FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2000961686> ?p ?o ?g. }

• W2000961686 endingPage "423" @default.
• W2000961686 startingPage "415" @default.
• W2000961686 abstract "By the turn of the millennium, 7% of the earth's population will carry a significant globin gene mutation. These mutations, known collectively as the haemoglobinopathies, are the most prevalent monogenic disorders worldwide. They affect the synthesis of gene products from two multigene loci, the α- and β-globin clusters. The globin loci act in concert to produce differing subtypes of haemoglobin in response to changing oxygen requirements during human development. The most significant of these ‘switches’ in subtype occurs at birth, with the change from fetal (HbF) to adult (HbA) haemoglobin production. At the level of gene transcription this equates to a switch solely in the β-cluster, with silencing of the fetal γ-genes and dramatic up-regulation of adult β-gene transcription ( 75). The fetal/adult switch heralds the onset of the devastating haemoglobinopathies, β-thalassaemia and sickle cell disease (SCD). The β-thalassaemias are characterized by dyserythropoiesis, transfusion dependence, iron overload while in SCD, painful crises and organ infarction predominate. Amelioration of the symptoms of these haemoglobinopathies has been documented in patients with elevated levels of HbF ( 10). In β-thalassaemia elevated levels of γ-chains combine with the redundant α-chains leading to reduced dyserythropoiesis and transfusion requirements. In SCD, HbF is a potent inhibitor of the polymerization of deoxyhaemoglobin S, thus reducing sickling and vascular occlusion ( 11). These observations suggest that treatment strategies aimed at preserving or reactivating fetal haemoglobin production after birth should be explored for these diseases. Elevated levels of fetal haemoglobin are observed in adult humans in two significantly different contexts. The first is in the setting of a diverse group of genetic mutations which are unified phenotypically by persistent synthesis of HbF after birth, known collectively as hereditary persistence of fetal haemoglobin (HPFH) ( 68). The second involves the reactivation of the quiescent fetal genes in adult erythroid progenitors as seen in stress erythropoiesis, drug induction and various other disease states. This paper examines the molecular and cellular mechanisms underlying the production of fetal haemoglobin during normal erythroid development and differentiation, and in the context of HPFH or γ-gene reactivation. In addition, we discuss how these mechanisms could be potentially harnessed as therapeutic strategies for the haemoglobinopathies. The human β-globin locus consists of five genes arranged on chromosome 11 in the order in which they are expressed developmentally (ε, Gγ, Aγ, δ, β). Gene expression is restricted to the erythroid lineage and governed by a precise developmental programme. Embryonic haemoglobin (ε-globin) is synthesized in the yolk sac until week 5 of gestation when γ-globin gene activation occurs in the fetal liver and expression of the ε-gene is silenced. At birth, the second haemoglobin switch occurs when β-globin gene expression commences in the bone marrow with a concomitant silencing of γ-gene expression. This highly specific pattern of gene expression is regulated by cis-acting sequences within the genes themselves, distal sequences which provide enhancer activity and trans-acting factors which bind to these regulatory elements ( 75; 57; 41). The ability to modify fetal globin expression necessitates that we understand the normal mechanisms governing globin gene regulation. To this end, studies in mice transgenic for the human globin genes have provided a number of seminal observations. From these studies it is evident that sequences within or immediately flanking the genes confer the appropriate developmental and tissue specificity of expression. However, expression levels are low in this context and subject to position effects ( 78). Restoration of high-level position-independent expression is observed when a 14 kb segment of DNA 6–20 kb 5′ of the ε-gene is included in transgenic constructs ( 35). This region, characterized by the presence of four erythroid-specific DNAse I hypersensitivity sites (HS1–4), is known as the locus control region (LCR) ( 80; 35). The importance of the LCR is highlighted by the thalassaemia phenotype observed in the Hispanic deletion of HS2–4. These patients produce no γ-, δ- or β-chains despite the fact that the individual globin genes remain intact ( 25). The LCR appears to influence globin gene expression by predominantly acting as a tissue-specific transcriptional activator ( 35). However, roles in insulation of the genes, chromatin opening and locus replication have also been proposed ( 30; 29; 1). The cis-acting sequences in the individual globin genes and the LCR serve as templates for binding of transcription factors. The functions of these proteins are diverse and include transcriptional activation, mediation of the interaction between the LCR and promoters and alteration of chromatin structure ( 58; 4; 43; 76). Many of the actions of these DNA-binding proteins are achieved through the formation of heteromeric complexes between ubiquitous subunits and factors largely confined to the erythroid/haemopoietic compartment ( 40; 44). This ability to interact with multiple partners provides the potential for enormous diversity of function. Many of the factors binding to key globin regulatory sequences have been identified. These include the erythroid-specific proteins NF-E2 and GATA-1 which bind to the LCR HSs and appear critical for high-level globin gene expression ( 48; 5; 55). GATA-1 also binds to the globin promoters where it functions as both a repressor and activator depending on the sequence context and its relationship with other proteins ( 18). Ubiquitous factors such as Sp1, USF, YY1, AP-1 and others also bind to the LCR and promoter sequences ( 57). In the LCR these proteins appear to function as a multiprotein enhanceosome responsible for high-level gene expression ( 86). In the promoters individual factors appear to have diverse functions involving both gene activation and repression. Despite the importance of these factors for globin expression and erythropoiesis, no clear role for any of these proteins in developmental regulation of the β-globin cluster has been established. Transgenic mouse data has defined sequences in the gene promoters as those critical for developmental specificity. Currently only two promoter-binding factors have been identified which appear to influence the development profile of globin gene expression. The first of these, erythroid Kruppel-like factor (EKLF), is a zinc-finger protein with homology to the Kruppel family of proteins ( 54). EKLF binds to the β-promoter CACCC box and is essential for high-level gene expression ( 8). A naturally occurring point mutation at nucleotide position −88 relative to the β-promoter CAP site which ablates EKLF binding results in a β-thalassaemia phenotype ( 59). Similarly, mice nullizygous for the EKLF gene die in utero of a β-thalassaemic-like illness ( 64; 56). Paradoxically, the expression of EKLF is equivalent at all developmental stages, suggesting that the protein is modified in a developmentally specific fashion or that other cofactors or inhibitors restrict its activity to the adult stage of erythropoiesis ( 23). The second developmentally specific protein is the stage selector protein (SSP) ( 43). This protein complex binds to the stage selector element (SSE) in the proximal γ-promoter and is involved in the preferential interaction of that promoter with the LCR ( 43). Phylogenetic footprinting studies demonstrate lack of SSP binding in species which lack a distinct fetal stage of γ-gene expression. These studies also reveal SSP sites in the ε-promoter and HS2 and 3. The SSP consists of a heteromeric complex between the ubiquitous transcription factor CP2 and an as yet unidentified fetal and erythroid-specific partner protein ( 44). Interestingly, NF-E4, the protein responsible for stage specificity in the chicken globin cluster, is the avian homologue of CP2, suggesting that the mechanisms of competitive globin gene regulation are conserved in evolution ( 32). Although the globin genes are depicted as a linear array on chromosome 11, current models of globin regulation indicate that the locus is a dynamic structure, rapidly changing during the switches in β-chain subtype. Recent studies of primary RNA transcripts using intron-specific probes in individual transgenic fetal liver cells have provided insights into the mechanisms governing the transition from fetal to adult globin gene expression ( 86). These experiments indicate that the LCR functions as a holocomplex, activating only a single globin gene at any given time. During the fetal/adult switch the LCR appears to flip-flop back and forth between the γ- and β-genes. As the transcription factor milieu changes to favour adult globin expression, the stability of the γ-gene–LCR interaction decreases and β-globin becomes the predominantly transcribed gene. The importance of the transcription factor environment in mediating these effects is evident in studies of fetal liver cells from EKLF −/− mice transgenic for the human β-globin locus. In cells harvested at the time of the switch, a significant increase is observed in LCR/γ-promoter interactions compared with wild-type mice. This is reflected in an increase in fetal globin expression in the EKLF null mice indicative of a shift in the competitive balance between the two genes for the LCR ( 85; 63). Competition between promoters for enhancers as a mechanism of developmental regulation was initially proposed in the chicken globin system in which repression of the embryonic globin gene in the adult erythroid environment is dependent on competition from the β-gene for the locus enhancer ( 14). In the human locus, competition has been demonstrated in switching models in vitro and in transgenic mice ( 43; 7; 26; 21). In these settings the β-globin gene linked in isolation to the LCR is expressed at all developmental stages. Specificity of β-gene expression is restored only when the γ-gene is also linked in the normal configuration. The polarity of the genes with respect to the LCR appears to be important for gene competition, with the more proximal genes having the advantage ( 37; 22). The trans-acting environment, in particular the stage-specific factors SSP and EKLF, also play a critical role, conferring a competitive advantage on the γ- and β-promoters respectively ( 43; 85; 63). Although the mechanisms by which these factors recruit the LCR have not been established, one model with increasing experimental support suggests that it may be achieved through direct interactions between the stage-specific factors and LCR-binding proteins, and the co-activators of the polymerase complex ( 4, 3). The discovery of naturally occurring mutations associated with the clinical phenotype, hereditary persistence of fetal haemoglobin (HPFH), provided insights into the importance of regulatory sequences in the γ-gene promoter. Two clusters of non-deletional HPFH sites have been identified (Fig 1). The first is centred around the CCAAT box at nucleotide position −114 and the second is centred at −200 relative to the transcription initiation site. A single mutation at −175 has also been documented ( 68). The mechanisms underlying the continued expression of HbF with these mutations are thought to involve alterations to protein-binding motifs, ultimately ablating binding of repressors or enhancing binding of transcriptional activators. Studies of the mutations around −200 suggest that a number of proteins may be involved at each site. Several of the base substitutions in this region alter DNA tertiary structure with loss of formation of an intramolecular triplex in in-vitro assays ( 81). This structural change ablates the binding of a cold shock protein which may function as a repressor at this site ( 38). In addition, the −202 (C → G) mutation also creates a binding site for the SSP. Binding of the SSP to this site differs from the SSE in that it is insensitive to the effects of DNA methylation. This would enable the SSP to bind to the mutated but not wild-type promoter in an adult erythroid environment allowing recruitment of the LCR to the HPFH promoter ( 42). Another HPFH mutation, the −198 (T → C) substitution, enhances the binding of the transcriptional activator, Sp1, providing a third potential mechanism by which the γ-promoter could be activated ( 70). The role of DNA-binding proteins in the CCAAT box and −175 HPFH mutations is less well defined. Although several proteins which bind the distal CCAAT box have been identified, including GATA-1, NF-E3, NF-E6, CP1 and CDP, no clear correlation has emerged between altered binding of these factors and the HPFH phenotype ( 69). Similarly, the −175 (T → C) HPFH mutation alters the binding of GATA-1, but the link between this observation and the increased promoter activity has not been established ( 48; 51). –4) and the major regulatory sequences in the individual promoters (ε, γ and β). The expanded sequence below depicts the common γ-promoter point mutations associated with hereditary persistence of fetal haemoglobin (HPFH). Although unravelling the mechanisms behind these mutations may provide therapeutic strategies for increasing fetal haemoglobin production, it must be remembered that the factors responsible for these effects are acting in the setting of a mutated promoter. However, another group of HPFH patients exist in which no mutations are identifiable within the β-globin cluster, and in many cases the determinant is not linked to the β-locus ( 33; 48; 79). These patients presumably carry a mutation in a trans-acting factor which can alter the developmental profile of an intact locus. More significantly, the effects of this mutation appear to be confined to the globin genes as, apart from increased levels of HbF, the patients are phenotypically normal. Recently, considerable progress has been made in the mapping of these heterocellular HPFH determinants by linkage analysis using DNA polymorphisms. One locus has been mapped to 1.5 Mb on chromosome 6q, another to the X chromosome, and a third to an autosome other than 6 ( 16, 17; 24). The cloning of the trans-factors responsible for the increase in HbF in these patients may offer exciting possibilities for future genetic and pharmacological therapies in SCD and β-thalassaemia. The developmental switch from fetal to adult globin synthesis is effectively completed at 12 weeks post gestation, with HbA representing 97% of the circulating haemoglobin after this time. Despite this, fetal haemoglobin synthesis is still detectable during the maturation process of individual adult erythroid cells. In studies of erythropoietin (Epo) induced erythroid differentiation in vitro, the transition from multipotent progenitor through BFUe, CFUe and proerythroblast phases is accompanied by a switch from γ- to β-globin expression in individual cells ( 15; 28; 61). In these cultures, synthesis of γ-globin occurs predominantly in proerythroblasts and basophilic normoblasts at day 7, whereas β-globin synthesis occurs later, peaking at day 14. As this pattern mimics the developmental profile in a shorter time frame, it is referred to as the ‘compressed switch’. Studies of transcription factor levels in these cells reveal that an increase in GATA-1, RBTN2 and EKLF expression coincides with the transition from γ- to β-globin synthesis ( 45; 19). This is accompanied by a reciprocal fall in GATA-2 levels. The changes in the levels of GATA factors are consistent with those observed during erythroid differentiation in mice and chickens ( 9; 84). However, the enhanced expression of EKLF in this setting provides a link between the globin developmental programme and erythroid maturation. The importance of this observation is that it suggests that the factors which regulate the developmental switch may also influence the pattern of globin gene expression in adult progenitors, the target cells for therapeutic intervention. A subset of the cells within the BFUe show persistent fetal haemoglobin expression and are consequently referred to as F cells ( 82). These cells otherwise display the characteristics of adult cells, including membrane antigen expression (I), red cell enzyme expression (carbonic anhydrase) and size ( 36). Significant variability exists within the normal population as to F-cell number. Determinants of higher levels of F cells may be genetically linked to the β-globin cluster as seen with the GγXmnl (C → T) polymorphism or unlinked to the cluster, as discussed above ( 34; 73; 17). Regulation of F-cell formation appears to be mediated in early erythropoiesis, but the molecular mechanisms governing this process are unknown. Clearly, environmental influences such as cytokine exposure plays an important role ( 74). Of interest from a therapeutic perspective is the mechanism by which F cells and consequently HbF levels are induced in response to stimuli such as erythropoietic stress. One model for this phenomenon is that within BFUe, specific populations of F cells exist which are selectively induced to proliferative with the appropriate environmental stimuli. A second model suggests that cells are reprogrammed in response to environmental influences resulting in an altered ratio of F cells to non-F cells in the colony ( 77). Studies of fetal haemoglobin induction with different therapeutic agents suggests that both these mechanisms may be operational (see below). An increased level of fetal haemoglobin in adult erythroid cells is observed in many divergent human conditions. Unlike the pancellular HPFH syndromes discussed above, these conditions do not represent a failure of completion of the erythroid developmental programme, but rather are due to an alteration in the balance of HbF and HbA expression in adult erythroid precursor cells. Erythropoietic stress as seen in acute blood loss, chemotherapeutic marrow ablation, erythroblastopenia of childhood, severe iron deficiency, acute haemolysis and β-thalassaemia and SCD is associated with increased F-cell production ( 52). At a cellular level, studies suggest that this is due to selective recruitment of erythroid progenitors with the capacity to synthesize fetal globin. At a molecular level, the mechanisms responsible for this increment in fetal haemoglobin remain unknown. Another disease in which striking elevations of HbF are observed is juvenile chronic myeloid leukaemia. The mechanisms underlying fetal globin activation in this setting appear to be different from those active in erythroid stress as the characteristics of the circulating red cells are more fetal, with i antigen expression and low levels of carbonic anhydrase ( 62). Similar findings are occasionally seen in myeloproliferative disorders, germ cell tumours and hepatoma. The ability to enhance fetal globin synthesis in vivo by pharmacological manipulation was initially demonstrated in baboons treated with 5-azacytidine (5-AzaC) ( 20). Subsequent studies confirmed the ability of 5-AzaC to increase HbF in patients with β-thalassaemia and SCD ( 46, 1983). The mechanism of action of 5-AzaC was initially attributed to its ability to demethylate the γ-genes, thus restoring them to the fetal methylation profile ( 83). The hypomethylated promoters were then accessible to the transcriptional activators necessary for gene expression. This was supported by the observation that the binding of the SSP to the proximal γ-promoter was inhibited by DNA methylation ( 42). However, the demonstration of hypomethylated γ-promoters in a patient who had failed to respond to 5-AzaC suggested that additional mechanisms should also be invoked ( 39). The cellular toxicity of 5-AzaC produces erythroid stress reminiscent of acute anaemia. This in turn may result in recruitment of more primitive progenitors committed to a fetal programme. Progenitors cultured in 5-AzaC continue to produce HbF when replated in drug-free media for 12 d, suggesting that the reprogramming model described above is also active ( 39; 31). Thus, it appears that the activity of 5-AzaC is multifactorial, which may explain its superior in-vivo induction of HbF. The use of alternate cytoreducing agents to induce fetal haemoglobin was a direct progression from treatment with 5-AzaC. The most widely studied of these agents is hydroxyurea which has shown significant induction of HbF and a clear clinical benefit in SCD ( 13). The maximal efficacy of hydroxyurea is achieved in the setting of myelotoxicity, suggesting that similar mechanisms to those outlined above may be in operation. However, the inability of hydroxyurea to induce fetal globin in late progenitors in culture indicates that the reprogramming observed with 5-AzaC does not occur ( 53). Interestingly, the level of HbF induction seen with hydroxyurea is enhanced in the presence of Epo, suggesting that once recruitment of F cells has occurred, a proliferative stimulus can further expand this pool ( 2). In addition to the cell-cycle-specific agents, butyrate analogues have also been found to induce fetal globin expression. The clinical use of these compounds stemmed from the observation that elevated levels of α-aminobutyric acid were found in the plasma of infants of diabetic mothers. These infants were noted to have a significant delay in their γ- to β-switch ( 6; 65). Subsequent studies infusing this compound into sheep fetuses reproduced the delayed switch and induction of fetal globin has now been documented in cell lines, transgenic mice and cultured erythroid progenitor cells ( 66, 1989). Unlike the cell cycle agents, stimulation of HbF with butyrate compounds occurs in the absence of erythropoietic stress. Studies of the mechanism of action of these compounds suggest that their effects are primarily transcriptional and may be linked to their ability to inhibit histone deacetylation ( 12; 50). The butyrate-mediated displacement of histones presumably allows access of transcriptional activators to key cis-acting sequencing. γ-promoter truncation studies in cell lines indicate that the distal CCAAT box may be one butyrate response element, although studies in transgenic mice have identified a second element further upstream ( 50; 60). Although proteins binding to the γ-CCAAT box have been extensively studied, no clear candidate protein has emerged as a mediator of butyrate activity. The identification of such a protein could have significant therapeutic implications. Pharmacological induction of HbF and studies of HPFH clearly demonstrate that the stringent developmental programme governing globin gene expression can be altered. Although minimal increases in HbF levels are beneficial in SCD, much higher levels are required in β-thalassaemia. Unfortunately the drugs in current clinical trials are hampered by lack of efficacy in all patients, difficulties with administration and unwanted toxicity. Whilst continued study of these agents singly and in combination in larger trials is imperative, there is a clear need for additional drug discovery programmes and the development of alternate treatment strategies. Recently, high throughput multidrug screening programmes have been established to identify novel inducers of fetal haemoglobin. Several active compounds have been identified in these programmes and these are currently the subject of further study. From the viewpoint of alternate strategies, most effort has been directed into the development of genetic therapy. Successful gene therapy for haemoglobinopathies requires isolation and effective targeting of the haemopoietic stem cell with vectors carrying cis- or trans-acting sequences capable of providing regulated and sustained expression of globin genes. Recent advances have occurred in the areas of stem cell isolation and modification and in vector development ( 71). However, the choice of DNA sequence to be utilized in vectors remains difficult. The studies of HPFH and drug induction of fetal haemoglobin suggest that gene therapy approaches using trans-acting factor genes are worth pursuing. This is particularly germane in light of the difficulties experienced in the use of cis-acting sequences in this context ( 72). A trans-factor based approach may simplify the transcriptional machinery required for transgene expression. Specificity of trans-factor activity is presumably achievable, as evidenced by the unlinked HPFH patients who are otherwise phenotypically unremarkable ( 16, 17; 24). Identification of the causative genes in these kindreds and reproduction of the factor perturbation using a gene therapy approach may offer an exciting treatment alternative. The use of genes involved in HPFH in gene therapy raises significant questions about the timing of treatment, as it may be easier to maintain fetal globin gene expression than it is to reactivate it. It has been shown previously that after the γ- to β-switch the γ-promoters become methylated ( 27). Thus chromatin effects may block the efficacy of proteins which are capable of maintaining fetal globin expression, but not reactivating it. Interestingly, in studies administering sodium butyrate to ovine fetuses the most profound effects were observed when treatment was instituted while fetal hemoglobin levels were >85% of total globin. Little if any alteration in the chronology of switching was observed when infusions were begun after adult globin levels had exceeded 40% ( 67). Thus the proteins which mediate the butyrate response may be far more effective at inducing HbF in the context of a locus with open chromatin. As the likely target of gene therapy vectors will be adult haemopoietic stem cells in which the developmental switch is complete, factors capable of influencing the compressed switch may offer better therapeutic alternatives. As yet, no proteins which can induce γ-gene expression in this context have been identified. However, the observed up-regulation of EKLF in progenitors during the γ- and β-gene compressed switch suggests that the major developmental proteins may also be involved in the changing globin profile during differentiation. Based on this observation, gene therapy approaches using the fetal component of the SSP (when cloned) or dominant negative mutants of EKLF should be pursued. The ability to perturb ontogeny and prevent or reverse the fetal/adult switch in globin gene subtype offers therapeutic hope in the haemoglobinopathies. Study of normal development globin gene expression, HPFH and pharmacological induction of HbF all provide insights into how this may be achieved. In the next decade, drug development programmes and the use of trans-acting genes in gene therapy will be expanded, utilizing models of haemoglobin switching. Complementary to this will be the search for additional factors which influence the temporal profile of the globin genes and further study of gene transfer efficiency, control of transgene expression, and stem cell expansion and transplantation. Advances in all these areas will bring the reactivation of fetal haemoglobin closer as a therapeutic modality in SCD and β-thalassaemia. This work was supported in part by The Wellcome Trust, NHMRC project grant 940573, Cancer Centre Support CORE Grant P30 CA 21765, NHLBI Program Project Grant PO1 HL 53749-01, American Lebanese Syrian Associated Charities (ALSAC) and the Assisi Foundation of Memphis." @default.
• W2000961686 created "2016-06-24" @default.
• W2000961686 creator A5059629859 @default.
• W2000961686 creator A5075314060 @default.
• W2000961686 date "1998-07-01" @default.
• W2000961686 modified "2023-09-26" @default.
• W2000961686 title "UNDERSTANDING FETAL GLOBIN GENE EXPRESSION: A STEP TOWARDS EFFECTIVE HbF REACTIVATION IN HAEMOGLOBINOPATHIES" @default.
• W2000961686 cites W108640602 @default.
• W2000961686 cites W1479457498 @default.
• W2000961686 cites W1505715400 @default.
• W2000961686 cites W1508246134 @default.
• W2000961686 cites W1534790099 @default.
• W2000961686 cites W1543772844 @default.
• W2000961686 cites W15754739 @default.
• W2000961686 cites W1581617558 @default.
• W2000961686 cites W1597956418 @default.
• W2000961686 cites W1628302795 @default.
• W2000961686 cites W1650227569 @default.
• W2000961686 cites W170079721 @default.
• W2000961686 cites W1763021161 @default.
• W2000961686 cites W176375325 @default.
• W2000961686 cites W182584870 @default.
• W2000961686 cites W1868405784 @default.
• W2000961686 cites W1966313798 @default.
• W2000961686 cites W1970103367 @default.
• W2000961686 cites W1973273611 @default.
• W2000961686 cites W1973595140 @default.
• W2000961686 cites W1979295881 @default.
• W2000961686 cites W1993532214 @default.
• W2000961686 cites W1996043889 @default.
• W2000961686 cites W1998607972 @default.
• W2000961686 cites W2003918839 @default.
• W2000961686 cites W2011907865 @default.
• W2000961686 cites W2015569707 @default.
• W2000961686 cites W2015980850 @default.
• W2000961686 cites W2016197401 @default.
• W2000961686 cites W2018637153 @default.
• W2000961686 cites W2024793020 @default.
• W2000961686 cites W2028553942 @default.
• W2000961686 cites W2028693053 @default.
• W2000961686 cites W2033858458 @default.
• W2000961686 cites W2048437870 @default.
• W2000961686 cites W2048441226 @default.
• W2000961686 cites W2050987515 @default.
• W2000961686 cites W2054254765 @default.
• W2000961686 cites W2055643269 @default.
• W2000961686 cites W2056046551 @default.
• W2000961686 cites W2057321580 @default.
• W2000961686 cites W206006204 @default.
• W2000961686 cites W2062031128 @default.
• W2000961686 cites W2067755233 @default.
• W2000961686 cites W2071000969 @default.
• W2000961686 cites W2071126171 @default.
• W2000961686 cites W2074378056 @default.
• W2000961686 cites W2081103358 @default.
• W2000961686 cites W2081723155 @default.
• W2000961686 cites W2084364631 @default.
• W2000961686 cites W2085631191 @default.
• W2000961686 cites W2091399084 @default.
• W2000961686 cites W2096998693 @default.
• W2000961686 cites W2102707713 @default.
• W2000961686 cites W2107270084 @default.
• W2000961686 cites W2114799987 @default.
• W2000961686 cites W2118299921 @default.
• W2000961686 cites W2120959775 @default.
• W2000961686 cites W2133279102 @default.
• W2000961686 cites W2141609688 @default.
• W2000961686 cites W2142071465 @default.
• W2000961686 cites W2147951044 @default.
• W2000961686 cites W2153569552 @default.
• W2000961686 cites W2154187040 @default.
• W2000961686 cites W2158053354 @default.
• W2000961686 cites W216922957 @default.
• W2000961686 cites W2264211064 @default.
• W2000961686 cites W2313541759 @default.
• W2000961686 cites W2318281771 @default.
• W2000961686 cites W2326476344 @default.
• W2000961686 cites W2341675267 @default.
• W2000961686 cites W2387570970 @default.
• W2000961686 cites W2424409471 @default.
• W2000961686 cites W272815522 @default.
• W2000961686 cites W4235165503 @default.
• W2000961686 cites W4235713981 @default.
• W2000961686 cites W4252346318 @default.
• W2000961686 cites W58134837 @default.
• W2000961686 cites W73366333 @default.
• W2000961686 doi "https://doi.org/10.1046/j.1365-2141.1998.00811.x" @default.
• W2000961686 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9695955" @default.
• W2000961686 hasPublicationYear "1998" @default.
• W2000961686 type Work @default.
• W2000961686 sameAs 2000961686 @default.
• W2000961686 citedByCount "38" @default.
• W2000961686 countsByYear W20009616862013 @default.
• W2000961686 countsByYear W20009616862014 @default.
• W2000961686 countsByYear W20009616862020 @default.
• W2000961686 countsByYear W20009616862023 @default.
• W2000961686 crossrefType "journal-article" @default.
• W2000961686 hasAuthorship W2000961686A5059629859 @default.

• next