim (1980) Insulin - - Is it a nerve survival factor? Brain Res., 196: 565-571. Sotelo, J., J.G. Clarence, D. Carleton Gajdusek, B.H. Toh and M. Wurth (1980) Method for preparing cultures of central neurons - - Cytochemical and immunochemical studies, Proc. Nat. Acad. Sei. (U.S.A.), 77: 653-657. Waymouth, C. (1977) Nutritional requirements of cells in culture, with special reference to neural cells. In: S. Fedoroffand L. Hertz (Eds.), Cell, Tissue and Organ Cultures in Neurobiology, Academic Press, New York, pp. 631-648.
339
JNS 2590
Effect of Insulin, Proinsulin and Pancreatic Extract on Myelination and Remyelination in Organotypic Nerve Tissue in Culture German A. Roth, Vincent H. Jorgensen and Murray B. Bornstein The Saul R. Korey Department of Neurology and The Rose F. Kennedy Center for Research in Mental Retardation and Human Development, Albert Einstein College of Medicine of Yeshiva University, Bronx, NY
(U.S.A.) (Received 31 December, 1984) (Revised, received 5 August, 1985) (Accepted 7 August, 1985)
SUMMARY
The effect of insulin, proinsulin and crude pancreatic extract was studied in organotypic nerve tissue cultures, principally in relation to the development of myelin. Cultures were exposed to media supplemented with these substances beginning on the first day of explantation. By 4 days in vitro, there was a good neuritic outgrowth from all the fragments. That from the insulin and pancreatic extract-fed were more profuse and extended further than from the control group. By 8-12 days in vitro it was also possible to observe more myelinated axons in these treated groups. The pattern of changes in the myelin associated enzyme activity, 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNPase) paralleled the differential increase in myelination. Insulin-fed cultures showed a more rapid increase in CNPase activity, which, after 21 days in vitro reached a plateau about 3 0 - 5 0 ~ over that of the controls. Cultures treated with pancreatic extract showed a similar pattern of increased activity, while in proinsulintreated explants the activity was only significantly higher after 21 days in vitro. To study the effect of these substances on remyelination, well myelinated cultures were completely demyelinated by exposure to anti-white matter antiserum and were subsequently exposed to the same normal control or supplemented media. The amount of myelin and
This work was supported by Grant NS 11920 from N.I.H.G.A.R. was a recipient of a Research Fellowship from the Consejo Nacional de Investigaciones Cientificas y Tecnicas, Argentina. Address correspondence and reprint request to: Dr. Murray B. Bornstein, Department of Neurology (K-401), Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461, U.S.A., Tel. (212) 430-2509. 0022-510X/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
340 concomitantly the CNPase activity increased rapidly and in the same proportion between the various groups as was observed previously during primary myelination. Insulin as well as crude pancreatic extract and, to some extent, proinsulin demonstrated a marked effect on the time of onset and principally on the total amount of myelin developed by treated cultures as compared to those maintained in normal nutrient medium.
Key words:
Insulin
-
Myelination
-
Nerve
tissue -
Pancreatic
extract
-
Proinsulin
-
Remyelination
INTRODUCTION
The development and maintenance of organotypic cultures of mammalian nerve tissues in the Maximow slide assembly are enhanced by the presence of serum, embryonic extract and other additives to the medium (Herschman 1974; Waymouth 1977). These active components provide greater nutritional support as well as other factors which may be required for the attachment, survival, proliferation and differentiation of the cultured tissue. Several approaches have been taken to eliminate serum requirements by adapting cultures to media without serum, or by the addition of active components of serum to the medium (Waymouth 1977; Obata 1981 ; Huck 1983). Among these components, insulin and insulin-related substances have been extensively examined. It is well established that insulin exerts a marked effect on carbohydrate utilization, protein, RNA and DNA synthesis as well as on lipogenesis (reviewed by Pilkis and Park 1974). The results obtained in CNS cultures exposed to insulin support the idea that insulin may be a growth stimulant and survival factor for primary cultures of cells dissociated from embryonic brain (Snyder and Kim 1980; Sotelo et al. 1980; Obata 1981; Huck 1983; Bhat et al. 1983). In this regard, to study the effect of insulin and insulin-related substances on the CNS development of organotypic cultures of nerve tissue, particularly on myelination, insulin, proinsulin and a preparation of crude pancreatic extract were added to the usual culture nutrient medium. Cultures were exposed to these media from the fit'st day of explantation in order to study their effect on primary myelination or on remyelination in cultures which had previously been demyelinated by exposure to anti-white matter antiserum. The development of myelin was determined by light-microscopic examination of the living explants and biochemically by measuring the activity of the myelin associated enzyme 2',3'-cyclic nucleotide 3'-phosphohydrolase (CNPase, 3.1.4.37). EXPERIMENTAL PROCEDURES Tissue cultures
Cultures of 13-14-day-old embryo mouse spinal cord without meninges were prepared and maintained in Maximow slide assemblies (Bornstein 1973). The normal
341 nutrient medium (NNM) consisted of 33 ~ human placental serum (heat-inactivated at 56 °C for 30 min), 50~0 Eagle's minimum essential medium plus L-glutamine, 17~o Simms' balanced salt solution and 40 mM glucose. Chick embryo extract was excluded.
Myelination On the In'st day in vitro (DIV) cultures were randomly distributed into 4 groups. The f'trst received the NNM; the second, NNM supplemented with 10-6g/ml crystalline bovine insulin*; the third, NNM supplemented with 10-6g/ml bovine proinsulin*; the fourth, NNM supplemented with 10 - 6 g/ml crude pancreatic extract*. Thereafter, all cultures were maintained on their particular nutrient solutions. The medium was changed twice weekly. Samples were taken at 7, 14, 21, 28, and 35 DIV. The degree of myelination was determined in a double blind fashion by lightmicroscopic observation of the living explants and scored from 0, meaning no visible myelinated fibers on the surface of the explant to 5 for cultures completely myelinated showing a dense network of myelinated fibers in all the exposed areas; 2, 3, and 4 scored the increasing intermediate amount of myelinated fibers.
Remyelination Cultures were maintained in NNM. After 18 DIV, when cultures were well myelinated, they were exposed for 2 days to NNM supplemented with rabbit anti-white matter antiserum and guinea pig serum as a source of complement at a final concentration in the medium of 25 ~o and 10 ~o, respectively. This produced total demyelination of all cultures. Cultures were then washed free of the antiserum and distributed into 4 groups which were exposed to the same media as was indicated previously for the myelination studies.
Spinal cords Spinal cords were obtained from 13-day-old fetuses and 1, 8, 15, 22 and 29-day-old-mice, which corresponded approximately in age to the tissue culture samples. Adult spinal cords were taken from 3-4-month-old female mice.
Biochemical analysis At the time of sampling, 8-10 cultured fragments in each group were peeled from the collagen-coated coverslips, pooled, and ultrasonically homogenized in 0.9~ w/v sodium chloride (Roth et al. 1983). Aliquots of the homogenates were removed for protein determination (Lowry et al. 1951) and CNPase activity by the method of Prohaska et al. (1973) as previously described (Roth et al. 1983). For the determination of CNPase activity, the homogenates were first treated with sodium deoxycholate. One unit of enzyme activity was defined as the amount that produced 1 #tool of 2'-AMP from 2',3'-cyclic AMP/min under the experimental conditions.
* Insulin, proinsulin and crude pancreatic extract were supplied through the courtesy of Eli Lilly Laboratories, Indiana,U.S.A.
342 RESULTS
Myelination Cultures were divided into 4 groups and maintained in NNM or in NNM supplemented with insulin (INS), proinsulin (PROINS) or pancreatic extract (PEXT). All cultures were observed daily by light microscopy at 600 diameters magnification under ordinary bright-field illumination. By 4 days in vitro, there was a good outgrowth, principally neuritic, from all the fragments. That from the INS and PEXT-fed (Fig. 1B) was more profuse and extended significantly further than in the NNM group (Fig. 1A). On the basis of the length of the neuritic outgrowth, there would be no difficulty at that timepoint in distinguishing the NNM from the INS and PEXT groups. Cultures treated with PROINS had a pattern variable and intermediate between the INS and NNM groups. After 8 DIV, it was possible to observe some myelinated axons in the NNM-fed cultures whereas the other, treated cultures had appreciably more myelin (Fig. 2B). By 12 DIV, PEXT cultures had a profuse outgrowth offme neuritic processes, less in INS and PROINS groups, and practically none in the NNM group. All cultures were myelinating. The number of myelinated axons appeared greater in the treated groups principally in INS and PEXT-fed cultures (Fig. 2B). At 19-20 DIV some central necrosis was clearly distinguishable in most fragments and the outgrowth of neurites had decreased markedly. The differences among the groups in myelin and outgrowth seemed to narrow in the following days and between 30 and 35 DIV all fragments contained about the same amount of myelinated axons as observed by light microscopy. The explants also looked more granular and some myelin breakdown was observed. Determination of total protein per explant (Fig. 2A) showed that the amount of protein increased during the first week in vitro coincident with the major development and maturation ofthe tissue. No significant difference was observed among the different groups, although in some experiments protein decreased in all after the third week in vitro. In experiments where this decrease was observed, concomitant flattening and faster degeneration of the tissue was also evident in the light microscope. Determination of CNPase activity, previously demonstrated to be a reliable index of myelination and demyelination of CNS tissue (Roth et al. 1983; Roth and Bornstein 1984), yielded a more accurate and quantitative evaluation of the effect of insulin and the related substances. The pattern of changes in the CNPase activity developed in parallel with the increase in the myelination (Fig. 2C). INS-fed cultures at 14 DIV showed an increase of about 90~o (P < 0.01) in CNPase activity. After 21 DIV the CNPase activity reached a plateau, about 30-50~o over that of the NNM-fed cultures. Cultures treated with PEXT showed a similar pattern of increased activity, while in PROINS-fed explants the activity was more variable and significant difference with respect to NNM were not seen until 28 DIV ( + 20~o, P < 0.05). To rule out the possibility that the effect induced by insulin and pancreatic extract might be merely a direct activation of the CNPase activity rather than on the total process of myelination, some experiments were carried on in vitro (Table 1). For this purpose, determinations of CNPase activity in NNM-maintained cultures (19 DIV) or in adult mouse spinal cord were performed in the presence of either a detergent, such
343
Fig. 1. Brightfield photomicrographs of living, unstained spinal cord cultures ( × 40). A: 4 DIV culture maintained in NNM; B: 4 DIV INS-fed explant.
as deoxycholate (regular condition), or insulin, or both. N o activation or similarity to the activation induced by deoxycholate was observed following the addition o f insulin to the cultured or spinal cord tissues. In addition, the C N P a s e activity was measured in 1 : 1 mixtures of N N M - m a i n t a i n e d explants with those previously maintained in
14
A
[ ~NNM / [] , ~ , ~2i- •,,,o,,~ lo
II1 In
In
i
|
0
1.6
1.2
| 1.0
.¢,
0.8
~ 0.6 o. 0.4
0.2
0
7
, 14
28 DIV
Fig. 2. Myelination. The degree of myelination in each point was determined in a group of 8-10 explants. After homogenization, total protein and CNPase activity were determined on the same explants. The amount of myelin was scored by light-microscopic examination from 0, meaning no visible myelinated fibers in the surface of the explant, to 5 for cultures heavily myelinated. Each point is the mean of 6 individual determinations and bars are the SE of these means. P was obtained by one-way ANOVA analysis of the variance. The probability of significance with respect to the NNM group was: * P < 0.001; ** P < 0.01; • ** P < 0.02; **** P < 0.05; ***** P < 0.10.
345 TABLE 1 CNPase ACTIVITYIN CNS MOUSE SPINAL CORD TISSUES IN DIFFERENT CONDITIONS For determination of CNPase activity,the different homogenates were subjected to previous treatment as indicated. Each value represents the mean of 3 separated experiments. Group
DIV
Preincubate medium
CNPase activity (units/mg protein)
NNM NNM NNM NNM
19 19 19 19
Water Deoxycholate Insulin Deoxycholate + insulin
0.45 0.72 0.25 0.75
Adult spinal cord Adult spinal cord Adult spinal cord Adult spinal cord
-----
Water Deoxycholate Insulin Deoxycholate + insulin
4.82 10.22 4.27 10.46
NNM INS NNM + INS (1 : 1)
14 14 14
Deoxycholate Deoxycholate Deoxycholate
0.40 0.75 0.60
NNM INS NNM + INS (1 : 1)
21 21 21
Deoxycholate Deoxycholate Deoxycholate
0.90 1.16 1.05
N N M plus insulin. The values obtained in these mixtures coincided with the expected activity as related to the differing amounts of myelin, but without any further direct activation of the enzyme. Similar results were obtained with mixtures of tissues maintained 14 DIV in their respective media, when the effect of insulin is beginning to occur or with tissues at 21 DIV, when the plateau of myelination had been reached. In other words, the values obtained in these mixtures demonstrated that the addition of insulin or insulin-treated tissue to normal control NNM-fed cultures did not affect the latters' CNPase activity.
Remyelination Cultures maintained for 18-19 DIV in N N M were completely demyelinated by exposure to anti-white matter antiserum in the presence of complement. The loss of myelin was accompanied by a decrease of about 70-75% in the CNPase activity, without any significant change in the amount of total protein per explant (Fig. 3). After demyelination, the explants were exposed to N N M , INS, P R O I N S or PEXT media. After 2 - 3 days' exposure, some myelinated fibers became visible, principally in the INS-treated group. At the same time, CNPase activity increased about 50~o (P < 0.01) as compared to NNM-fed cultures. The amount of myelin and concomitantly the CNPase activity increased rapidly and in the same proportion among the various groups as had been observed previously during primary myelination. By 8 - 9 days after demyelination, the different groups reached the same degree of CNPase activity as was
14
A ~ EAE/NNM EAE/INS ~ EAE/IROINS EAE/PEXT ~1 NM EAE
12
lO
i
Z
8
4
2
0
1.8
i
i
i
i
i
i
6~7
8:9
10~11
B
1.4
1.2
i
1.0
0,8
~ g.
~
0.6
0.4
0.2
/
0
0
2-'3
4-5
13~14
DIV after demyelinatlon Fig. 3. Remyelination. Cultures maintained in normal nutrient medium (NNM) ([S]) were exposed to anti-white matter antiserum. After 48 h exposure (Day 0), the demyelinate cultures (m), were fed with N N M alone or supplemented with insulin (INS), proinsulin (PROINS) and crude pancreatic extract (PEXT). Each point is the mean of 5 determinations and bars are the SE of the means. P was calculated by one-way A N O V A analysis of the variance. The probability of significance with respect to the E A E / N N M group was: * P < 0.001; ** P < 0.01; *** P < 0.02; **** P < 0.05; ***** P < 0.10.
347 present in sister cultures (28 DIV) that had not previously been demyelinated. After this time, all explants, independent of the medium, began to show some degeneration. Loss of ceils, some myelin breakdown and decrease in total protein content and CNPase activity were common features. DISCUSSION
Neural tissue culture would seem to be well suited for exploring the nutritional and humoral needs of an intact developing nervous system. The difficulty, however, in maintaining organized tissue in vitro in the absence of complex and poorly characterized biological fluids; e.g., serum, has precluded the realization of such potential. One of the challenges is to identify growth factors and their possible effects on myelin formation and maintenance. Although organotypic nerve tissue cultures possess all the CNS parenchymal cellular components, the amount of myelin produced is significantly less than in comparable tissues in vivo. Spinal cord in vivo has a higher level of total lipids ( + 313 ~o) than spinal cord in vitro. Nevertheless, although the total lipid content is less, individual lipids in tissue culture are represented in the same ratio distribution as that found in vivo, except for changes in the contents of a few lipids, principally in the myelin-related galactolipids which are relatively decreased (-42~o) (Roth et al. 1983). Also, in organ cultures of rat cerebellum, the phase of relative galactolipid enrichment is delayed for at least 6 weeks, whereas during normal myelination in vivo myelin galactolipids increase from 14 days post-partum (Bradbury and Lumsden 1979). Correlative quantification of the CNPase activity in spinal cord tissue in vivo and in vitro (Fig. 4) showed that, although the onset and development of myelination is similar in both tissues, adult spinal cord in vivo reached a total activity about 10-fold higher than the corresponding tissue maintained in vitro. This underdevelopment in tissue culture principally in the amount of myelin might be related to the absence in the nutrient medium of active factors which might contribute to the maturation and maintenance of the cultured tissue. One group of substances widely being studied for trophic influences is insulin and insulin-related compounds. The effects exerted by insulin have been examined principally in muscle but also in mammary gland, liver, adipose tissue and on cultured cells (reviewed by Pilkis and Park 1974). Insulin appears to stimulate synthesis of RNA and proteins, has a marked effect on carbohydrate transport and utilization, inhibits lipolysis, and on cAMP-mediated processes in general. Insulin also has a dramatic stimulatory effect on the growth of cultured fibroblasts (Jimenez de Asua et al. 1973), neurons (Snyder and Kim 1980; Sotelo et al. 1980) and for the general survival and proliferation of brain cells (Obata 1981; Bhat et al. 1983; Huck 1983). Recently, synthesis of insulin by brain cells has been suggested (Rosenzweig et al. 1980; Birch et al. 1984) and specific insulin binding sites have been demonstrated in brain tissue (Havrankova et al. 1978) as well as in cultures of fetal brain cells (Raizada et al. 1980). However, data concerning the effect of insulin on the complex intercellular activities needed for myelinogenesis have not appeared. Although determinations of additional myelin-related parameters; e.g., myelin
348 14
12
10
I
i
j6 4
0
7
14
21
28
35 Days in vitro
Fig. 4. CNPase determination. CNPase activity was quantified in nerve tissue cultures (broken line) and in spinal cords from animals of correspondent age (solid line).
basic protein, might be desirable to certify the increase in myelination, as determined in these experiments, insulin as well as crude pancreatic extract seem to have a marked effect on the time of onset and principally on the total amount of myelin produced in these cultures as compared to those maintained in a medium without insulin (see Fig. 2). Proinsulin had a similar but smaller effect. Although experiments to elucidate the mechanism of insulin action in these cultures were not performed, it seems that insulin had no direct effect on the CNPase activity in vitro nor did it induce a secondary compound, which might alter CNPase activity in normal control tissue (see Table 1). This effect does not seem to be exclusive to INS considering that PEXT that has little insulin was able to induce a similar effect. Conversely, PROINS had a lower or no effect, which could be due principally to variable endogenous production of insulin from proinsulin. The effect of these compounds was evident on primary myelinogenesis as on remyelination although they
349 could not prevent some deterioration of the tissue similar to that observed in the NNM-fed group (see Fig. 3). It is possible that the lower values for CNPase activity in cultures as compared to tissues in vivo might result in part from a lack of some factors such as insulin. Therefore, the addition of these factors might produce a maturation pattern more closely resembling those found in vivo. Similar effects could also explain the lower lipid amounts observed in CNS cultures by Latovitzki and Silberberg (1973) and Satomi and Kishimoto (1981) whose cultures were grown in the absence of insulin in comparison to Giesing and Zilliken (1976) which supplemented their medium with insulin. Current studies are extending these experiments to demonstrate the effect of insulin and related substances on CNS tissue in vivo, particularly for their possible effects on remyelination in experimental allergic encephalomyelitis and its implications for immunologically-mediated human demyelinating diseases of the central and peripheral nervous systems. REFERENCES Bhat, N.R., G. Shanker and R.A. Pieringer (1983) Cell proliferation in growing cultures of dissociated embryonic mouse brain - - Macromolecule and ornithine decarboxylase synthesis and regulation by hormones and drugs, J. Neurosci. Res., 10: 221-230. Birch, N.P., D.L. Christie and A. G. C. Renwick (1984) Immunoreactive insulin from mouse brain cells in culture and whole rat brain, Biochem. J., 218: 19-27. Bornstein, M.B. (1973) Organotypic mammalian central and peripheral nerve tissue. In: P.F. Kruse and M.K. Patterson (Eds.), Tissue Culture - - Methods and Applications, Academic Press, New York, pp. 86-92. Bradbury, K. and C.E. Lumsden (1979) The chemical composition of myelin in organ cultures of rat cerebellum, J. Neurochem., 32: 145-154. Giesing, M. and F. Zilliken (1976) Analysis of hpid components in organotypic cultures of cerebellum during development, Brain Res., 111: 212-219. Havrankova, J., J. Roth and M. Brownstein (1978) Insulin receptors are widely distributed in the central nervous system of the rat, Nature (Lond.), 272: 827-829. Hershman, H.R. (1974) Culture of neural tissue and cells. In: N. Marks and R. Rodnight (Eds.), Research Methods in Neurochemistry, Vol. 2, Plenum Press, New York, pp. 101-160. Huck, S. (1983) Serum-free medium for cultures of the postnatal mouse cerebellum - - Only insulin is essential, Brain Res. Bull., 10: 667-674. Jimenez de Asua, L., E. S. Sudan, M. M. Flawia and H.N. Torres (1973) Effect of insulin on the growth pattern and adenylate cyclase activity of BHK fibroblasts, Proc. Nat. Acad. Sci. (U.S.A.), 70: 1388-1392. Latovitzki, N. and D.H. Silberberg (1973) Quantification of galactolipids in myelinating cultures of rat cerebellum, J. Neurochem., 20: 1771-1776. Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall (1951) Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193: 265-275. Obata, K. (I 981) Requirements for growth of neurites from embryonic chick cerebellum in culture, Develop. Brain Res., 1: 444-449. Pilkis, S.J. and C.R. Park (1974) Mechanism of action of insulin, Ann. Rev. Pharmacol., 14: 365-388. Prohaska, J. R., D.A. Clark and W.W. Wells (1973) Improved rapidity and precision in the determination of brain 2' ,Y-cyclic nucleotide Y-phosphohydrolase, Anal. Biochem., 56: 275-282. Raizada, M.K., J.W. Yang and R.E. Fellows (1980) Binding of [125I]insulin to specific receptors and stimulation of nucleotide incorporation in cells cultures from rat brain, Brain Res., 200: 389-400. Rosenzweig, J. L., J. Havrankova, M.A. Lesniak, M. Bronstein and J. Roth (1980) Insulin is ubiquitous in extrapancreatic tissues of rats and humans, Proc. Nat. Acad. Sci. (U.S.A.), 77: 572-576. Roth, G. A., R. K. Yu and M. B. Bornstein (1983) Chemical analysis of organotypic cultures of mouse spinal cord in normal, demyelinative and nondemyelinative conditions, J. Neurochem., 41: 1710-1717.
350 Roth, G. A. and M. B. Bornstein (1984) Quantitative chemical analysis of demyclination m cultured mouse spinal cord, Develop. Brain Res., 15: 105-111. S atomi, D. and Y. Kishimoto (198 I) Change ofgalactolipids and metabolism of fatty acids in the organotypic culture of myelinating mouse brain, Biochim. Biophys. Aeta, 666: 446-454. Snyder, E.Y. and S.U. Kim (1980) Insulin - - Is it a nerve survival factor? Brain Res., 196: 565-571. Sotelo, J., J.G. Clarence, D. Carleton Gajdusek, B.H. Toh and M. Wurth (1980) Method for preparing cultures of central neurons - - Cytochemical and immunochemical studies, Proc. Nat. Acad. Sei. (U.S.A.), 77: 653-657. Waymouth, C. (1977) Nutritional requirements of cells in culture, with special reference to neural cells. In: S. Fedoroffand L. Hertz (Eds.), Cell, Tissue and Organ Cultures in Neurobiology, Academic Press, New York, pp. 631-648.