Transcriptional activation of the human cytotoxic serine protease gene CSP-B in T lymphocytes

Acquisition of murine NK cell cytotoxicity requires the translation of a pre-existing pool of granzyme B and perforin mRNAs

Although activated murine NK cells can use the granule exocytosis pathway to kill target cells immediately upon recognition, resting murine NK cells are minimally cytotoxic for unknown reasons. Here, we showed that resting NK cells contained abundant granzyme A, but little granzyme B or perforin; in contrast, the mRNAs for all three genes were abundant. Cytokine-induced in vitro activation of NK cells resulted in potent cytotoxicity associated with a dramatic increase in granzyme B and perforin, but only minimal changes in mRNA abundance for these genes. The same pattern of regulation was found in vivo with murine cytomegalovirus infection as a physiologic model of NK cell activation. These data suggest that resting murine NK cells are minimally cytotoxic because of a block in perforin and granzyme B mRNA translation that is released by NK cell activation.

Activated status of tumour-infiltrating lymphocytes and apoptosis in testicular seminoma

Testicular seminoma is characterized by a prominent lymphoid infiltrate and an excellent prognosis. Cytotoxic T-lymphocytes (CTLs) infiltrating seminoma tumour nests constitute a major subset of the lymphoid infiltrate. The objective of this study was to determine whether CTLs express markers of cytotoxic potential and activity and whether the number of activated CTLs correlates with the extent of apoptosis in testicular seminomas, as opposed to non-seminomatous testicular germ cell tumours (NSTGCTs). Twenty cases of pure seminoma as well as 20 cases of NSTGCTs including 16 mixed germ cell tumours (MGCTs) were studied. Immunohistochemistry for the cytotoxic markers TIA-1 (cytotoxic potential) and granzyme B (cytotoxic activity) and the T-cell markers CD3 and CD8 was performed on formalin-fixed, paraffin-embedded sections. The apoptotic index (AI) was determined by the TUNEL method. The number of CD3(+), CD8(+), TIA-1(+), and granzyme B(+) cells in tumour cell nests was markedly increased in testicular seminomas, compared with NSTGCTs (p<0.01). Activated granzyme B(+) cells numbered 25.6+/-5.2 per high power field in seminomas and 8.9+/-3.2, 8.1+/-3.9, and 0.4+/-0.2 for embryonal carcinomas, yolk sac tumours, and immature teratomas, respectively. Double immunohistochemical staining for granzyme B and CD8 revealed that 82.6+/-8.5% of granzyme B-expressing cells were CD8(+). The tumour cell AI was significantly increased in embryonal carcinoma, compared with the seminoma, yolk sac tumour, and immature teratoma subgroups (6.7+/-1.3, 2.3+/-0.3, 3.0+/-1.1, and 2.3+/-1.1, respectively, p<0.001). TUNEL/CD3 double immunostaining revealed that a significant proportion of the apoptotic seminomatous tumour cells were in direct contact with one or more CD3(+) lymphocytes (47.2+/-6.2%). The number of activated granzyme B(+) CTLs showed a strong linear correlation with the AI in the seminoma group (r=0.71, p<0.0001) but not in other subgroups. TUNEL/granzyme B double immunolabelling revealed that a proportion of activated granzyme B(+) lymphocytes (20%) were often seen in close contact with apoptotic tumour cells. The presence of increased numbers of activated cytotoxic lymphocytes in testicular seminomas suggests that apoptotic tumour cell death in this neoplasm may be triggered by cytotoxic granule effectors. This phenomenon may be one of the key host immune mechanisms leading to the excellent prognosis in this tumour.

The nature of the myenteric infiltrate in achalasia: an immunohistochemical analysis

Achalasia is an esophageal motor disorder characterized by abnormal relaxation of the lower esophageal sphincter and absence of progressive peristalsis in the esophageal body. Previous studies evaluating esophagomyotomy and esophageal resection specimens have shown the presence of myenteric inflammation to be a consistent and early pathologic change in patients with achalasia. Thus, the goal of this study was to determine the immunohistochemical characteristics of the inflammatory infiltrate within the myenteric plexus in patients with clinically early and end-stage achalasia. Using formalin-fixed tissue, we analyzed the immunohistochemical features of the myenteric lymphocytes using antibodies that recognize B cells (CD20), T cells (CD3), T cell subsets (CD8), and the activation state of T cell subpopulations (TIA-1 and granzyme B) in nine patients with clinically early achalasia who underwent esophagomyotomy and 13 patients with clinically endstage achalasia who underwent esophageal resection. The myenteric infiltrate in all nine esophagomyotomy specimens was composed predominantly of T cells (CD3-positive), the majority of which also stained for CD8. In five of nine specimens, the majority of CD8-positive cells stained for TIA-1. In the esophageal resection specimens, the myenteric infiltrate was composed predominantly of CD3-positive T cells in seven of 13 cases. In three cases, there was a predominance of CD20-positive B cells, and in the remaining three cases there were relatively equal numbers of T and B cells. In eight of 13 cases, the majority of T cells stained for CD8. TIA-1 immunoreactivity was found in the majority of CD8-positive cells in nine of 13 cases. In all esophagomyotomy and esophageal resection specimens studied, rare granzyme B-positive cells were detected. In conclusion, the majority of myenteric inflammatory cells in patients with achalasia are CD3-positive T cells, most of which are also CD8-positive, although the relative percentage of such cells appears to decrease with disease progression. Furthermore, many of the CD3-positive/CD8-positive myenteric lymphocytes also express TIA-1, suggesting they are resting or activated cytotoxic T cells. The immunohistochemical demonstration of granzyme B in a subpopulation of these cells supports the contention that achalasia is an immune-mediated disease, although the inciting antigen remains an enigma.

Characterization of Ly-6M, a novel member of the Ly-6 family of hematopoietic proteins

The Ly-6 family includes a number of highly homologous, low molecular weight glycophosphatidylinositol-linked proteins expressed on hematopoietic and lymphoid cells. The best characterized family member is Sca-1 (Ly-6A/E), an antigen commonly used for purification of murine pluripotent hematopoietic cells. We sought to characterize the genomic locus surrounding the Sca-1 gene. We identified several overlapping P1 artificial chromosomes containing theSca-1 gene and mapped one of these to mouse chromosome 15D3.1-3.3, the region previously shown to contain members of the murine Ly-6 gene family. We then mapped this clone and found that the Sca-2 gene lies 35.4 kilobase (kb) downstream ofSca-1 in the opposite transcriptional orientation. This is the first direct demonstration of physical linkage of Ly-6 genes. A novel gene, highly homologous to Sca-1 was identified and localized 13.4 kb downstream of Sca-1. This gene, which we designated Ly-6M, shares several structural features conserved among members of the Ly-6 family. Ly-6M messenger RNA (mRNA) is easily detectable in hematopoietic tissue (bone marrow, spleen, thymus, peritoneal macrophages) as well as kidney and lung. No mRNA expression was detected in heart, stomach, liver, small intestine, brain, or skin. Ly-6M protein is detectable on 10% to 15% of peripheral blood leukocytes, including monocytes and a subpopulation of B220+ cells. Ly-6M is broadly distributed in the bone marrow, with prominent expression on monocytes and myeloid precursors. The identification and characterization of Ly-6M adds a new member to a complex family of homologous, tightly linked genes that have proven extremely useful reagents for defining populations within the hematopoietic system.

Characterization of Ly-6M, a novel member of the Ly-6 family of hematopoietic proteins

The Ly-6 family includes a number of highly homologous, low molecular weight glycophosphatidylinositol-linked proteins expressed on hematopoietic and lymphoid cells. The best characterized family member is Sca-1 (Ly-6A/E), an antigen commonly used for purification of murine pluripotent hematopoietic cells. We sought to characterize the genomic locus surrounding the Sca-1 gene. We identified several overlapping P1 artificial chromosomes containing theSca-1 gene and mapped one of these to mouse chromosome 15D3.1-3.3, the region previously shown to contain members of the murine Ly-6 gene family. We then mapped this clone and found that the Sca-2 gene lies 35.4 kilobase (kb) downstream ofSca-1 in the opposite transcriptional orientation. This is the first direct demonstration of physical linkage of Ly-6 genes. A novel gene, highly homologous to Sca-1 was identified and localized 13.4 kb downstream of Sca-1. This gene, which we designated Ly-6M, shares several structural features conserved among members of the Ly-6 family. Ly-6M messenger RNA (mRNA) is easily detectable in hematopoietic tissue (bone marrow, spleen, thymus, peritoneal macrophages) as well as kidney and lung. No mRNA expression was detected in heart, stomach, liver, small intestine, brain, or skin. Ly-6M protein is detectable on 10% to 15% of peripheral blood leukocytes, including monocytes and a subpopulation of B220+ cells. Ly-6M is broadly distributed in the bone marrow, with prominent expression on monocytes and myeloid precursors. The identification and characterization of Ly-6M adds a new member to a complex family of homologous, tightly linked genes that have proven extremely useful reagents for defining populations within the hematopoietic system.

Aberrant esophageal HLA-DR expression in a high percentage of patients with Crohn's disease

Esophageal histology is not well studied in patients with Crohn's disease (CD). We, therefore, analyzed the histologic and immunohistologic appearance of esophageal mucosa in CD. Biopsy specimens taken from the esophagus of 57 consecutive patients with known CD of the large and/or small bowel, of 200 Crohn's-free controls, of 15 cases with ulcerative colitis, and of 5 cases with viral esophagitis were evaluated. In controls, most patients had either HLA-DR negative esophageal epithelium or showed focal or diffuse basal staining. HLA-DR expression of all epithelial layers (transepithelial staining) was observed in only four (2%) control subjects, in one case with herpes esophagitis, but not in patients with ulcerative colitis. In contrast, transepithelial HLA-DR expression was found in 19 (33%) patients with CD (p < 0.0001). In CD patients, it was associated with a significantly increased epithelial content in T-cells (CD3+, TIA-1+, granzyme B+), B-cells (CD79a+), natural killer cells (CD57+), and macrophages (CD68+). There was no correlation with either histological findings elsewhere in the upper gastrointestinal tract or with laboratory findings, symptoms, CDAI, or medication. Transepithelial esophageal HLA-DR expression is common in CD. Immunohistochemistry may prove useful in supporting the histologic diagnosis of CD in staging procedures, for initial diagnosis as well as in doubtful cases.

Assessment and diagnostic utility of the cytotoxic T-lymphocyte phenotype using the specific markers granzyme-B and TIA-1 in esophageal mucosal biopsies

Most esophageal intraepithelial lymphocytes (IELs) express T-cell markers. Increased numbers of esophageal IELs have been shown in reflux esophagitis. The cytotoxic potential and activity of esophageal IELs have not as yet been examined. Our objectives were to determine whether esophageal IELs express the recently described cytotoxic T-cell (CTLs) markers, TIA-1 and granzyme-B, and whether the number of CTLs correlates with well-defined endoscopic, clinical, and histological features of esophagitis. In this study, most CD-3+ esophageal IELs exhibit the CD-8+/TIA-1+ T cell with cytotoxic potential phenotype in both histologically normal biopsy specimens and in biopsy specimens with esophagitis. A subpopulation of esophageal IELs that express cytotoxic activity was identified by granzyme-B immunostaining. A significant positive association was found between the number of esophageal IELs seen by light microscopy in biopsy specimens with histological features of reflux (21 IELs/HPF) and Candida esophagitis (31 IELs/HPF) as compared with normal-appearing biopsy specimens (10 IELs/HPF) (P< or =.05). Furthermore, the number of TIA-1 or granzyme-B-positive IELs were significantly increased in biopsy specimens with reflux esophagitis (34 and 15 cells/HPF) and Candida esophagitis (44 and 18 cells/HPF) as compared with normal (11 and 2 cells/HPF) (P< or =.05). Granzyme-B and CD-3-positive IELs were also significantly elevated in biopsy specimens with reflux-associated squamous hyperplasia (P< or =.05). Finally, biopsy specimens of patients with dysphagia and to a lesser extent dyspepsia/heartburn exhibited increased numbers of IELs bearing the cytotoxic phenotype when compared with asymptomatic patients. In conclusion, we provide immunohistochemical evidence that most esophageal IELs exhibit the cytotoxic phenotype and that activated cytotoxic IELs are increased in reflux and Candida esophagitis.

The 5′ Flanking Region of the Human Granzyme H Gene Directs Expression to T/Natural Killer Cell Progenitors and Lymphokine-Activated Killer Cells in Transgenic Mice

Human granzyme H is a neutral serine protease that is expressed predominantly in the lymphokine-activated killer (LAK)/natural killer (NK) compartment of the immune system. The gene that encodes this granzyme is located between the granzyme B and cathepsin G genes on human chromosome 14q11.2. Although the murine orthologue of human granzyme H has not yet been identified, murine granzymes C, D, E, F, and G also lie between the murine granzyme B and cathepsin G genes on murine chromosome 14; murine granzymes C, D, and F are also highly expressed in LAK cells, but minimally in cytotoxic T lymphocytes (CTL). We therefore tested whether the 5′ flanking region of human granzyme H contains the cis-acting DNA sequences necessary to target a reporter gene to the LAK/NK compartment of transgenic mice. A 1.2-kb fragment of 5′ flanking human granzyme H sequence was linked to an SV40 large T-antigen (TAg) reporter gene and used to create six transgenic founder lines. SV40 TAg was specifically expressed in the LAK cells of these mice, but not in resting T or NK cells, in CTL, or in any other tissues. Most mice eventually developed a fatal illness characterized by massive hepatosplenomegaly and disseminated organ infiltration by large malignant lymphocytes. Cell lines derived from splenic tumors were TAg+ and NK1.1+ large granular lymphocytes and displayed variable expression of CD3, CD8, and CD16. Although these cell lines contained perforin and expressed granzymes A, B, C, D, and F, they did not exhibit direct cytotoxicity. Collectively, these results suggest that the 5′ flanking sequences of the human granzyme H gene target expression to an NK/T progenitor compartment and to activated NK (LAK) cells. Mice and humans may therefore share a regulatory “program” for the transcription of NK/LAK specific granzyme genes.

The 5′ Flanking Region of the Human Granzyme H Gene Directs Expression to T/Natural Killer Cell Progenitors and Lymphokine-Activated Killer Cells in Transgenic Mice

Human granzyme H is a neutral serine protease that is expressed predominantly in the lymphokine-activated killer (LAK)/natural killer (NK) compartment of the immune system. The gene that encodes this granzyme is located between the granzyme B and cathepsin G genes on human chromosome 14q11.2. Although the murine orthologue of human granzyme H has not yet been identified, murine granzymes C, D, E, F, and G also lie between the murine granzyme B and cathepsin G genes on murine chromosome 14; murine granzymes C, D, and F are also highly expressed in LAK cells, but minimally in cytotoxic T lymphocytes (CTL). We therefore tested whether the 5′ flanking region of human granzyme H contains the cis-acting DNA sequences necessary to target a reporter gene to the LAK/NK compartment of transgenic mice. A 1.2-kb fragment of 5′ flanking human granzyme H sequence was linked to an SV40 large T-antigen (TAg) reporter gene and used to create six transgenic founder lines. SV40 TAg was specifically expressed in the LAK cells of these mice, but not in resting T or NK cells, in CTL, or in any other tissues. Most mice eventually developed a fatal illness characterized by massive hepatosplenomegaly and disseminated organ infiltration by large malignant lymphocytes. Cell lines derived from splenic tumors were TAg+ and NK1.1+ large granular lymphocytes and displayed variable expression of CD3, CD8, and CD16. Although these cell lines contained perforin and expressed granzymes A, B, C, D, and F, they did not exhibit direct cytotoxicity. Collectively, these results suggest that the 5′ flanking sequences of the human granzyme H gene target expression to an NK/T progenitor compartment and to activated NK (LAK) cells. Mice and humans may therefore share a regulatory “program” for the transcription of NK/LAK specific granzyme genes.

High proportion of granzyme B-positive (activated) intraepithelial and lamina propria lymphocytes in lymphocytic gastritis

Intraepithelial lymphocytes (IELs) and lamina propria lymphocytes (LpLs) have not been well studied in gastric mucosa, particularly in lymphocytic gastritis. Therefore, they were immunohistologically characterized with antibodies recognizing CD3, CD8, CD57, T cell-restricted intracellular antigen (TIA-1), and granzyme B (GrB). The TIA-1 labels cytotoxic granules of resting and activated T-cells, whereas GrB decorates activated cytotoxic T cells. Thirty patients with celiac disease, including 20 taking gluten and 10 on a gluten-free diet, 15 patients with nonceliac disease-associated lymphocytic gastritis, and 20 controls were studied. Stained cells were counted and results were given as IELs/100 epithelial cells or percentage of lamina propria cells. Sixty percent to 90% of CD3+ IELs and up to 12% of lamina propria cells contained TIA-1-positive cytotoxic granules. The number of GrB+ IELs and LpLs was increased in Helicobacter pylori-positive controls (p < 0.03 vs. H pylori-negative controls) and celiac disease patients taking gluten (p < 0.05 vs. controls). The highest number of GrB+ IELs and LpLs was found in nonceliac disease-associated lymphocytic gastritis (p < 0.009 vs. controls, p < 0.05 vs. celiac disease). This study shows that a high proportion of gastric IELs and LpLs is potentially cytotoxic in nature. Through stimuli not yet identified, a proportion of them becomes activated after H pylori infestation and in lymphocytic gastritis.

T-cell receptor ligation by peptide/MHC induces activation of a caspase in immature thymocytes: the molecular basis of negative selection

T-cell receptors (TCRs) are created by a stochastic gene rearrangement process during thymocyte development, generating thymocytes bearing useful, as well as unwanted, specificities. Within the latter group, autoreactive thymocytes arise which are subsequently eliminated via a thymocyte-specific apoptotic mechanism, termed negative selection. The molecular basis of this deletion is unknown. Here, we show that TCR triggering by peptide/MHC ligands activates a caspase in double-positive (DP) CD4+ CD8+ thymocytes, resulting in their death. Inhibition of this enzymatic activity prevents antigen-induced death of DP thymocytes in fetal thymic organ culture (FTOC) from TCR transgenic mice as well as apoptosis induced by anti-CD3epsilon monoclonal antibody and corticosteroids in FTOC of normal C57BL/6 mice. Hence, a common caspase mediates immature thymocyte susceptibility to cell death.

Altered Myeloid Development and Acute Leukemia in Transgenic Mice Expressing PML-RARα Under Control of Cathepsin G Regulatory Sequences

Acute promyelocytic leukemia (APML) is characterized by abnormal myeloid development, resulting an accumulation of leukemic promyelocytes that are often highly sensitive to retinoic acid. A balanced t(15; 17) (q22; q21) reciprocal chromosomal translocation is found in approximately 90% of APML patients; this translocation fuses the PML gene on chromosome 15 to the retinoic acid receptor α (RARα) gene on chromosome 17, creating two novel fusion genes, PML-RARα and RARα-PML. The PML-RARα fusion gene product, which is expressed in virtually all patients with t(15; 17), is thought to play a direct role in the pathogenesis of APML. To determine whether PML-RARα is sufficient to cause APML in an animal model, we used the promyelocyte-specific targeting sequences of the human cathepsin G (hCG) gene to direct the expression of a PML-RARα cDNA to the early myeloid cells of transgenic mice. Mice expressing the hCG–PML-RARα transgene were found to have altered myeloid development that was characterized by increased percentages of immature and mature myeloid cells in the peripheral blood, bone marrow, and spleen. In addition, approximately 30% of transgene-expressing mice eventually developed acute myeloid leukemia after a long latent period. The splenic promyelocytes of mice with both the nonleukemic and leukemic phenotypes responded to all-trans retinoic acid (ATRA) treatment, which caused apoptosis of myeloid precursors. Although low-level expression of the hCG–PML-RARα transgene is not sufficient to directly cause acute myeloid leukemia in mice, its expression alters myeloid development, resulting in an accumulation of myeloid precursors that may be susceptible to cooperative transforming events.

In vivo regulation of murine granzyme B gene transcription in activated primary T cells

A murine granzyme B promoter fragment that extends 243 base pairs upstream of the transcription start site confers high levels of luciferase reporter gene activity in transient transfection assays into T cells and mouse L cell fibroblasts. This promoter fragment contains canonical binding sites for the transcription factors AP-1, core binding factor (CBF), Ikaros, and the cyclic AMP responsive element binding protein (CREB). Oligonucleotides containing the granzyme B AP-1 or CBF elements form specific complexes with proteins present in nuclear extracts from activated CD8(+) splenocytes, MTL cells, EL4 T cells, and L cells. A strong DNase1 hypersensitive site that coincides with the closely associated AP-1, CBF, Ikaros, and CRE elements is present in activated CD8(+) T cells but not in resting T cells or L cells. Both in vitro and in vivo footprints are observed at these sequence elements in activated cytotoxic T cells (CTL) but not in resting T cells. The endogenous granzyme B gene is CTL-specific as no mRNA is detectable in EL4 or L cells. We propose that a condensed chromatin structure at the granzyme B promoter is responsible for transcription factor inaccessibility and repression of transcription in non-T cells.

Evidence that intestinal intraepithelial lymphocytes are activated cytotoxic T cells in celiac disease but not in giardiasis

To further define intraepithelial lymphocytes (IELs) in celiac disease (CD) and giardiasis, IELs were probed for the presence of cytolytic granules containing granzyme B (GrB) and T-cell-restricted intracellular antigen (TIA)-1. The expression of TIA-1, GrB, CD3 (T-cell-receptor-associated complex), and Leu-7 (subset of natural killer cells) was studied by a sensitive three-step immunoperoxidase technique. Stained IELs were determined quantitatively, and results were expressed as number of stained IELs per 100 epithelial cells (ECs). The relative content in labeled lamina propria lymphocytes was determined and expressed as the percentage of all lamina propria cells counted. When compared with controls, CD3+ and GrB+ IELs were significantly increased (P < 0.0004) in CD paralleled by an increase in TIA-1+ IELs (P < 0.0004). In CD, the highest numbers of IELs containing GrB were found in subjects with a flat mucosa (median, 38 IELs/100 ECs, P < 0.0004), followed by cases with shortened and blunted villi (median, 8 IELs/100 ECs, P < 0.0004) and, finally, CD patients with an intact villous architecture (median, 0.5 IELs/100 ECs, P < 0.02). Except for cases with giardiasis, Leu-7+ IELs were virtually absent in all groups as were GrB+ IELs in the controls and in subjects with giardiasis. In the lamina propria of CD subjects, GrB+ lymphocytes were also significantly increased (P < 0.001), whereas controls and cases with giardiasis were essentially free of GrB+ cytotoxic T lymphocytes. The percentage of CD3+ lamina propria lymphocytes was nearly equal in all groups. In humans and mice, extensive studies revealed a GrB expression to be absolutely restricted to activated cytotoxic T lymphocytes and natural killer cells. TIA-1, on the other hand, is considered a marker of resting T lymphocytes possessing cytolytic potential. We therefore conclude that IELs are cytotoxic T cells that are in a resting state in the normal small bowel and in giardiasis. In CD, they become activated as suggested by the GrB positivity of their granules.

Cytotoxic lymphocytes require granzyme B for the rapid induction of DNA fragmentation and apoptosis in allogeneic target cells

We have generated H-2b mice with a homozygous null mutation in the granzyme (gzm) B gene. Gzm B is a neutral serine protease with Aspase activity that is found only in the granules of activated cytolytic T cells, natural killer cells, and lymphokine-activated killer cells. Gzm B-/- mice develop normally and have normal hematopoiesis and lymphopoiesis. In vitro, cytotoxic T lymphocytes (CTL) derived from gzm B-/- animals are able to induce 51Cr release from allotarget cells, but with reduced efficiency. However, gzm B-/- CTL have a profound defect in their ability to induce rapid DNA fragmentation and apoptosis in allogeneic target cells. This defect is kinetic since DNA fragmentation is partially compensated and 51Cr release is completely rescued with long incubation times. We conclude that gzm B serves a critical and nonredundant role for the rapid induction of target cell DNA fragmentation and apoptosis by alloreactive cytotoxic T lymphocytes.

A promoter element of the human serine esterase granzyme B gene controls specific transcription in activated T cells

The human granzyme B gene encodes a serine protease expressed specifically in cytoplasmic granules of cytotoxic T lymphocytes, released upon effector-target cell interaction. Previous studies have shown that granzyme B mRNA was induced in T lymphocytes after antigenic or mitogenic stimulation. To study the regulation of human granzyme B gene expression during lymphocyte activation we analyzed its 5' flanking region using chloramphenicol acetyl transferase (CAT) reporter gene constructs. We show that a 208-bp fragment (-148 to +60) containing an NF-AT (nuclear factor of activated T cells)-binding site promotes CAT expression in phytohemagglutinin-activated T lymphocytes, in immobilized monoclonal anti-CD3 antibody-activated Jurkat T cell line while it is inactive in unstimulated PEER and Jurkat T cells lines or B Epstein-Barr virus-transformed cell lines.

Genomic structure of NKG5, a human NK and T cell-specific activation gene

We reported previously the isolation of a cDNA clone, designated NKG5, encoding a secreted protein that is expressed only in natural killer and T cells and is strongly upregulated upon cell activation. In this report we have isolated the NKG5 gene from a human placental genomic library and sequenced the gene and two kilobases of 5'-flanking DNA. Comparison with the cDNA sequence reveals that the NKG5 gene consists of five exons and four introns. Intron 1 contains a DNA segment that was reported to occur as an exon in 519, a closely related cDNA clone that was isolated from a T-cell library. This result indicates that NKG5 and 519 are alternative splicing products of a single gene. The 5'-flanking region of the NKG5 gene was analyzed for homology with the promoter regions of cytokines and other activation-induced genes showing lymphocyte-specific expression. Several segments displaying sequence similarity were identified. We also identified numerous sequence elements that have strong similarity to known binding sites for transcriptional regulatory proteins including T cell-specific and activation-specific regulatory factors. These findings are consistent with the cell-specific expression and the tight regulatory control that is observed for the NKG5 gene.