Interleukin-15, a T-cell growth factor, is expressed in human neural cell lines and tissues

Journal of Neurological Sciences 155 (1998) 170–177

Interleukin-15, a T-cell growth factor, is expressed in human neural cell lines and tissues Jun-ichi Satoh*, Kazuhiro Kurohara, Motohiro Yukitake, Yasuo Kuroda Division of Neurology, Department of Internal Medicine, Saga Medical School, 5 -1 -1 Nabeshima, Saga 849, Japan Received 17 April 1997; received in revised form 19 August 1997; accepted 27 August 1997

Abstract Interleukin-15 (IL-15) is a novel cytokine which shares activities and receptor components with IL-2. To investigate the biological roles of IL-15 in the human nervous system, we examined the expression of mRNAs for IL-15 and the IL-15 receptor three subunits (IL-15Ra, IL-2Rb and IL-2Rg) in human neural cell lines and tissues using reverse transcription–polymerase chain reaction and Southern blot analysis. The constitutive expression of high levels of IL-15 mRNA was observed in all the cell lines examined, including Y79 retinoblastoma, IMR-32 neuroblastoma, SK-N-SH neuroblastoma, U-373MG glioma, KG-1-C glioma, NTera2 teratocarcinoma and neurons derived from NTera2 cells following treatment with retinoic acid (RA). Among these cell lines, IL-15 protein was detectable at high levels in culture supernatants of SK-N-SH cells and NTera2-derived neurons. The expression of an alternatively-spliced transcript of the IL-15 gene was up-regulated in NTera2 cells during RA-induced neuronal differentiation, suggesting the existence of differentiationdependent transcriptional regulation. The expression of IL-15 mRNA was also identifed in the human cerebral and cerebellar tissues, peripheral nerve and skeletal muscle, while the mRNAs for the complete set of IL-15R components were detectable only in U-373MG cells, cerebral and cerebellar tissues at significant levels. These results indicate that the expression of IL-15 but not of IL-15R mRNA is universal in human neural cell lines and tissues and raise the possibility that IL-15 acts as a neuroimmune regulatory factor in the human central nervous system.  1998 Elsevier Science B.V. Keywords: Human nervous system; IL-15; IL-15 receptor; Neuroimmune interaction; RT–PCR

1. Introduction Interleukin-15 (IL-15) is a 14- to 15-kDa novel cytokine that shares biological activities and receptor components with IL-2 (for reviews, Giri et al., 1995a; Tagaya et al., 1996). IL-15 was originally identified in culture supernatants of a simian kidney epithelial cell line as an activity to support proliferation of the IL-2-dependent murine T cell line (Grabstein et al., 1994). The IL-15 receptor (IL-15R) is composed of a heterotrimeric complex consisting of three subunits: the IL-15-specific nonsignal transducing a chain and the b and g chains of the IL-2 receptor (IL-2R), both of which are essential for signal transduction via the JAK / STAT pathways (Anderson et al., 1995; Giri *Corresponding author. Tel.: 181 952 316511; fax: 181 952 342017. 0022-510X / 98 / $19.00  1998 Elsevier Science B.V. All rights reserved. PII S0022-510X( 97 )00310-9

et al., 1995b; Johnston et al., 1995). IL-15 stimulates proliferation of activated CD4 1 T cells, CD8 1 T cells, gd T cells and natural killler (NK) cells and promotes induction of cytotoxic T cells and lymphokine activated killer (LAK) cells (Carson et al., 1994; Edelbaum et al., 1995; Kanegane and Tosato, 1996; Leclercq et al., 1996; ´ Mrozek et al., 1996; Nishimura et al., 1996; Warren et al., 1996). IL-15 induces cytokine production by both CD4 1 T cells and NK cells and acts as a potent chemoattractant for recruitment and migration of T cells (Carson et al., 1995; Wilkinson and Liew, 1995; McInnes et al., 1996; Mori et al., 1996; Nieto et al., 1996). Despite lacking any sequence homology with IL-2, IL15 is a member of the four-helix bundle cytokine family with a close structural relationship to IL-2, although the expression patterns of both cytokines are found to be quite

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

different. The primary source of IL-2 is activated T cells in which the expression of IL-15 mRNA is undetectable (Grabstein et al., 1994), while the IL-15 message is expressed preferentially in the nonlymphoid tissues (Grabstein et al., 1994; Giri et al., 1995a; Tagaya et al., 1996). With regard to their receptors, the expression of the IL-2Ra subunit is identified exclusively in lymphoid cells such as T cells, B cells and monocytes / macrophages, while IL-15Ra mRNA expression exhibits a much broader cellular distribution, including the nonlymphoid tissues (Anderson et al., 1995; Giri et al., 1995b). The IL-15Ra subunit with a structural similarity to IL-2Ra does not by itself participate in IL-15 signal transduction, but it shows more than 1000-fold higher affinity of binding to IL-15 than that of IL-2Ra binding to IL-2, regardless of the presence of the other IL-15 receptor components (Anderson et al., 1995; Giri et al., 1995b). By contrast, the IL-2Ra subunit requires association with the IL-2Rbg heterodimeric complex for acquisition of the high affinity IL-2 binding ability. Remarkable differences between the IL-15 / IL-15R and IL-2 / IL-2R systems suggest that IL-15 may exert unique biological activities in the nonlymphoid system in addition to IL-2-like activities in the immune system. The expression of IL-15 and IL-15R in the human central and peripheral nervous systems (CNS and PNS) remains unknown. There exist only two previous reports which showed that the messages for neither IL-15 nor IL-15R were detectable in the human and mouse brain tissues (Grabstein et al., 1994; Giri et al., 1995b). However, in disagreement with these observations, we have recently demonstrated that cultured fetal human astrocytes and microglia express IL-15 at both mRNA and protein levels (Lee et al., 1996). In the present study, for the initial step to investigate biological roles of IL-15 in the human nervous system, we examined the expression of IL-15 / IL15R mRNAs in a panel of well-characterized human neural cell lines.

2. Materials and methods

2.1. Human cell lines and tissues of neural and nonneural origins All human cell lines and primary cultures were maintained in Dulbecco’s modified Eagle’s medium (DMEM) or RPMI medium 1640 supplemented with 5% or 10% fetal bovine serum (FBS), 100 U / ml penicillin and 100 mg / ml streptomycin (feeding medium). A panel of human neural cell lines were selected in view of their wellcharacterized phenotypes. The nonneural cell lines of human origin were also utilized for controls. The human T-cell lymphoma cell line HUT102 was provided by Dr Tokunaga, Department of Pathology, Saga Medical School (Saga, Japan). The cell lines of Y79 retinoblastoma, IMR-

171

32 neuroblastoma, SK-N-SH neuroblastoma, KG-1-C glioma, MOLT-4 T-cell leukemia, HeLa cervical carcinoma, HepG2 hepatoblastoma and A549 lung carcinoma were provided by the RIKEN Cell Bank (Tsukuba, Japan) and the Japanese Cancer Research Resources Bank (Tokyo, Japan). U-373MG glioma, NTera2 teratocarcinoma, K-562 erythroleukemia and HL-60 promyelocytic leukemia were obtained from the American Type Culture Collection (Rockville, MD). The peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood by centrifugation on a Ficoll–Hypaque gradient. The primary culture of fibroblasts was prepared from a skin explant of a patient with the presumptive diagnosis of Fabry disease. For induction of neuronal differentiation, NTera2 cells maintained in the undifferentiated state (NTera2-undif) were incubated for 4 weeks in feeding medium containing 10 25 M all trans retinoic acid (RA; Sigma, St. Louis, MO), replated twice and then plated on the surface coated with Matrigel Basement Membrane Matrix (Becton Dickinson, Bedford, MA). They were incubated further for 2 weeks in feeding medium containing a cocktail of mitotic inhibitors to establish the differentiated neuron-enriched cultures (NTera2-N) as described previously (Pleasure et al., 1992; Satoh et al., 1997). The expression of IL-15 mRNA was also examined in several human tissues. They include the biopsy-derived human skeletal muscle of a patient with ocular myasthenia gravis, the autopsy-derived human peripheral nerve of the cauda equina, cerebral and cerebellar tissues isolated from unaffected regions of a 76 year-old patient who died of cerebral infarction and the surgically-resected tissues of neurinoma and thymoma.

2.2. Reverse transcription–polymerase chain reaction ( RT–PCR) analysis Reverse transcription–polymerase chain reaction (RT– PCR) analysis was performed according to the methods described previously (Lee et al., 1996; Satoh et al., 1995, 1997). In brief, total RNA was extracted from the cells and tissues described above using TRIZOL reagent (GIBCOBRL) by the acid quanidinium thiocyanate–phenol–chloroform method. The concentration and purity of isolated RNA were assessed by reading UV absorbance at 260 and 280 nm on a Beckman UV spectrophotometer. Five micrograms of RNA from each sample were subjected to DNase treatment and then processed for cDNA synthesis using oligo(dT) 12 – 18 primers and SuperScript II reverse transcriptase (GIBCO-BRL, Gaithersburg, MD). Fifty nanograms of cDNA was amplified by PCR in a programmable thermal cycler (Takara, Tokyo, Japan) using specific sense and antisense primers for IL-15, IL-15 receptor subunits (IL-15Ra, IL-2Rb and IL-2Rg) and glyceraldehyde-3-phosphate dehydrogenase (G3PDH) as listed in Table 1. They were synthesized by a

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

172

Table 1 Primer and internal oligonucleotide sequences used for PCR amplification and Southern blot analysis Primers and internal oligonucleotides

Sequence

Product size (b.p.)

References

IL-15 IL-15 IL-15 IL-15

59AAACCCCTTGCCATAGCCAGCTCTT39 59CTTCTGTTTTAGGAAGCCCTGCACT39 59CAGCCCAAAATGAAGACATGAATGCC39 59GAGTAAAACTGAACTGAGAGTCAG39

202 / 321

Grabstein et al., 1994

IL-15Ra sense IL-15Ra antisense IL-15Ra internal

59TTGAACAAGGCCACGAATGTCGCC39 59ATGGCTTCCATTTCAACGCTGGCC39 59GCTTATCTCTGTGGTTCCTGTGGA39

404 / 503 / 623

Anderson et al., 1995

IL-2Rb sense IL-2Rb antisense IL-2Rb internal

59ACTTCCCAGTTCACATGCTTCTACAACTCG39 59GAGATTTCCCAGCTTATGTTGCATCTGTGG39 59GTGACGATGTCAACTGTGGTCAGT39

374

Hatakeyama et al., 1989

IL-2Rg sense IL-2Rg antisense IL-2Rg internal

59AGAGCAAGCGCCATGTTGAAGCCA39 59CTCCGAACACGAAACGTGTAGCGT39 59CAGAGCTGCTGTTCCAAGTGCAAT39

686

Takeshita et al., 1992

G3PDH sense G3PDH antisense

59CCATGTTCGTCATGGGTGTGAACCA39 59GCCAGTAGAGGCAGGGATGATGTTC39

251

Ercolani et al., 1988

sense antisense internal-1 internal-2

Meazza et al., 1996

Two different oligonucleotide probes designated as IL-15 internal-1 and internal-2 were utilized to identify PCR products for the human IL-15 gene. The IL-15 internal-1 oligonucleotide probe was designed to hybridize with the sequence which exists outside exon A (Meazza et al., 1996), while the IL-15 internal-2 oligonucleotide probe was designed to hybridize specifically with the sequence within exon A. The primers for the PCR of human IL-15Ra gene were designed to amplify three differentially spliced isoforms of this gene with their expected sizes of 623 bp, 503 bp and 403 bp, all of which were detectable by the IL-15Ra internal oligonucleotide probe listed above.

commercial supplier (Sawady Technology, Tokyo, Japan). PCR was carried out in 50 ml of reaction mixture containing Taq DNA polymerase buffer (20 mM Tris–HCl, pH 8.4, 50 mM KCl, 200 mM dNTP, 0.8 to 1.5 mM MgCl 2 , 0.5 mM each primer) and 1.25 U Taq DNA polymerase (GIBCO-BRL). The amplification program consisted of a denaturing step at 948C for 1 min, an annealing step at 608C for 40 s and an extension step at 72.98C for 50 s for 40 cycles for IL-15 and IL-15 receptor or for 35 cycles for G3PDH, except for the initial amplification in which cDNA was denatured at 948C for 3 min, annealed at 608C for 1 min and extended at 72.98C for 3 min. In each sample, total RNA without reverse transcription was amplified in paralell in order to confirm the absence of genomic DNA. The G3PDH gene was used as a reaction standard (Ercolani et al., 1988).

2.3. Southern blot analysis Southern blot analysis was performed to confirm the specificity of PCR products. The PCR products were separated on a 1.5% agarose gel, stained with ethidium bromide, transferred onto a nylon membrane and immobilized by baking at 808C for 60 min in an oven. The membranes were then hybridized with specific internal oligonucleotide probes (Table 1) labeled with digoxigenin (DIG)-11-dUTP by the DIG oligonucleotide tailing kit (Boehringer Mannheim, Mannheim, Germany). The specific reaction was visualized by the DIG chemiluminescence detection kit (Boehringer Mannheim).

2.4. Enzyme-linked immunosorbent assay ( ELISA) The IL-15 protein in undiluted culture supernatants was assessed using the human IL-15 ELISA kit (Genzyme, Cambridge, MA) in which recombinant human IL-15 was utilized as a standard according to the manufacturer’s instruction. The lower limit of detection of IL-15 protein was less than 10 pg / ml in this assay.

3. Results

3.1. The constitutive expression of IL-15 mRNA in human cells and tissues of neural and nonneural origins The constitutive expression of high levels of IL-15 mRNA was identified by RT–PCR and subsequent Southern blot analysis with the expected product size of 202 bp in a panel of human neural cell lines such as Y79 retinoblastoma cells, IMR-32 neuroblastoma cells, SK-NSH neuroblastoma cells, U-373MG glioma cells, KG-1-C glioma cells, NTera2 teratocarcinoma cells in the undifferentiated state (NTera2-undif) and peripheral blood mononuclear cells (PBMC) (Fig. 1A and B, lanes 1–6 and 9). The culture of NTera2-derived differentiated neurons (NT2-N) expressed two distinct amplification products: a weak band with the size of 202 bp and an intense band with the 119 bp-longer size (321 bp) (Fig. 1A and B, lane 7). The latter was identified by the oligonucleotide probe specific for the sequence of a recently reported alternative exon termed exon A (Meazza et al., 1996; Onu et al.,

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

173

The expression of substantial levels of IL-15 mRNA was also identified in human tissues, including the skeletal muscle, peripheral nerve, cerebrum, cerebellum, neurinoma and thymoma and as well as in cultured skin fibroblasts (Fig. 2A and B, lanes 1–7). An extremely weak band corresponding to the exon A-containing transcript was detectable in the skeletal muscle (Fig. 2C, lane 1). The levels of G3PDH mRNA expression were almost constant among the tissues examined (Fig. 2D, lanes 1–7). No products were amplified in total RNA isolated from each sample processed for PCR without reverse transcription (data not shown). The constitutive expression of high levels of IL-15 mRNA was also observed in human nonneural cell lines of either hematopoietic or nonhematopoietic origins such as HUT102 T-cell lymphoma cells, MOLT4 T-cell leukemia

Fig. 1. IL-15 mRNA expression in human neural cell lines. Fifty nanograms of cDNA was amplified by PCR using the primers specific for the human IL-15 gene or the human G3PDH gene which are listed in Table 1. The amplified products were separated on a 1.5% agarose gel, transferred onto a nylon membrane and hybridized with specific internal oligonucleotide probes. Lanes (1 to 9) represent as follows: (1) Y79 retinoblastoma cells, (2) IMR-32 neuroblastoma cells, (3) SK-N-SH neuroblastoma cells, (4) U-373MG glioma cells, (5) KG-1-C glioma cells, (6) NTera2 teratocarcinoma cells in the undifferentiated state, (7) NTera2-derived differentiated neurons induced by treatment with retinoic acid, (8) peripheral blood mononuclear cells without inclusion of the reverse transcriptase in the procedure of cDNA synthesis and (9) peripheral blood mononuclear cells with reverse transcription. The DNA size marker is shown on the left. Panels A and D: ethidium bromide staining of the gels; (A) IL-15, (D) G3PDH. Panels B and C: Southern blot of the gel corresponding to the panel A using the oligonucelotide probe of IL-15 internal-1 (B) which can detect both of two alternatively spliced IL-15 transcripts with exon A (321 bp) (Meazza et al., 1996) or without exon A (202 bp) or using the IL-15 internal-2 probe (C) which is specific for the exon A-containing transcript (321 bp) as listed in Table 1.

1997) which exists within the known IL-15 mRNA coding sequence as an additional exon (Fig. 1C, lane 7). A very low level of the exon A-containing transcript was also expressed in Y79 cells (Fig. 1C, lane 1). The levels of a housekeeping gene G3PDH mRNA expression were almost constant among the cell lines examined (Fig. 1D, lanes 1–7 and 9). No products were amplified in total RNA isolated from each sample when processed for PCR without reverse transcription, confirming that a contamination of genomic DNA was excluded (Fig. 1A–D, lane 8 and other data not shown).

Fig. 2. IL-15 mRNA expression in human neural tissues. Fifty nanograms of cDNA was amplified by PCR using the primers specific for the human IL-15 gene or the human G3PDH gene which are listed in Table 1. The amplified products were separated on a 1.5% agarose gel, transferred onto a nylon membrane and hybridized with specific internal oligonucleotide probes listed in Table 1. Lanes (1 to 7) represent as follows: (1) skeletal muscle, (2) peripheral nerve, (3) cerebrum, (4) cerebellum, (5) neurinoma, (6) thymoma and (7) cultured-skin fibroblasts. The DNA size marker is shown on the left. Panels A and D: ethidium bromide staining of the gels; (A) IL-15, (D) G3PDH. Panels B and C: Southern blot of the gel corresponding to the panel A using the oligonucelotide probe of IL-15 internal-1 (B) or IL-15 internal-2 (C) which are listed in Table 1.

174

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

cells, K-562 erythroleukemia cells, HL-60 promyelocytic leukemia cells, HeLa cervical carcinoma cells, HepG2 hepatoblastoma cells and A549 lung carcinoma cells (data not shown). The exon A-containing transcript was detectable in HL-60 promyelocytic leukemia cells at a very low level.

403 bp could be identified (Fig. 4A and B, lane 4). They correspond to three differentially spliced isoforms of the human IL-15Ra gene that were reported previously (Anderson et al., 1995). Lower levels of IL-15Ra mRNA were detectable in KG-1-C, NTera2-undif cells and PBMC in which the expression of both 503 bp- and 403 bp-tran-

3.2. The detection of IL-15 protein in human cell lines of neural and nonneural origins Varying amounts of IL-15 protein were detectable by ELISA in culture supernatants of Y79, SK-N-SH, U373MG, NTera2-N, HUT102, MOLT-4, HepG2 and A549 cells (Fig. 3). Among the human neural cell lines, SK-NSH cells and NTera2-derived neurons (NTera2-N) produced significantly greater amounts of IL-15 protein, while it was undetectable in culture supernatants of IMR-32, KG-1-C, NTera2-undif, K-562, HL-60 and HeLa cells (Fig. 3). This indicates that not all human neural and nonneural cell lines with IL-15 mRNA expression can secrete significant amounts of IL-15 protein in the culture medium.

3.3. The differential expression of IL-15 R mRNA in human cells and tissues of neural and nonneural origins Among the human neural cell lines examined, the expression of a high level of IL-15Ra mRNA was observed in U-373MG cells in which at least three distinct IL-15Ra-specific bands with the size of 623 bp, 503 bp, or

Fig. 3. IL-15 protein expression in culture supernatants of human neural and nonneural cell lines. The expression of IL-15 protein in undiluted supernatants harvested from the confluent cultures of a panel of human neural and nonneural cell lines was analyzed by a human IL-15 ELISA kit using recombinant human IL-15 as a standard. The lower limit of detection was less than 10 pg / ml in this assay. The data represent the average derived from the assays performed in duplicate.

Fig. 4. IL-15R mRNA expression in human neural cell lines. Fifty nanograms of cDNA was amplified by PCR using the primers specific for the human IL-15Ra, IL-2Rb or IL-2Rg gene listed in Table 1. The amplified products were separated on a 1.5% agarose gel, transferred onto a nylon membrane and hybridized with specific internal oligonucleotide probes. Lanes (1 to 9) represent as follows: (1) Y79 retinoblastoma cells, (2) IMR-32 neuroblastoma cells, (3) SK-N-SH neuroblastoma cells, (4) U-373MG glioma cells, (5) KG-1-C glioma cells, (6) NTera2 teratocarcinoma cells in the undifferentiated state, (7) NTera2-derived differentiated neurons induced by treatment with retinoic acid, (8) peripheral blood mononuclear cells without inclusion of the reverse transcriptase in the procedure of cDNA synthesis and (9) peripheral blood mononuclear cells with reverse transcription. The DNA size marker is shown on the left. Panels A, C and E: ethidium bromide staining of the gels. Panels B, D and F: Southern blot of the corresponding gels using the specific internal oligonucelotide probes listed in Table 1. (A, B) IL-15Ra, (C, D) IL-2Rb and (E, F) IL-2Rg.

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

scripts was predominant (Fig. 4A and B, lanes 5, 6 and 9). A very low level of IL-15Ra message was detectable in SK-N-SH cells, while it was below the detection limit in Y79, IMR-32 and NTera2-N cells (Fig. 4A and B, lanes 1, 2 and 7). Although our PCR analysis is not absolutely quantitative, the differences in IL-15Ra mRNA levels among these cell lines were significant because they expressed almost constant levels of G3PDH mRNA (Fig. 1D, lanes 1–7 and 9). These observations indicate that the expression of IL-15Ra mRNA is not universal in human neural cell lines. By contrast, IL-2Rb mRNA expression was detectable at varying levels in all the human neural cell lines examined (Fig. 4C and D, lanes 1–7). It was higher in IMR-32, SK-N-SH, KG-1-C and NTera2-undif cells than the levels in Y79, U-373MG and NTera2-N cells (Fig. 4C and D, lanes 1–7). High levels of IL-2Rg mRNA were expressed in Y79, U-373MG cells and PBMC, while it was undetectable in other neural cell lines (Fig. 4E and F, lanes 1–7). No products were amplified in total RNA isolated from each sample processed for PCR without reverse transcription (Fig. 4A–F, lane 8 and other data not shown). These observations indicate that the expression pattern of genes encoding three IL-15R components is varied among human neural cell lines. A high level of IL-15Ra mRNA was expressed in the human cerebral and cerebellar tissues and cultured skin fibroblasts (Fig. 5A and B, lanes 3, 4 and 7). Lower levels of IL-15Ra mRNA were detectable in the human skeletal muscle, peripheral nerve and thymoma tissues and in addition, an extremely low level in the neurinoma tissue (Fig. 5A and B, lanes 1, 2, 5 and 6). Three distinct IL-15Ra-specific bands of 623 bp, 503 bp and 403 bp were expressed in cultured skin fibroblasts, while the expression of two bands (503 bp and 403 bp) was predominant in the skeletal muscle, cerebrum, cerebellum and thymoma (Fig. 5A and B, lanes 1, 3, 4, 6 and 7). One weak band with the smallest size (403 bp) was detectable in the peripheral nerve (Fig. 5A and B, lane 2). The expression of IL-2Rb mRNA was detectable at substantial levels in all the human tissues examined (Fig. 5C and D, lanes 1–7). It was higher in the peripheral nerve, cerebrum, neurinoma, thymoma and cultured skin fibroblasts as compared to the levels in the skeletal muscle and cerebellar tissues (Fig. 5C and D, lanes 1–7). Appreciable levels of IL-2Rg mRNA were expressed in all the human tissues examined but undetectable in cultured skin fibroblasts (Fig. 5E and F, lanes 1–7). These observations indicate that the expression pattern of mRNAs for the three IL-15R components is varied among human neural and nonneural tissues. High levels of IL-15Ra and IL-2Rb mRNAs were identified in all the human nonneural cell lines examined (data not shown). Three intense IL-15Ra-specific bands of 623 bp, 503 bp and 403 bp were detectable in HUT102, K-562, HeLa and A549 cell lines and two bands (503 bp and 403 bp) were in MOLT-4 and HepG2, while only one

175

Fig. 5. IL-15R mRNA expression in human neural tissues. Fifty nanograms of cDNA was amplified by PCR using the primers specific for the human IL-15Ra, IL-2Rb or IL-2Rg gene listed in Table 1. The amplified products were separated on a 1.5% agarose gel, transferred onto a nylon membrane and hybridized with specific internal oligonucleotide probes. Lanes (1 to 7) represent as follows: (1) skeletal muscle, (2) peripheral nerve, (3) cerebrum, (4) cerebellum, (5) neurinoma, (6) thymoma and (7) cultured skin fibroblasts. The DNA size marker is shown on the left. Panels A, C and E: ethidium bromide staining of the gels. Panels B, D and F: Southern blot of the corresponding gels using the specific internal oligonucelotide probes listed in Table 1. (A, B) IL-15Ra, (C, D) IL-2Rb and (E, F) IL-2Rg.

major band of 403 bp was identified in HL-60 cells. High levels of the IL-2Rg message were also detectable in most of these cell lines except for A549 cells in which the level is extremely low (data not shown).

4. Discussion The present study for the first time has demonstrated that the expression of IL-15 mRNA was constitutive and

176

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177

universal in the human neural cell lines and tissues, whereas the messages for three IL-15R components were found to be expressed differentially. The results of our study disagree with two previous studies which reported that the messages for IL-15 and IL-15R were undetectable in the human and mouse brains (Grabstein et al., 1994; Giri et al., 1995b). The discrepancy between those studies might be in part attributable to differences in the materials examined (brains of undefined ages vs adult), methods utilized (Northern blot analysis vs RT–PCR analysis), or both. High levels of IL-15Ra mRNA were identified in U-373MG glioma cells and both cerebral and cerebellar tissues, while it was significantly lower in the skeletal muscle and peripheral nerve tissues. IL-2Rb mRNA was expressed at varying levels in all the human cell lines and tissues examined, while IL-2Rg mRNA expression was confined to Y79 and U-373MG cells, in addition to the skeletal muscle, peripheral nerve and cerebral and cerebellar tissues. Substantial levels of mRNAs for the complete set of IL-15R subunits were identified exclusively in U373MG cells and cerebral and cerebellar tissues. Our findings raise a possibility that IL-15 can act as an autocrine or paracrine growth / differentiation factor for certain neural cell types bearing the complete or partial IL-15 / IL-15R system. Because IL-15 was detectable in fetal human astrocytes and microglia in culture at both mRNA and protein levels (Lee et al., 1996), it should be determined whether IL-15 can stimulate proliferation of human astrocytes and microglia in vitro and whether IL-15 can be induced under pathological conditions in vivo such as in inflammatory or gliosis lesions of multiple sclerosis. Our study has shown that in a human teratocarcinoma cell line NTera2 which represents neuroepithelial stem cells with a differentiative capacity in response to treatment with retinoic acid (Pleasure et al., 1992), neuronal differentiation induced the expression of an alternatively spliced IL-15 mRNA isoform containing an additional exon termed exon A that exists between the second and third exons of the originally identified IL-15 mRNA coding sequence (Meazza et al., 1996; Onu et al., 1997). The IL-15 protein derived from the transcript of IL-15 with exon A, which contains three stop codons followed by a new initiation codon, differs only in the first portion of the signal peptide from the originally reported IL-15 precursor (Meazza et al., 1996). It has been shown that human small cell lung cancer cell lines expressed preferentially the exon A-containing IL-15 transcript (Meazza et al., 1996), although biological implications of the existence of two distinct isoforms remain unknown. In NTera2 cells, the induction of exon A-containing transcript was followed by increased production of the IL-15 protein. Inversely, the expression of IL-15Ra and IL-15Rb messages was downregulated in NTera2-derived differentiated neurons. Our observations suggest that the splicing events in IL-15 gene transcription might be under the control of as yet undefined differentiation-dependent molecules in human neurons.

Recently, we have shown that retinoic acid-induced neuronal differentiation up-regulates expression of the mRNAs for neurotrophins and their receptors in NTera2 cells (Satoh et al., 1997). These findings suggest that human neuronal development is controlled by a complex regulation of a variety of genes coding for cytokines, growth factors and their receptors in a differentiation-dependent manner. In our study, IL-15 protein was identified in culture supernatants of several but not all human cell lines which expressed significant levels of IL-15 mRNA. In agreement with this observation, previous studies have shown that it is not always easy to identify IL-15 protein in vivo and in vitro in spite of the widespread distribution of IL-15 mRNA expression (Giri et al., 1995a; Bamford et al., 1996b; Tagaya et al., 1996). The discordance between IL-15 mRNA and protein levels observed in the human neural cell lines suggests that IL-15 protein synthesis is regulated posttranscriptionally in certain cell types. It has been shown that the existence of multiple upstream AUGs in the 59 untranslated region (UTR) of the human IL-15 gene results in a reduction of its translation (Bamford et al., 1996a,b). IL-15 is readily detectable at the protein level in HTLV-I-infected T-cell lymphoma cell line, HUT102, in which IL-15 protein is produced following formation of a chimeric mRNA composed of a segment of the R region of the long terminal repeat of HTLV-I joining with the shortened form of the 59 UTR of the IL-15 gene (Bamford et al., 1996a,b). In conclusion, IL-15 mRNA expression is constitutive and universal in human neural cell lines and tissues in which IL-15R mRNA is expressed differentially. Our observations put forth a hypothesis that IL-15, a novel T-cell growth factor, may play a role as a neuroimmune regulatory cytokine. The cell types producing and responding to IL-15 in the human CNS under physiological and pathological conditions remain to be investigated.

Acknowledgements This work is supported by grants to J.S. from the Naito Foundation, Tokyo, Japan and from the Ministry of Education, Science and Culture of Japan (C2-08670715).

References Anderson, D.M., Kumaki, S., Ahdieh, M., et al., 1995. Functional characterization of the human interleukin-15 receptor chain and close linkage of IL15RA and IL2RA genes. J. Biol. Chem. 270, 29862– 29869. Bamford, R.N., Battiata, A.P., Burton, J.D., Sharma, H., Waldmann, T.A., 1996a. Interleukin (IL) 15 / IL-T production by the adult T-cell leukemia cell line HuT-102 is associated with a human T-cell lymphotrophic virus type I R region / IL-15 fusion message that lacks

J.-i. Satoh et al. / Journal of Neurological Sciences 155 (1998) 170 – 177 many upstream AUGs that normally attenuate IL-15 mRNA translation. Proc. Natl. Acad. Sci. USA. 93, 2897–2902. Bamford, R.N., Battiata, A.P., Waldmann, T.A., 1996b. IL-15: the role of translational regulation in their expression. J. Leukoc. Biol. 59, 476– 480. Carson, W.E., Giri, J.G., Lindemann, M.J., et al., 1994. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J. Exp. Med. 180, 1395–1403. Carson, W.E., Ross, M.E., Baiocchi, R.A., et al., 1995. Endogeneous production of interleukin 15 by activated human monocytes is critical for optimal production of interferon-g by natural killer cells in vitro. J. Clin. Invest. 96, 2578–2582. Edelbaum, D., Mohamadzadeh, M., Bergstresser, P.R., Sugamura, K., Takashima, A., 1995. Interleukin (IL)-15 promotes the growth of murine epidermal gd T cells by a mechanism involving the b- and gc -chains of the IL-2 receptor. J. Invest. Dermatol. 105, 837–843. Ercolani, L., Florence, B., Denaro, M., Alexander, M., 1988. Isolation and complete sequence of a functional human glyceraldehyde-3-phosphate dehydrogenase gene. J. Biol. Chem. 263, 15335–15341. Giri, J.G., Anderson, D.M., Kumaki, S., Park, L.S., Grabstein, K.H., Cosman, D., 1995. IL-15, a novel T-cell growth factor that shares activities and receptor components with IL-2. J. Leukoc. Biol. 57, 763–766. Giri, J.G., Kumaki, S., Ahdieh, M., et al., 1995. Identification and cloning of a novel IL-15 binding protein that is structurally related to the a chain of the IL-2 receptor. EMBO J. 14, 3654–3663. Grabstein, K.H., Eisenman, J., Shanebeck, K., et al., 1994. Cloning of a T-cell growth factor that interacts with the b chain of the interleukin-2 receptor. Science 264, 965–968. Hatakeyama, M., Tsudo, M., Minamoto, S., et al., 1989. Interleukin-2 receptor b chain gene: generation of three receptor forms by cloned human a and b chain cDNA’s. Science 244, 551–556. Johnston, J.A., Bacon, C.M., Finbloom, D.S., et al., 1995. Tyrosine phosphorylation and activation of STAT5, STAT3 and Janus kinases by interleukins 2 and 15. Proc. Natl. Acad. Sci. USA. 92, 8705–8709. Kanegane, H., Tosato, G., 1996. Activation of naive and momory T cells by interleukin-15. Blood 88, 230–235. Leclercq, G., Debacker, V., De Smedt, M., Plum, J., 1996. Differential effects of interleukin-15 and interleukin-2 on differentiation of bipotential T / natural killer progenitor cells. J. Exp. Med. 184, 325– 336. Lee, Y.B., Satoh, J.-I., Walker, D.G., Kim, S.U., 1996. Interleukin-15 gene expression in human astrocytes and microglia in culture. Neuroreport 7, 1062–1066.

177

McInnes, I.B., Al-Mughales, J., Field, M., et al., 1996. The role of interleukin-15 in T-cell migration and activation in rheumatoid arthritis. Nature Med. 2, 175–182. Meazza, R., Verdiani, S., Biassoni, R., et al., 1996. Identification of a novel interleukin-15 (IL-15) transcript isoform generated by alternative splicing in human small cell lung cancer cell lines. Oncogene 12, 2187–2192. Mori, A., Suko, M., Kaminuma, O., et al., 1996. IL-15 promotes cytokine production of human T helper cells. J. Immunol. 156, 2400–2405. ´ Mrozek, E., Anderson, P., Caligiuri, M.A., 1996. Role of inteleukin-15 in the development of human CD56 1 natural killer cells from CD34 1 hematopoietic progenitor cells. Blood 87, 2632–2640. ´ Nieto, M., del Pozo, M.A., Sanchez-Madrid, F., 1996. Interleukin-15 induces adhesion receptor redistribution in T lymphocytes. Eur. J. Immunol. 26, 1302–1307. Nishimura, H., Hiromatsu, K., Kobayashi, N., et al., 1996. IL-15 is a novel growth factor for murine gd T cells induced by Salmonella infection. J. Immunol. 156, 663–669. Onu, A., Pohl, T., Krause, H., Bulfone-Paus, S., 1997. Regulation of IL-15 secretion via the leader peptide of two IL-15 isoforms. J. Immunol. 158, 255–262. Pleasure, S.J., Page, C., Lee, V.M.-Y., 1992. Pure, postmitotic, polarized human neurons derived from NTera2 cells provide a system for expressing exogeneous proteins in terminally differentiated neurons. J. Neurosci. 12, 1802–1815. Satoh, J.-I., Lee, Y.B., Kim, S.U., 1995. T-cell co-stimulatory molecules B7-1 (CD80) and B7-2 (CD86) are expressed in human microglia but not in astrocytes in culture. Brain Res. 704, 92–96. Satoh, J.-I., Yukitake, M., Kurohara, K., Kuroda, Y., 1997. Retinoic acid-induced neuronal differentiation regulates expression of mRNAs for neurotrophins and neurotrophin receptors in a human embryonal carcinoma cell line NTera2. Neuropathology 17, 80–88. Tagaya, Y., Bamford, R.N., DeFilippis, A.P., Waldmann, T.A., 1996. IL-15: A pleiotropic cytokine with diverse receptor / signaling pathways whose expression is controlled at multiple levels. Immunity 4, 329–336. Takeshita, T., Asao, H., Ohtani, K., et al., 1992. Cloning of the g chain of the human IL-2 receptor. Science 257, 379–382. Warren, H.S., Kinnear, B.F., Kastelein, R.L., Lanier, L.L., 1996. Analysis of the costimulatory role of IL-2 and IL-15 in initiating proliferation of resting (CD56 dim) human NK cells. J. Immunol. 156, 3254–3259. Wilkinson, P.C., Liew, F.Y., 1995. Chemoattraction of human blood T lymphocytes by interleukin-15. J. Exp. Med. 181, 1255–1259.