Elsevier

Human Immunology

Volume 59, Issue 9, September 1998, Pages 571-579
Human Immunology

Human Immunology
Tumor Necrosis Factor locus: genetic organisation and biological implications

https://doi.org/10.1016/S0198-8859(98)00056-1Get rights and content

Abstract

TNF genes determine strength, effectiveness, and duration of local and systemic inflammatory reactions, as well as repair and recovery from infectious and toxic agents. Multiple pro- and anti-inflammatory activities of TNF factors are conditioned by their profound effects on metabolism of many cell types, their activation state, cell survival and others. TNF genes show strong linkage disequilibrium with HLA class I and class II genes and with other genes in the MHC region relevant to immuneregulation. Structural or regulatory defects in TNF genes may contribute to pathogenesis of MHC associated diseases especially those with inflammatory and autoimmune components.

Introduction

The monocyte/macrophage lineage or reticuloendothelial system in concert with blood granulocytes (neutrophils, eosinophils and basophils) represents a non-specific branch of immune responsiveness. Macrophages can ingest and destroy micro-organisms by phagocytosis or, in a similar way to granulocytes, by secretion of cytotoxic compounds from their intracellular granules to their local environment when they encounter micro-organisms. Macrophages can also act more specifically by collaborating with lymphocytes and their product-cytokines. An organism defence reaction, inflammation, which results from co-operative and co-ordinated efforts of this first line defence system is directed to localise and restrict the spread of infection or of transformed cells, but when inappropriately strong and prolonged and/or located in vitally important organs, may cause morphological and functional damage leading to organ insufficiency and finally to death. Cells of the monocyte/macrophage lineage play a central role in inflammatory cytokine production and the most abundant product of activated macrophages is Cachectin or Tumor Necrosis Factor (TNF) [1]. TNF belongs to the TNF ligand family and has multiple biologic activities. Although initially TNF avoked attention as a factor able to elicit haemorrhagic necrosis of tumors in recipient animals [2] it is now believed that TNF is one of the main proinflammatory cytokines and plays a central role in initiating and regulating the cytokine cascade during an inflammation response and is involved in local and systemic events attendant an inflammation. Studies on TNF-deficient mice [3] showed that TNF also plays an essential homeostatic or anti-inflammatory role limiting the extent and duration of an inflammation and promotes repair and recovery from infectious and toxic agents. TNF deficient mice showed little or no initial response to infectious challenge but then developed a vigorous and disorganized inflammatory response leading to death.

TNF receptors belong to the TNF-nerve growth factor (NGF) receptor family and are expressed on virtually all somatic cells [1]. High affinity binding of TNF and of the LT-α cytokine to their receptors [4] mediates a wide range of their biological functions including immunological responses, inflammatory reactions, and anti-tumor and anti-viral activity.

The TNF locus is located within the central MHC (class III region), spans 12 kb and maps centromeric to the HLA-B locus and telomeric to the C2/Bf complex. (Fig. 1 ). The TNF locus contains genes coding for Cachectin/TNF (also called TNFα), lymphotoxin α (previously referred to as TNF β) and lymphotxin β proteins [5] and the human homologue of mouse B144 - Lst1 gene [6]. TNF (TNF A) as well as Lymphotoxin A genes (also called TNF B ) are each approximately 3 kb pairs long and are separated by only about 1100 base pairs. The LTA gene lies 5′ to TNF and both genes consist of four exons and three introns 1, 7, 8. The lymphotoxin B gene (LTB) lies 3′ to TNF, spans 2 kb and has a very similar arrangement to that of TNF and LTA but is oriented in the opposite direction (Fig.1). The LTB gene is contained within four exons and lacks an intron in the 5′ untranslated region [5]. The organisation of the TNF, LTA and LTB genes, especially of the last large exon that encodes 80–89% of the extracellular domain of the mature protein, is conserved in all three genes 1, 5, 7, 8.

Human TNF is a polypeptide hormone composed of three equal non-glycosylated sub-units arranged noncovalently in a compact and stable bell-shaped complex. Each sub-unit has a molecular weight of 17350 Dalton and consist largely of a beta pleated sheet structure 1, 7, 9. The monomer is composed of ten strands (named “a” through “h”) of which five are “inner” strands involved in trimmer interaction and five compose an outer sheet. The inner sheet facing the trimmer axis is essentially flat while the outer sheet is highly curved [9]. The amino- and carboxyl- terminal strands of the molecule (a,h) are packed together and form the first two strands of the inner sheet. Monomers are tightly packed into trimmer complex by hydrophobic residues. The mature hormone contains 157 amino acids, but initially, the hormone is synthetised as a pro-hormone and contains 76 additional amino acids attached to the N terminus of the molecule. The additional amino acids constitute an N-terminal hydrophobic domain that anchors the molecule on the cell surface. The mature or secreted form of the hormone is generated by proteolytic cleavage of the transmembrane portion of the molecule. Studies on TNF mutants resistant to proteolytic processing showed that the membrane bound form, like the secreted protein could induce cytotoxic function 10, 11. Furthermore recent data [12] demonstrate a crucial role for membrane-bound TNF in brain pathology of experimental cerebral malaria.

LT-α (or TNFβ) and LT-β are structurally related to TNF (TNFα) proteins and share some regions of sequence conservation especially in the region defined by the fourth large exon [5]. LT-α consists of 25 kD subunits, contains 171 residues and has one N-glycosilation site at residue 62 [13]. The lymphotoxin-α subunits form a trimmer similar to those of TNF which is stabilized primarily by interactions between hydrophobic and aromatic side chains [14]. Both hormones have similar spectra of biological activities and in native form are known to be homotrimers. The homology regions that contain many of the contact residues involved in forming the trimmers lie on the internal surfaces. 1, 7, 9, 15. Both TNF and LT-α exist in soluble or secreted forms although distinct mechanisms generate the soluble and membrane bound forms of these hormones. As mentioned above soluble TNF is generated by the proteolysis of the 76 amino acid peptide from the N-terminus of the prohormone while LT-α lacks the transmembrane portion and is exclusively secreted as a homotrimer. The membrane associated form of LT is produced by the assemblage of LT-α with the related LT-β protein in a heteromeric complex and by virtue of the transmembrane portion of LT-β molecule the complex is localised on the cell surface [5]. The LT-β protein has a short 15–19 amino acid N terminal cytoplasmic domain followed by 30 hydrophobic amino acids that presumably act as a membrane-anchoring domain [5]. The stoichiometry of the overall complex is believed to be LT α1:β2, but a small portion of the complex (about 10% of the total LT) may also exist in an α2:β1 ratio [16]. The heteromeric complex most likely retains a trimmeric structure similar to that of TNF and LT-α, with the homology regions interacting in a heterotypic fashion [5]. Although TNF and LT-α lack sequence homology in the region of the receptor binding site both cytokines bind and signal through two receptors, the 55–60 kD TNF receptor (TNFR60; CD120a or type 1) and the 75–80 kD TNFR (TNFR80; CD120b or type 2) and this may indicate that they employ alternative receptor binding conformation 14, 17. By contrast, the surface LT α1: β2 complex binds a related but distinct receptor, termed LTβ receptor that does not bind neither TNF nor LT-α, whereas both TNF receptors bind the LT α1β2 heteromeric complex 16, 18, 19. The existence of secreted and membrane bound forms of LT that engage distinct receptors may indicate that the membrane-associated LTα-LTβ complex has a separate role in immune regulation 5, 16. Studies on LTA deficient mice showed that LT-α has fundamental actions on lymphoid organogenesis [20]. Disruption of the LTA coding sequence by gene targeting in embryonic stem cells lead to abnormal development of peripheral lymphoid organs. In LTA deficient mice neither lymphoid aggregates nor rudiments of lymph node stromal tissue were detected [20]. In contrast genetic knockout of either TNFR 60 or TNFR80 did not result in a similar phenotype, which suggests a role for the surface LTα-LTβ complex and the LTβR signalling pathway in immune development [18]. The LT α1: β 2 complex is the most abundant form expressed by activated T cells and unlike TNF or LT-α it is not produced naturally in soluble form. The existence of a LTβ homotrimmer is uncertain, since the LTβ protein is apparently always associated with LT-α in T cells [16].

Studies of the polymorphism in the human TNF genes revealed the existence of four alleles for TNF B (LTA) gene at the DNA level detected by Nco I and Eco IR RFLP [21]. According to Nco I RFLP 2 alleles TNF B∗1 and TNFB∗2 were determined. The NcoI RFLP restriction site is located in the first intron of the TNFB∗1 allele which also contains AAC at position 26. The TNFB∗1 allele is less frequent (p = 0.06–0.1) [21] and shows strong association with the HLA-A1B8DR3 haplotype which in turn is associated with a wide range of autoimmune diseases such as IDDM, SLE, Graves disease, celiac disease, rheumatoid arthritis and others and with immune deviations and/or abnormalities in apparently healthy subjects 22, 23, 24, 25, 26. The TNFB∗1 allele leads to a higher TNF β response upon PHA stimulation of T lymphocytes. The TNFB∗2 allele lacks the Nco I restriction site and contains ACC at position 26. Thus TNFB alleles differ in structure and in the level of cytokine inducibility. Allelically varying differences in TNF-β inducibility in activated T lymphocytes might be relevant to the severity and/or duration of an inflammatory reaction of a local autoimmune process and may contribute to the HLA-associated predisposition for autoimmune diseases [21]. A two-allele polymorphism was described for the TNFA promoter region involving the substitution of Guanine by Adenosine at position −308 in the rare TNF2 allele [27] and at position −238 (presence of A sequence = TNFA-A allele and presence of G sequence = TNFA-G allele) [28]. As was shown for human cytochrome gene [29] a single base polymorphism within a promoter region may result in differences of the promoter activity and alter both the rate of gene transcription and the rate of protein production. Recent studies indicate that indeed TNF2 is a much stronger transcriptional activator than the common allele (TNF1) in a human B cell line [30]. The rare TNF2 and TNFA-A alleles showed strong association with the HLA-A1B8DR3 haplotype. The HLA-A1B8DR3 alleles are part of the ancestral or conserved haplotype 8.1 that can be regarded along with other conserved haplotypes as a unique genomic structure carrying a particular set of genes of the MHC region 31, 32. TNF alleles are in strong linkage disequilibrium with HLA class I and class II alleles as well as with other alleles in the region and it is possible that structural or regulatory defects in the TNF genes contribute to pathogenesis of HLA associated diseases 32, 33, 34.

Although coding regions of the human TNF genes show low polymorphism several polymorphic areas are documented within the TNF gene cluster. Polymorphic elements of the TNF locus include tandemly repeated dinucleotide sequences usually a CA or CT occurring at a unique location within the locus known as microsatellites. Five polymorphic microsatellite sequences (TNFa, TNFb, TNFc, TNFd, TNFe) were identified 35, 36, 37. TNFa and TNFb microsatellites are closely linked and located 3.5 kb upstream of the LTA gene, TNFe and TNFd are located 8–10 kb downstream of the TNF gene and TNFc is located in the first intron of the LTA gene (Fig. 1). Microsatellites differ in the length of their simple dinucleotide sequence elements and are said to be polymorphic. Based on the length of the polly (CA) or polly (CT) repeats there were identified 13 TNFa and 7 TNFb alleles 36, 37. The length of consecutively numbered TNFa alleles differ by 2 nucleotides (one repeat) and TNF b alleles differ by only 1 nucleotide [37]. TNFc shows less polymorphism and contains only 2 alleles. Six alleles for TNFd and 3 alleles for closely linked TNFe were characterised [35]. Microsatellites in the human TNF locus are inherited as stable alleles and segregate in a codominant manner with HLA haplotypes 36, 37. Microsatellites in the TNF locus show population polymorphism 38, 39, 40. The allele frequencies for the TNFa microsatellite differ significantly among populations. As was shown in 4 European populations [40] the most common TNFa microsatellite allele is TNFa2 (p = 0.208 –0.320). The TNFa1 allele showed high frequency among Basques (p = 0.116), a very low frequency in the French and Danish populations (p = 0.031 and p = 0.0067, respectively) and was completely absent in a Greek population. Some but not all of the TNF microsatellite alleles display strong linkage disequilibrium to each other and to other genes of the region. It was shown that the TNFd2a2b3 haplotype is associated with the ancestral or conserved haplotype HLA DQB1∗ 0201 DQA1∗0501 DRB1∗0301 C4A∗Q0 and clinical peculiarities of SLE such as photosensitivity and Raynaud’s phenomenon [41]. A strong negative association between the TNFc2 microsatellite allele and faster HIV disease progression was recently reported. [42]. It is believed that polymorphism of microsatellites in the TNF locus by itself does not affect TNF genes structure or regulation [36] but they are widely used as reliable and convenient from a methodological stand point as genetic markers for thorough genetic analysis of the involvement of TNF factors in MHC associated pathologies and for population genetic studies.

Stringent regulation of TNF gene expression upon infectious or tumorigenic challenge is a prerequisite for appropriate and effective immune responses and inflammatory reactions of the host. Although TNF, LTA, and LTB genes are closely linked in a 12 kb space and share some features in their organisation 1, 7 and gene regulatory elements (NF-κB like binding motifs) [43], they are differentially regulated in a tissue specific and inducer specific manner 43, 44, 45 suggesting distinct function for each cytokine depending on the course of immune responsiveness. Studies on murine T cell clones activated through the TCR [46] showed dramatic differences in the molecular mechanisms of regulation of TNF, LTA and LTB genes at the transcriptional and post-transcriptional levels. Anti CD3 treatment of T lymphocytes activated TNF and LTA gene transcription at different rate and at the same time did not cause any increase of LTβ mRNA which is actively transcribed without an activating signal. Striking differences in the decay rates of TNF, LT-α and LTβ mRNAs observed in the study could be related to the difference in the number of AUUUA motifs present in their 3′ sequences and in the context within which this motif occurs [46]. Monocytes as well as tissue macrophages although last ones less efficiently, secrete TNF only in response to infectious challenge [47]. Powerful inducers of TNF gene transcription and monokine secretion are bacterial endotoxin lipopolysacharide-LPS and bacterial exotoxins like staphylococcal superantigens although they engage different cell surface receptors. LPS binds to CD14 antigen in complex with soluble lipopolysacharide binding protein (LBP) and the exotoxin superantigen binds to MHC class II molecules 48, 49, 50. Transcriptional regulation of the human TNF gene is a complex process and involves multiple promoter regions and factors including NF-κB protein binding sites (consensus κB element GGGGACTTTCC) [51], a GC box/Sp-1 binding site, a “Y-box” like decanucleotide, a cAMP-responsive element-binding protein binding site, and binding sites for AP-1 and AP-2 and others [52]. The TNF promoter region contains four elements or potential binding sites for the NF-κB transcription factor that mediates signal transduction between cytoplasm and nucleus in many cell types. Studies on murine thioglycolate exudate peritoneal macrophages stimulated by LPS, cycloheximide (CH) and IFN-γ showed that these κB-like motifs have very different binding affinities [53]. The strongest ones are at site -510 (κB3) and -850 (κB1), whereas the sites at -655 (κB2) and -210 (κB4) are weakly recognised 53, 54. One or more enhancers of the “κB” family play major roles in LPS and bacterial exotoxin mediated transcriptional activation of macrophages. NF-κB seems to be involved in the initial activation event triggered by endo and exotoxins and in the transcriptional activation of the TNF gene 51, 55. Newly synthesised TNF protein as it was shown in TSST-1 treated human monocytic cell line [51] then acts in an autocrine fashion to further stimulate NF-κB. Both LPS and superantigens have been shown to activate protein tyrosine kinases and protein kinase C. Inhibitors of PKC and PTK down regulated induction of both NF-κB DNA-binding proteins and NF-κB enhancer function 51, 56. An engagement of the LPS ligand and MHC class II molecules by superantigen therefore causes transcriptional activation of the TNF gene and also induces translational activation of TNF mRNA as demonstrated by an increase in polysome loading and in translation rate. These indicate that TNF gene expression is tightly regulated by MHC class II ligands, both at the transcriptional and translational levels [57]. LPS stimulation of macrophages also increases the concentration of a protein that forms Y box-specific complex. Factors interacting with the MHC class II-like “Y box” can additionally modulate the activity of the gene. TNF and MHC class II genes therefore may have common elements involved in their regulation [55].

Both neonatal and adult monocytes and macrophages are equally effective in producing TNF upon stimulation [47]. Although in unstimulated macrophages TNF mRNA is detectable 47, 52 it is maintained in an inactive or translationally dormant form [52]. One of the possible mechanisms of TNF mRNA inactivation in resting monocytes is a rapid truncation of the initially synthesised 3′ polly (A) tail once the TNF mRNA is exported to the cytoplasm [52]. The TNF gene belongs to transiently expressed genes and a cytoplasmic TNF mRNA turnover and its rapid degradation are important points of TNF gene regulation. The rates of mRNA degradation are often influenced by locally produced cytokines such as IFN-γ and IL4. Degradation of the LPS-induced TNF mRNA was found to be increased by IL-4 and by contrast it was significally stabilised by IFN-γ [58]. Studies of murine lymphotoxin-β gene expression and cloning [43] revealed that by contrast to TNF, LT-β is constitutively expressed on lymphoid and hematopoietic tissues. The highest level of LT-β transcription occurs in adult spleen and thymic medulla, suggesting a role for LT-β in the development and function of T cells [43]. LT-α and LT-β have similar patterns of expression in the embryonic and adult thymus and adult spleen. TNF has also been reported to be transcribed in the embryonic thymus [59]. These findings imply that there may be also common transcriptional regulation mechanisms shared by all three genes [43].

Unravelling of the precise molecular mechanisms involved in the regulation of genes that define effectiveness and essential levels of immune responses and inflammatory reactions open wide perspectives for goal-directed correction or substitutive therapy of possible severe consequences conditioned by breaches in this complex chain of gene regulatory events.

Although the main producers of TNF are activated monocytes and macrophages TNF secretion may be induced in a large variety of cell types including lymphocytes, NK cells, polymorphonuclear leukocytes (PNL), keratinocytes, astrocytes and macroglial cells, smooth muscle cells, intestinal paneth cells, tumor cells of various origins, mesangial cells and others 1, 10, 60. A unique source of TNF are mast cells that contain TNF in a stored form in their intracellular granules. Production of TNF by activated T lymphocytes and NK cells may be crucial in allograft rejection where macrophages are less involved primarily [60].

Main points of the pathophisiology of local and systemic inflammatory reactions of an organism are determined by TNF with contribution of IL-1, IL-6 and other proinflammatory cytokines, which synergise with TNF in many respects. TNF exerts profound effects on vascular endothelial cells including their morphological changes, modulation of expression of surface antigens and elaboration of procoagulant activity [1]. An elevated coagulation activity is important for small vessels obstruction and restriction of spread of infection. At the same time TNF promotes transsudation of neutrophils from the intravascular space to the site of inflammation and release of biologically active substances like lysozyme and hydrogen peroxide and degranulation [1]. Important events in local inflammation are the activation and differentiation of monocytes and macrophages themselves induced by TNF and the local leukocyte emigration favoured by vascular endothelial cell expression of adhesion molecules. The systemic inflammatory response is also directed by ubiquitous TNF. TNF influences the metabolism of many cell types favouring the mobilisation of energy stores for use in the immune response. TNF causes suppression of lipoprotein lipase, acetyl Co A carboxilase, and fatty acid synthetase. TNF also activates the release of glycerol from differentiated fat cells [1]. A pyrogenic response characteristic for a generalised inflammatory condition in part may be through a direct effect of TNF on hypothalamic neurones and through induction of IL-1 release 1, 60. TNF is an important modulator of T and B cell function. It induces the expression of additional TNF receptors on primary cultures of T lymphocytes and increases the expression of HLA-DR molecules and high affinity IL-2 receptor. TNF treated T cells show an enhanced proliferative response to IL-2. TNF also enhances IL-2 dependent production of interferon-γ but on the other hand, the B cell proliferation and differentiation occurring in response to the B cell activator PwM are inhibited by TNF [1].

Recent data demonstrate that TNF and both TNFRs (TNFR1 and TNFR2) are involved in the regulation of the TCR-mediated programmed cell death of mature T lymphocytes at least within the CD8+ population. 61, 62, 63. The TNFR1 (TNF-R55) contains near its intracellular C-terminus a region of about 90 amino acids referred to as the ‘death domain’ (DD). Upon clustering, this DD is sufficient for signalling cell death [64]. TNFR2 does not contain DD in its cytoplasmic tail and the cell death induction may have distinct mechanisms [61]. Alterations in lymphocyte apoptosis may contribute to the pathogenesis of a variety of diseases, including cancer, infections, autoimmune diseases, neurodegenerative disorders and AIDS 65, 66, 67, 68.

Through its multiple biologic activities TNF plays a critical role in defensive responses but causes severe damage to host when produced in excess or when the balance between pro- and anti- inflammatory cytokines is broken. Experimental and clinical data show 69, 70, 71 that overexpression of TNF may play a major role in the pathogenesis of inflammatory bowel diseases (IBD) which include Chron’s disease (CD) and ulcerative colitis (UC). Antibodies to TNF ameliorated clinical and histological signs of colitis. No chronic colitis could be induced in TNF knockout mice. Evidence has accumulated that TNF contributes to pathogenesis of rheumatoid arthritis and that neutralisation of TNF by antibodies induced a high degree of clinical benefit 72, 73, 74. IDDM is characterised by a lymphocyte infiltration of the pancreatic islet cells—insulitis that may result in the progressive destruction of the insulin-secreting β cells. Studies on NOD mice showed that the TNF is deeply involved in the pathogenesis of this disease. Permanent neutralisation of TNF by high blood levels of soluble TNF receptor p55-human FcIgG3-fusion molecules protected NOD mice from spontaneous diabetes [75]. On the other hand local expression of TNF in islets protected NOD mice from autoimmune diabetes by preventing the development of autoreactive islet-specific T cells. Furthermore it was shown that a repeated exposure of T cells with TNF in culture suppress the responses of both Th1 and Th2 subsets possibly by attenuating TCR signalling pathways 76, 77. At first sight controversial data regarding beneficial or deteriorative effects of TNF in the development of IDDM, in reality suggest that precise molecular mechanisms of pro- and anti-inflammatory functions of TNF in the complex and multifactorial pathogenesis of IDDM are still to be determined.

TNF is an important and powerful component of the host’s immune response to combat and clear an infection but at the same time may mediate severe damage to tissues of affected organs. Accumulating experimental and clinical data 78, 79, 80, 81 demonstrate that organ damages correlate with increased levels of TNF in serum. Contribution of TNF to neurovascular lesions in cerebral malaria (CM) and early hepatic necrosis in virus-induced liver diseases have been shown. CM and acute mortality can be prevented by anti-TNF antibodies in experimental models [78]. An infection requires a very intensive co-ordinated and effective function of the host immune system to combat the invader. In this regard genetically determined peculiarities or abnormalities in cytokine production and specifically in TNF/TNFR levels may bring hard dissonance in the harmonious orchestra of ubiquitous cytokine regulators of immune responses. Such dissonance may have dramatic consequences to the host and mediate the severe course of disease, subsequent complications and death. Studies on CM and mucocutaneous leishmaneasis (MCL) 82, 83 showed the influence of TNF and LTA genes polymorphism on severity and increased relative risk (RR) for disease development. The particularly rare TNF2 allele and homozygocity for TNFB∗1 allele were associated with increased susceptibility to MCL. Furthermore homozygocity for the TNF2 allele that is a more powerful transcriptional activator than the common allele [30] was associated with a sevenfold increased risk for death from neurovascular complications due to CM in Gambian children.

Section snippets

Summary

The TNF gene cluster is located in the central MHC in close vicinity to other genes responsible for immuneregulation. TNF predetermines in many respects the course of infection and the local inflammatory component present in autoimmune diseases. TNF is involved in the elimination of self-reactive and hyper activated T cells in the periphery by apoptosis. Multiple pro- and anti-inflammatory activities of TNF and related cytokines in the TNF region make them attractive targets along with other

Acknowledgements

I thank Dr. Zulay Layrisse for critical reading and improving of the manuscript and Lubomira Rybak for secretarial assistance.

References (83)

  • B. Crouau-Roy et al.

    Tumor necrosis factor microsatellites in four European populations

    Hum Immunol

    (1993)
  • E.K. Crawford et al.

    The role of 3′ poly (A) tail metabolism in tumor necrosis factor-α regulation

    J Biol Chem

    (1997)
  • B. Wong et al.

    Pathways leading to cell death in T cells

    Current Opinion in Immunology

    (1997)
  • D. Wallach et al.

    Cell death induction by receptors of the TNF familyTowards a molecular understanding

    FEBS Letters

    (1997)
  • P.L. Cohen et al.

    The lpr and gld genes in systemic autoimmunityLife and death in the Fas lane

    Immunol Today

    (1992)
  • B. Derkx et al.

    Tumor-necrosis-factor antibody treatment in Crohn’s disease

    Lancet

    (1993)
  • H.M. Van Dullemen et al.

    Treatment of Crohn’s disease with anti-tumor necrosis factor chimeric monoclonal antibody (cA2)

    Gastroenterology

    (1995)
  • F.M. Brennan et al.

    Cytokines in autoimmunity

    Current Opinion in Immunology

    (1996)
  • D. Kwiatkowski et al.

    TNF concentration in fatal cerebral, non-fatal cerebral, and uncomplicated plasmodium falciparum malaria

    Lancet

    (1990)
  • G. Marinos et al.

    Tumor necrosis factor receptors in patients with chronic hepatitis B virus infection

    Gastroenterology

    (1995)
  • B. Beutler et al.

    The biology of cachectin/TNF-a primary mediator of the host response

    Ann Rev Immunol

    (1989)
  • E.A. Carswell et al.

    An endotoxin-induced serum factor that causes necrosis of tumors (activated macrophage)

    Proc Natl Acad Sci USA

    (1975)
  • M.W. Marino et al.

    Characterisation of tumor necrosis factor-deficient mice

    Proc Natl Acad Sci

    (1997)
  • C.A. Smith et al.

    A receptor for tumor necrosis factor defines an unusual family of cellular and viral proteins

    Science

    (1990)
  • I. Holzinger et al.

    Cloning and genomic characterisation of LST1A new gene in the human TNF region

    Immunogenetics

    (1995)
  • B. Beutler et al.

    Cachectin and tumor necrosis factor as two sides of the same biological coin

    Nature

    (1986)
  • C.F. Ware et al.

    Expression of surface lymphotoxin and tumor necrosis factor on activated T, B, and natural killer cells

    J Immunol

    (1992)
  • R. Lucas et al.

    Crucial role of tumor necrosis factor (TNF) receptor 2 and membrane-bound TNF in experimental cerebral malaria

    Eur J Immunol

    (1997)
  • N.L. Paul et al.

    Lumphotoxin

    Ann Rev Immunol

    (1988)
  • E.Y. Jones et al.

    Structure of tumor necrosis factor

    Nature

    (1989)
  • J.L. Browning et al.

    Characterisation of surface lymphotoxin formsUse of specific monoclonal antibodies and soluble receptors

    J Immunol

    (1995)
  • P.D. Crowe et al.

    A lymphotoxin-β-specific receptor

    Science

    (1994)
  • P. De Tongi et al.

    Abnormal development of peripheral lymphoid organs in mice deficient in lymphotoxin

    Science

    (1994)
  • G. Messer et al.

    Polymorphic structure of the tumor necrosis factor (TNF) locusAn NcoI polymorphism in the first intron of the human TNF-β gene correlates with a variant amino acid in position 26 and a reduced level of TNF-β production

    J Exp Med

    (1991)
  • S. Hashimoto et al.

    Mechanism of a lymphocyte abnormality associated with HLA-B8/DR3Role of interleukin-1

    Clin Exp Immunol

    (1990)
  • G. Candore et al.

    T-cell activation in HLA-B8, DR3-positive individualsEarly antigen expression defect in vitro

    Hum Immunol

    (1994)
  • D. Ulgiati et al.

    Comparative analysis of the disease-associated complement C4 gene from the HLA-A1,B8,DR3 haplotype

    Exper Clin Immunogenet

    (1996)
  • N.J. Makhatadze et al.

    Common Caucasoid HLA-A1B8DR3 linkage group and immune responsiveness in a hybrid Venezuelan population, preliminary data

    Exp Clin Immunogenet

    (1997)
  • A.G. Wilson et al.

    An allelic polymorphism within the human tumor necrosis factor α promoter region is strongly associated with HLA A1,B8, and DR3 alleles

    J Exp Med

    (1993)
  • S. D’Alfonso et al.

    A polymorphic variation in a putative regulation box of the TNFA promoter region

    Immunogenetics

    (1994)
  • S. Hayashi et al.

    Genetic polymorphisms in the 5′-flanking region change transcriptional regulation of the human cytochrome P450IIE1 gene

    J Biochem

    (1991)
  • Cited by (78)

    • Genetic Variations in Tumor Necrosis Factor Alpha, Interleukin-10 Genes, and Migraine Susceptibility

      2011, Pain Medicine
      Citation Excerpt :

      TNF-α and IL-10 have opposing role in the inflammatory response. Furthermore, they also have a synergistic role in the control of the immune-inflammatory responses [8,9]. These cytokines stimulations exert strong control on the Th1/Th2 balance [10], and the perturbation of this balance may affect migraine pathogenesis [3].

    View all citing articles on Scopus
    View full text