Effect of Forskolin and Exogenously Administered Oxytocin mRNA on Oxytocin Release by Dispersed Hypothalamic Cultures

Experimental Neurology 171, 246 –254 (2001) doi:10.1006/exnr.2001.7782, available online at http://www.idealibrary.com on

Effect of Forskolin and Exogenously Administered Oxytocin mRNA on Oxytocin Release by Dispersed Hypothalamic Cultures Zhilin Song and Celia D. Sladek Department of Physiology and Biophysics, Finch University of Health Sciences/The Chicago Medical School, 3333 Green Bay Road, North Chicago, Illinois 60064 Received May 22, 2001; accepted July 17, 2001

Differential vasopressin (VP) gene expression and oxytocin (OT) gene expression were observed in hypothalamic cultures derived from 14-day-old rat fetuses, with VP but not OT being induced by treatment with forskolin and 3-isobutyl-1-methylxanthine. These cultures were used to demonstrate that exogenous VP mRNA could be taken up and translated into releasable VP. In the current studies a similar culture preparation was used to test the hypothesis that, due to the similarity in the mRNA and prohormone structures of VP and OT, the VP-expressing neurons in the cultures would be capable of utilizing exogenous OT mRNA for synthesis of releasable OT. Although OT release was increased by the administration of exogenous OT mRNA, endogenous OT gene expression was also observed. To determine what had induced OT gene expression in the current cultures, the undefined components of the culture preparation, e.g., the glial feeder layer and the serum component of the culture medium, were evaluated. Restraining growth of the glial carpet with cytosine–arabinoside did not alter OT gene expression. Use of a defined medium supplemented with B-27 induced optimal OT gene expression. From this, it is possible to conclude that the components included in B-27 are sufficient for OT gene expression. Factors included in earlier lots of sera may have been responsible for suppression of OT gene expression. Cultures maintained in serum-free, B-27supplemented medium may provide a useful model system for studying OT gene regulation. © 2001 Academic Press Key Words: oxytocin; vasopressin; hypothalamic cultures; cyclic AMP; exogenous mRNA; supraoptic nucleus.

INTRODUCTION

The ability of neurons to take up and translate exogenous mRNA has been demonstrated using models of impaired vasopressin (VP) secretion. Brattleboro rats carry a genetically mutated VP gene and therefore are not capable of synthesizing VP (16, 19). Jirikowski et al. (9) demonstrated that VP mRNA injected into the hypothalamoneurohypophyseal tract of Brattleboro 0014-4886/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

rats was taken up and translated into secretable VP, resulting in correction of diabetes insipidus. Dispersed hypothalamic cultures that do not appear to express VP and oxytocin (OT) have also been used to demonstrate translation of exogenous mRNA for synthesis of VP. In these cultures VP secretion was essentially undetectable under basal conditions, but it could be driven by drugs that elevate intracellular cAMP [e.g., IBMX (3-isobutyl-1-methylxanthine), a phosphodiesterase inhibitor, and forskolin, an activator of adenylate cyclase (together designated as IF)] (15). This effect of cAMP on VP secretion is reversible (18). After treatment for 2–3 weeks with IF, the cultures contained numerous large immunoreactive VP–neurophysin neurons and secreted VP into the culture medium. VP secretion declined to an undetectable level when IF was withdrawn for 1 week but subsequent reexposure of the culture to IF resulted in reexpression of VP synthesis and secretion (18). Thus, this system provided an optimal way to evaluate the ability of neurons that contain all the necessary protein synthetic machinery, but are not actively secreting VP, to take up and translate exogenous VP mRNA. Using this system, VP secretion increased following administration of fulllength sense, but not antisense, VP mRNA to the culture medium (12). In addition, radiolabeled VP mRNA was colocalized in neurophysin (NP)-positive neurons (12). Thus both this study and the Brattleboro rat study demonstrated that, under conditions of low endogenous VP gene expression, exogenous VP mRNA could induce VP release. To evaluate the specificity of VP neurons for the uptake and translation of exogenous VP mRNA, experiments were performed to determine if the VP neurons in the dispersed cultures could utilize a different exogenous mRNA for peptide synthesis and secretion. Since several laboratories have not been successful in demonstrating OT expression in dispersed hypothalamic cultures (15, 18), and since VP and OT are two related hormones with similar genes and prohormone structures, it was hypothesized that VP neurons would be capable of taking up and translating exogenous OT mRNA and secreting OT as

246

OXYTOCIN RELEASE INDUCED BY EXOGENOUS OXYTOCIN mRNA

had previously been shown for exogenous VP mRNA. Therefore, this culture system was used to evaluate the specificity of the mechanisms involved in uptake and translation of mRNA by neurons. These cultures were also used to evaluate the role of the glial feeder layer and of serum supplementation of the culture in inducing or inhibiting OT expression. MATERIALS AND METHODS

1. Dispersed Hypothalamic Culture Primary dispersed hypothalamic cultures were derived from 14-day-old (E14) fetal Sprague–Dawley rats (Zivic Miller). Microdissected hypothalamic tissue was collected in calcium- and magnesium-free buffer aerated with 95% O 2/5% CO 2 and gently triturated following a 10-min incubation in 0.1% trypsin at 37°C. Cells were plated at 0.5 ⫻ 10 6 cells/well in 24-well culture plates (poly-D-lysine treated) in Eagles minimum essential medium (MEM, GIBCO) supplemented with 10% fetal bovine serum, 0.6% glucose, 0.1% L-glutamine, 50 U/ml penicillin, 50 ␮g/ml streptomycin, 2.5 ␮g/ml fungizone, and 0.28 mM bacitracin. Bacitracin was added to prevent degradation of VP in the culture medium. The culture medium was changed at 2- to 3-day intervals. Cytosine-␤ D-arabinofuranoside (Ara-C, 1 ␮M), an agent that inhibits cell division, was added to specific wells in some experiments starting at 5 days in vitro (DIV) to control growth of the glial feeder layer. At the end of the experiment, the culture was fixed with acrolein and positive neurons were identified by immunocytochemistry (ICC). Radioimmunoassays (RIA) were performed to determine the VP and OT contents in the culture medium.

247

feeder layer by treatment with Ara-C either inhibited or induced OT expression, OT release was evaluated with and without Ara-C treatment (1 ␮M starting on DIV 5). c. Evaluation of the effect of serum in the medium on OT expression. Since there could be variability among lots of sera in providing factors capable of stimulating peptide expression, the effect of different serum lots and different serum concentrations were compared: (1) Evaluation of different lots of sera. A dispersed culture was maintained for 21 days in medium supplemented with three different lots of sera: serum 1, Atlanta Biological, Lot 7008C; serum 2, Atlanta Biological, Lot 7013C; and serum 3, GIBCO, Lot 1014546. (2) Effect of filtering serum. A dispersed culture was maintained for 14 days in medium supplemented with serum 1 that had been filtered through a 2-␮m filter or that had not been filtered prior to use. (3) Effect of different concentrations of serum. A dispersed culture was maintained for 14 days in medium supplemented with 2, 5, or 10% of serum 4 (Atlanta Biological, Lot 6003K). (4) Experiment with serum-free medium. A dispersed culture was maintained for 14 days in a serumfree medium formulation consisting of equal volumes of F-12 nutrient mixture and MEM supplemented with 2% 50⫻ B-27 (Sigma). Although the exact formulation of B-27 is proprietary, it is similar to that reported for B-18 (2). It contains all the components known to be required for neuronal survival as originally described for the N2 supplement (1): insulin (5 ␮g/ml), transferrin (100 ␮g/ml), progesterone (6.3 ng/ml), and putrescine (16 ␮g/ml). It also contains corticosterone, retinol, retinyl acetate, triiodothyronine (T3), and antioxidants, as well as other components.

2. Experimental Design a. Effect of exogenous OT mRNA on OT release. Cultures were maintained for 21 days. Control cultures received no further additions except mRNA when indicated. To elevate intracellular cAMP, drug-treated cultures received 200 ␮M IBMX together with 25 ␮M forskolin at every medium change from 2 to 21 DIV. This treatment was necessary to maintain the VP phenotype of the hypothalamic cultures (synthesis and release). OT mRNA was added to the culture on the 21st day of culture (DIV 21). After removing the spent medium, either sense or antisense OT mRNA (7.5 ng/ well) was added in serum-free medium warmed to 37°C containing MEM, 0.6% glucose, 0.1% L-glutamine, 50 U/ml penicillin, 50 ␮g/ml streptomycin, 2.5 ␮g/ml fungizone, 0.28 mM bacitracin, 5 ␮g/ml insulin, and 4 U/ml RNasin. Culture medium was collected after 6 h of incubation and assayed by RIA for VP and OT. b. Evaluation of the role of glia in OT expression. To determine if manipulation of the growth of the glial

3. OT mRNA Synthesis OT mRNA was synthesized from a full-length OT cDNA (obtained from Dr. Thomas G. Sherman). It was linearized by EcoRI and in vitro transcribed with polymerase T7 to obtain sense mRNA. To make antisense mRNA, it was linearized by HindIII and in vitro transcribed with polymerase SP6. The cRNA transcript was labeled by [ 32P]CTP for quantification. 4. Radioimmunoassay VP and OT contents of the medium collected from the cultures were assayed using sensitive RIA. a. Vasopressin assay. The antiserum used in the VP RIA was developed in conjunction with Arnel Laboratories. To achieve a minimum assay sensitivity of 0.1 pg/ml the antiserum was used at a final concentration of 1:150,000. Phosphate buffer (0.1 M, pH 7.6) containing 0.15 M NaCl and 1 mg/ml bovine serum

248

SONG AND SLADEK

albumin and 1 mg/ml sodium azide was used, and standard curves were prepared in triplicate. Nonequilibrium incubation conditions were employed that involved a 24-h preincubation of the standards or samples with antiserum prior to the addition of 2000 cpm I 125-labeled AVP (Arginine vasopressin, New England Nuclear). With this protocol the upper limit of the assay is 10 pg/tube. Antibody-bound VP was separated from free VP using dextran-coated charcoal with 1 mg/ml ␥-globulin added prior to centrifugation to aid in complete separation of bound and free fractions. Culture medium (up to 200 ␮l) was assayed without extraction. Duplicate determinations of two aliquot sizes from each culture sample were assayed. Cross-reactivity with OT was less than 0.01%. b. Oxytocin assay. The antiserum used in the OT RIA was prepared in conjunction with Arnel Laboratories. The antiserum was used at a final concentration of 1:100,000. I 125-labeled OT was used at 1500 cpm/tube with the addition of ␥-globulin prior to centrifugation and preincubation with antibody for 24 h. Phosphate buffer (0.1 M, pH 7.6) containing 0.15 M NaCl and 1 mg/ml bovine serum albumin and 1 mg/ml sodium azide was used in the assay. Standard curves were prepared in triplicate using synthetic OT obtained from Ferring. The lower limit of detectability was 0.1 pg/assay tube and the upper limit was 50 pg/assay tube. VP cross-reactivity in the assay was less than 0.01%. 5. Immunocytochemistry Cultures were fixed in 5% acrolein (10). VP or OT neurons were visualized by ICC with primary antibodies against rat NP (RNP), rat VP–NP (PS-41), rat OT–NP (PS-38), rat VP (VA-4), or rat OT (VA-10) utilizing the Vectastain ABC technique (Vecta Laboratories, Inc., Burlingame, CA). The polyclonal rabbit antirat NP (RNP) antibody was generously supplied by Dr. Alan Robinson (UCLA, CA). The monoclonal mouse anti-rat VP–NP and anti-rat OT–NP and the polyclonal rabbit anti-rat VP (VA-4) and OT (VA-10) antibodies were supplied by Dr. Harold Gainer (NIH). The RNP antibody recognizes both VP–NP and OT–NP while the VP–NP and OT–NP antibodies are specific for the respective NP. In early experiments, other antibodies were tested but did not give positive staining. Briefly, following the acrolein fixation, cultures were treated with 0.01 M sodium metaperiodate and 1% sodium borohydride to remove residue aldehydes. This was followed by incubation in 5% normal donkey serum in 0.1 M phosphate-buffered saline (PBS) containing 0.4% Triton X-100 (Sigma) to block nonspecific staining. Cultures were then incubated with primary antisera (diluted 1:20,000 for RNP; 1:100 for VP–NP and OT–NP; and 1:1000 for VP and OT) for 48 h at 4°C. Cultures were then extensively washed in 0.05 M PBS

(containing 0.25% Triton X-100), incubated for 90 min with biotinylated antibody (donkey anti-rabbit IgG for RNP, OT, and VP; horse anti-mouse IgG for VP–NP and OT–NP), washed, incubated for 90 min with Vectastain ABC reagent, and then washed again. The reaction product formed after 5–7 min with 3,3⬘-diaminobenzidine tetrahydrochloride (DAB, Sigma) activated with 0.03% H 2O 2. Cultures were air-dried overnight and coverslipped with glass coverslip circles (Bellco). ICC stained neurons were counted by systematically scanning each culture well using an inverted light microscope equipped with interference-contrast optics. 6. Data Analysis Results are presented as means ⫾ standard deviations of the means. ANOVA with repeated measure followed by a post hoc t-test was used to determine significant differences. RESULTS

1. VP and OT Release in Response to Forskolin and IBMX Cultures were maintained under basal conditions (control) or with IF treatment from DIV 2 to DIV 21 (IF 21) or from DIV 2 to DIV 14 (IF 14). As shown in Fig. 1A, IF stimulation resulted in increased VP release. The stimulation of hormone release reached a peak by DIV 12 and lasted until the end of the experiment period (DIV 21; F ⫽ 35.486, P ⬍ 0.0001, IF 21 vs control). If IF was withdrawn at DIV 14, VP release gradually declined to baseline 7 days later (IF 14 vs control, F ⫽ 8.330, P ⫽ 0.0095; IF 14 vs IF 21, F ⫽ 20.315, P ⫽ 0.0001). Similarly in Fig. 1B, OT release was stimulated by IF and followed the same response pattern as that of VP. Stimulated OT release reached a peak around DIV 12 and lasted the whole experiment period (F ⫽ 27.055, P ⫽ 0.0001, IF 21 vs control). IF stimulation was essential to maintain sustained OT release. The withdrawal of IF at DIV 14 led to a drop of hormone release to baseline within 7 days (IF 14 vs control, F ⫽ 9.824, P ⫽ 0.0055; IF 14 vs IF 21, F ⫽ 14.594, P ⫽ 0.0007). Thus, OT release provided evidence for OT gene expression in dispersed hypothalamic cultures. This was reinforced by the positive staining of hypothalamic neurons with the OT–NP and OT antibodies. Figure 2 displays some typical neurons from cultures stained with antibodies against rat neurophysin, VP– neurophysin, OT–neurophysin, VP, and OT. 2. VP and OT Release upon Administration of OT mRNA Either sense or antisense OT mRNA was added at DIV 21 to each treatment group. After 6 h of incuba-

OXYTOCIN RELEASE INDUCED BY EXOGENOUS OXYTOCIN mRNA

FIG. 1. VP and OT release in response to IF. (A) VP release. With IF stimulation, VP release significantly increased around Day 7, reached maximum around Day 12 in vitro, and was sustained through Day 21 in the presence of IF (IF 21). If IF was withdrawn on Day 14 (IF 14), VP release declined to baseline 7 days later. #, Difference between IF 21 and control groups (P ⬍ 0.01); *, IF 14 versus control groups (P ⬍ 0.01); @, IF 21 versus IF 14 groups (P ⱕ 0.001). (B) OT release. Sustained stimulation of OT release was induced by IF that exceeded VP release by approximately threefold. Upon withdrawal of IF on Day 14, OT release declined to basal within 1 week. #, Difference between IF 21 and control groups (P ⱕ 0.001); *, IF 14 versus control groups (P ⱕ 0.005); @, IF 21 versus IF 14 groups (P ⱕ 0.001). Values are means ⫾ SEM except where error bars are smaller than symbols. (■) Control, n ⫽ 6; (F) IF 21, n ⫽ 15; (Œ) IF 14, n ⫽ 15.

tion, culture medium was collected for assay of VP and OT. As shown in Fig. 3, the administration of sense OT mRNA led to a significant increase in OT release in both the control and IF 14 groups (control group, F ⫽ 5.735, P ⫽ 0.0118; IF 14 group, F ⫽ 12.090, P ⫽ 0.0013). Administration of antisense OT mRNA had no effect on OT release in these groups, but in the IF 21 group, antisense OT mRNA caused a significant decrease in OT release (F ⫽ 3.928, P ⫽ 0.0487). Sense OT mRNA had no significant effect on OT release in the IF 21 group, and neither sense nor antisense OT mRNA altered VP release in control, IF 14, or IF 21 groups (Fig. 4). 3. Experiments to Evaluate Factors Responsible for OT Gene Expression Effect of Ara-C treatment on VP and OT gene expression. Ara-C treatment prevents cell division and thereby limits the overgrowth of the glial carpet. As shown in Fig. 5, it enhanced both VP and OT release

249

after DIV 12 (VP, F ⫽ 63.011, P ⬍ 0.0001; OT, F ⫽ 22.052, P ⫽ 0.0003). The greater VP and OT release in the Ara-C-treated wells correlates with increased numbers of neurons stained for NP in these wells (Table 1). This may reflect either increased viability of VPand OT-expressing neurons and/or increased numbers of neurons expressing VP and OT. However, Ara-C treatment was not required to induce OT gene expression in these cultures as OT release was detectable in both Ara-C-treated and untreated cultures. Effect of sera on VP and OT gene expression. To determine if the unexpected appearance of OT expression in the cultures was due to characteristics of a specific lot of fetal bovine serum (FBS), cultures were maintained in media supplemented with three different lots of sera. As shown in Fig. 6, IF was effective in stimulating VP and OT release regardless of the lot of serum used. However, there were marked differences in hormone release between the three lots of sera. Peak release was observed at DIV 12 for all sera, but the absolute peak values differed among the sera. There was no significant difference between the ratios of peak VP to OT release in cultures maintained with different sera (Fig. 6C). Although RNP-positive cells were observed with ICC, the numbers of VP- and OT-positive neurons were not determined. Another variable in the serum component of culture medium was the procedure of filtering the serum through a Millipore filter. To determine if this procedure may have removed some substance that regulated OT expression, the effect of this filtering procedure on VP and OT release was evaluated. VP release was increased by IF in cultures maintained in medium supplemented with nonfiltered serum (F ⫽ 2260.093, P ⬍ 0.0001) or filtered serum (F ⫽ 311.551, P ⫽ 0.0004). Similarly, OT release was increased by IF in cultures maintained in either nonfiltered (F ⫽ 19.431, P ⫽ 0.0217) or filtered serum (F ⫽ 36.541, P ⫽ 0.0091). However, there was no difference in release of either VP or OT between these two conditions (data not shown). The number of neurons positively stained with RNP or OTNP further confirmed this (Table 1). IF stimulation of VP and OT release as well as the number of positively stained OT neurons decreased dramatically as the concentration of serum was reduced from 10 to 5% and became undetectable at 2% fetal bovine serum (Figs. 7A and 7B, Table 1). However, in cultures maintained in serum-free medium supplemented with B-27, both VP and OT release were stimulated by IF (VP, F ⫽ 80.628, P ⫽ 0.0009; OT, F ⫽ 29.538, P ⫽ 0.0056; Fig. 8). Cell counts of neurons positively stained with RNP, OTNP, and VPNP confirmed this (Table 1). OT expression was optimal under this condition with the largest number of positively stained OT neurons recorded (Table 1) and peak OT release lagging behind but eventually exceeding VP

250

SONG AND SLADEK

FIG. 2. ICC staining of dispersed hypothalamic neuronal cultures. A and B were stained by antibody against OT–NP (oxytocin neurophysin); C against VP; D against VP–NP (vasopressin neurophysin) E against RNP (rabbit anti-neurophysin); and F against OT. Bars at the right bottom corners represent 50 ␮m.

release. The continued increase in OT release after DIV 12 may reflect slower growth of the glial carpet in the B-27-supplemented medium as a similar pattern was seen in the Ara-C-treated cultures (see Fig. 5). DISCUSSION

The original goal of these studies was to test the hypothesis that VP neurons could utilize exogenous OT mRNA to synthesize secretable OT. Due to the unexpected secretion of OT by the cultures, the experiments did not address this hypothesis. However, they did demonstrate that both exogenous OT mRNA and acti-

vation of adenylate cyclase with forskolin increase OT release from dispersed hypothalamic cultures. Since the current cultures contained both active VP and OT neurons, the synthesis of OT from exogenous OT mRNA in this system may be due to translation of OT mRNA by either OT neurons only or by both VP and OT neurons. The structures of the mRNA for the two peptides are similar, and the prohormones require similar processing enzymes (4, 17). Therefore it is possible that VP neurons can take up OT mRNA and translate it into hormone, or vice versa. In vivo there is a subpopulation of magnocellular neurons that contains both VP and OT peptide as observed by immunocyto-

OXYTOCIN RELEASE INDUCED BY EXOGENOUS OXYTOCIN mRNA

251

FIG. 3. Effect of sense and antisense OT mRNA on OT release. Exposure to sense OT mRNA for 6 h induced a significant increase in OT release in both control and IF-treated neurons following IF withdrawal for 7 days (indicated by *(F ⫽ 5.735, P ⫽ 0.0118) and #(F ⫽ 12.090, P ⫽ 0.0013), respectively). In those neurons treated with IF continuously for 21 days, antisense OT mRNA caused a significant decrease in OT release (indicated by @, F ⫽ 3.928, P ⫽ 0.0487).

chemistry. Furthermore, recent single-cell RT–PCR studies revealed that the majority of magnocellular neurons express both VP and OT genes, although most cells predominantly express either VP or OT (3, 5, 22). Thus, although classically neurons have been characterized as either VP or OT neurons, because the minor

FIG. 4. Effect of sense and antisense OT mRNA on VP release. Administration of exogenous OT mRNA did not affect VP release in any group. Control, wells not treated with IF; IF 14, wells treated with IF DIV 2–14; IF 21, wells treated with IF DIV 2–21. Sense or antisense OT mRNA was added on Day 21 for 6 h. The lower panel is an enlarged scale of the control and IF 14 groups. n ⫽ 7 for each control subgroup; n ⫽ 5 for each IF 14 subgroup; n ⫽ 5 for each IF 21 subgroup.

FIG. 5. The effect of Ara-C treatment on VP and OT release. (A) VP release was stimulated in both IF-treated or IF ⫹ Ara-C-treated cultures (IF vs its control group, F ⫽ 819.677, P ⬍ 0.0001; IF ⫹ Ara-C vs its control group, F ⫽ 373.796, P ⬍ 0.0001). Ara-C treatment did not affect VP release initially. After DIV 12, VP release in IF-stimulated, Ara-C-treated wells was enhanced in comparison with that of wells without Ara-C (F ⫽ 63.011, P ⬍ 0.0001). (B) OT release was stimulated in both IF-treated or IF ⫹ Ara-C-treated cultures (IF vs its control group, F ⫽ 27.328, P ⫽ 0.0003; IF ⫹ Ara-C vs its control group, F ⫽ 48.026, P ⬍ 0.0001). Initially, there is no difference in OT release between Ara-C-treated or nontreated wells. At Day 14, IF-stimulated OT release decreased in wells without Ara-C treatment, whereas OT release continued to increase in wells treated with Ara-C (F ⫽ 22.052, P ⫽ 0.0003). P ⬍ 0.05, simple main effects: *, IF group vs its control; #, IF ⫹ Ara-C group vs its control; @, IF vs IF ⫹ Ara-C group. Values are means ⫾ SEM except where error bars are smaller than symbols. (䊐) Control, n ⫽ 5; (■) IF, n ⫽ 8; (E) Con ⫹ Ara-C, n ⫽ 4; (F) IF ⫹ Ara-C, n ⫽ 8.

peptide is expressed in amounts that are undetectable by ICC or in situ hybridization studies, all magnocellular neurons appear to be producing mRNAs for both VP and OT. Although the promoter region of the bovine OT gene includes classic cAMP response elements, such as CRE (cAMP response element) and AP-2 (activator protein) sites, the functional importance of these sites has not been evaluated (21). In contrast, regulation of VP gene expression by similar elements has been the focus of several studies (14, 20). In the current studies, both VP and OT gene expression was stimulated by cAMP. IF increased both OT release and the number of neurons expressing OT–NP in a manner similar to that of IF’s effect on VP expression and release. Without IF stimulation, the OT gene expression was detectable by ICC only in occasional cells and OT release was undetectable in five out of seven culture preparations.

252

SONG AND SLADEK

TABLE 1 Cell Counts Following Immunocytochemistry for RNP, OTNP, or VPNP Experimental variable 1. ARA-C treatment No treatment ⫹ ARA 2. Serum filtration Not filtered Filtered 3. Serum concentration 10% 5% 2% 4. B27 supplement

RNP

VPNP

OTNP

299 ⫾ 31 (2) 386 ⫾ 3 (2)

77 ⫾ 1 (2) 108 ⫾ 21 (2)

180 ⫾ 21 (2) 310 ⫾ 15 (2)

156 (1) 203 (1)

— —

73 ⫾ 21 (2) 77 ⫾ 6 (2)

— — — —

— — — 104 ⫾ 34 (2)

161 ⫾ 8 (2) 56 ⫾ 1 (2) 2 ⫾ 1 (2) 576 ⫾ 34 (2)

Note. Results are expressed as means ⫾ SD of multiple wells (n) in single culture preparations. All wells were treated with IF. When the number of wells available in a single culture preparation was inadequate for staining with all three antibodies, we focused on evaluation of OT expression.

Forskolin stimulation of OT release is consistent with the previous report in cultures from E17–19 rat fetuses (7), however, it is inconsistent with the observation from several laboratories that hypothalamic cultures derived from E14 rat fetuses did not express OT. These previous findings of suppression of the OT gene in dispersed culture derived from E14 rat fetuses may reflect lack of a stimulating factor or existence of an inhibiting factor. Thus, the OT gene expression observed in the current studies with cultures derived from E14 rat fetuses may be due either to the unintentional removal of inhibitory factors or to the inclusion of stimulatory factors in the culture system. Several factors potentially responsible for providing this environment were investigated in the current studies. The first candidate considered was Ara-C, an agent that inhibits cell division and thereby limits glial carpet growth. Neurons are not affected since they are not actively dividing in these cultures. A glial carpet provides an ideal substrate for neuronal attachment and also produces growth factors that provide trophic support for neuronal growth. In vivo, there are active interactions between glial cells and neurons, both morphologically and chemically. In the hypothalamoneurohypophysial system, morphologic retraction of glial cells during stimulation of hormone release allows formation of multiple synapses and synchronization of firing pattern of OT neurons (6). In the posterior pituitary, taurine released from glial cells may actively inhibit hormone release (6, 13). However, in dispersed primary cultures, overgrowth of glial cells will reduce neuronal viability by literally growing over them and reducing access to nutrients and gas exchange. Thus, due to the importance of the balance between glia and neurons, application of Ara-C in this culture system may have had an impact on OT gene expression. This possibility was suggested by the fact that Ara-C treatment was omitted the first time that OT expression

was detected. However, in an experiment specifically designed to evaluate this possibility, Ara-C treatment did alter OT gene expression. Its effect on OT expression was similar to its effect on VP expression, with enhancement of both OT and VP release and an increase in the number of OT-expressing neurons. Thus, Ara-C treatment was not responsible for the suppression of OT gene expression in the previous cultures. Another undefined component in this culture system is the FBS component of the medium. FBS provides several components that are essential for culture viability and growth, but since it is undefined and may vary from batch to batch, removal of inhibitory factors or inclusion of stimulating factors in certain lots of sera may result in an environment that allows for cAMP regulation of OT gene expression. Indeed, it has been reported that exposure to sera may be responsible for permanent suppression of OT gene expression in bovine granulosa cell cultures (8, 11). However, all three lots of sera tested allowed both OT and VP expression. Although, none of these lots were unique in providing the regulatory environment for OT gene expression, the possibility still exists that earlier lots of sera may have included factors that suppressed OT expression. Filtering the sera did not remove factors required for OT gene expression, but both VP expression and OT expression were dependent upon factors supplied by the serum since expression of both peptides was suppressed if the concentration of supplemented sera was reduced. However, both VP expression and OT expression were maintained when serum was replaced with B-27, a defined supplement. The B-27 supplement provided an optimal condition for OT expression as indicated by the large number of OTNP-positive cells present and the large OT release. Although the factors responsible for OT gene expression were not revealed by the current manipulations, since B-27 is well defined and provided an optimal

OXYTOCIN RELEASE INDUCED BY EXOGENOUS OXYTOCIN mRNA

253

FIG. 7. Effect on VP and OT release of reducing the concentration of serum in the culture medium. (A) VP release was stimulated by IF in both 10 and 5% FBS-supplemented cultures, but not in 2% FBS-supplemented cultures. The peak response to IF in 5% FBS-supplemented culture was significantly smaller than that of 10% FBS-supplemented culture (F ⫽ 841.095, P ⫽ 0.0001). * and # P ⬍ 0.05 (simple main effects) for IF-treated (closed symbols) versus control (open symbols) in 10 and 5% serum. (B) OT release was only significantly stimulated by IF in 10% FBS-supplemented cultures (F ⫽ 18.588, P ⫽ 0.0182, *IF-treated (closed symbols) versus control (open symbols)). In cultures supplemented with 5% FBS, OT release had a tendency to increase, but did not reach a statistically significant difference. Values are means ⫾ SEM except where error bars are smaller than symbols. FIG. 6. Effect of three different lots of sera on VP and OT release. (A) VP release. Although all three lots of sera promoted VP expression in the presence of IF (closed symbols), serum 1 (squares) gave the highest peak VP release (on Day 12). *, Significant difference in VP release between IF-treated and nontreated neurons using serum 1 (F ⫽ 423.181, P ⫽ 0.0003); #, serum 2 (circles, F ⫽ 11.223, P ⫽ 0.0441); @, serum 3 (diamonds, F ⫽ 3.485, P ⫽ 0.0083). Open symbols indicate release without IF. N ⫽ 3 and 2 wells with and without IF treatment, respectively. (B) OT release. All of the sera promoted OT release under IF stimulation. Serum 1 gave the highest peak of OT release (on Day 12). *, Significant difference in OT release between IF-treated and nontreated neurons using serum 1; #, serum 2; @, serum 3 (P ⬍ 0.05, simple main effects). Symbols and N are the same as those in A. (C) Comparison of peak VP and OT release. The ratios between peak VP and OT release from cultures maintained in serums 1, 2, or 3 were not significantly different. The ratio of VP to OT release was evaluated on Day 12 in vitro.

environment for OT gene regulation, it is evident that B-27 includes the components essential for OT gene expression. Thus cultures supplemented with B-27

FIG. 8. VP and OT release was stimulated by IF in serum-free cultures supplemented with B-27. *P ⬍ 0.05 versus respective control (simple main effects). (䊐) Control–VP, n ⫽ 1; (■) IF–VP (n ⫽ 5); (E) Control–OT (n ⫽ 1); (F) IF–OT, n ⫽ 4.

254

SONG AND SLADEK

may be a useful model system to study OT gene regulation. In conclusion, these studies indicate that like VP mRNA, exogenous OT mRNA can be taken up and translated into releasable hormone. Furthermore, OT gene expression can be reliably induced by IF treatment of dispersed hypothalamic cultures derived from E14 rat fetuses.

10.

11.

12.

ACKNOWLEDGMENTS This work was supported by a grant from the NIH to C.D.S. (RO1-NS27975). The technical assistance of H. E. Sidorowicz is gratefully acknowledged.

13.

REFERENCES 14. 1. 2.

3.

4.

5.

6.

7.

8.

9.

Bottenstein, J. E., and A. L. Harvey. 1985. Cell Culture in the Neurosciences. Plenum, New York. Brewer, G. J., and C. W. Cotman. 1989. Survival and growth of hippocampal neurons in defined medium at low denisty: Advantages of a sandwich culture technique or low oxygen. Brain Res. 494: 65–74. Gainer, H. 1998. Cell-specific gene expression in oxytocin and vasopressin magnocellular neurons. Adv. Exp. Med. Biol. 449: 15–27. Gainer, H., and S. Wray. 1994. Cellular and molecular biology of oxytocin and vasopressin. In The Physiology of Reproduction (E. Knobil and J. D. Neill, Eds.), pp. 1099 –1129. Raven Press, New York. Glasgow, E., K. Kusano, H. Chin, E. Mezey, W. S. Young, and H. Gainer. 1999. Single cell reverse transcription–polymerase chain reaction analysis of rat supraoptic magnocellular neurons: Neuropeptide phenotypes and high voltage-gated calcium channel subtypes. Endocrinology 140: 5391–5401. Hatton, G. I. 1999. Astroglial modulation of neurotransmitter/ peptide release from the neurohypophysis: Present status. J. Chem. Neuroanat. 16: 203–222. Hu, S.-B., L. A. Tannahill, S. Biswas, and S. L. Lightman. 1991. Release of corticotrophin-releasing factor-41, arginine vasopressin and oxytocin from rat hypothalamic cells in cultures: Responses to activation of intracellular second messengers and to corticosteroids. J. Endocrinol. 132: 57– 65. Ivell, R., N. Walther, U. Wehrenberg, C. McArdle, and H. Ungefroren. 1993. The regulation of neurohypophyseal peptide gene expression in gonadal tissues. Regul. Pept. 45: 263–267. Jirikowski, G. F., P. P. Sanna, D. Maciejewski-Lenoir, and F. E. Bloom. 1992. Reversal of diabetes insipidus in Brattleboro rats:

15.

16.

17.

18.

19.

20.

21.

22.

Intrahypothalamic injection of vasopressin mRNA. Science 255: 996 –998. King, J. C., R. M. Lechan, G. Kugel, and E. L. P. Anthony. 1983. Acrolein: A fixative for immunocytochemical localization of peptides in the central nervous system. J. Histochem. Cytochem. 31: 62– 68. Luck, M. R. 1989. Greatly elevated and sustained secretion of oxytocin by bovine granulosa cells in serum-free culture. J. Exp. Zool. 251: 361–366. Mathiasen, J. R., D. Maciejewski-Lenoir, F. E. Bloom, and C. D. Sladek. 1996. Increased vasopressin secretion from hypothalamic cultures following administration of exogenous vasopressin mRNA. Exp. Neurol. 141: 165–172. Miyata, S., O. Matsushima, and G. I. Hatton. 1997. Taurine in rat posterior pituitary: Localization in astrocytes and selective release by hypoosmotic stimulation. J. Comp. Neurol. 381: 513– 523. Mohr, E., W. Meyerhof, and D. Richter. 1995. Vasopressin and oxytocin: Molecular biology and evolution of the peptide hormones and their receptors. Vitam. Horm. 51: 235–266. Oeding, P., K. Schilling, and H. Schmale. 1990. Vasopressin expression in cultured neurons is stimulated by cyclic AMP. J. Neuroendocrinol. 2: 859 – 865. Schmale, H., and D. Richter. 1984. Single base deletion in the vasopressin gene is the cause of diabetes insipidus in Brattleboro rats. Nature 30: 705–709. Sladek, C. D. 2000. Antidiuretic hormone: Synthesis and release. In Handbook of Physiology (J. C. S. Fray, Ed.), Section 7, Vol. III, pp. 436 – 495. Oxford Univ. Press, Oxford. Sladek, C. D., and M. J. Gallagher. 1993. The stimulation of vasopressin gene expression in cultured hypothalamic neurons by cyclic adenosine 3⬘,5⬘-monophosphate is reversible. Endocrinology 133: 1320 –1330. Valtin, H., and H. A. Schroeder. 1964. Familial hypothalamic diabetes insipidus in rats (Brattleboro strain). Am. J. Physiol. 206: 425– 430. Verbeeck, M. A. E., R. A. H. Adan, and J. P. H. Burbach. 1990. Vasopressin gene expression is stimulated by cyclic AMP in homologous and heterologous expression systems. FEBS Lett. 272: 89 –93. Walther, N., U. Wehrenberg, B. Brackmann, and R. Ivell. 1991. Mapping of the bovine oxytocin gene control region: Identification of binding sites for luteal nuclear proteins in the 5⬘ noncoding region of the gene. J. Neuroendocrinol. 3: 539 –549. Xi, D., K. Kusano, and H. Gainer. 1999. Quantitative analysis of oxytocin and vasopressin messenger ribonucleic acids in single magnocellular neurons isolated from supraoptic nucleus of rat hypothalamus. Endocrinology 140: 4677– 4682.