Single-chain antibody and its derivatives directed against vascular endothelial growth factor: application for antiangiogenic gene therapy.

Single-chain antibodies (scFv) have an enormous potential for clinical application. However, rapid blood clearance and difficulties in large-scale production of active scFvs have limited the practical use of these antibody fragments. Recently, an anti-vascular endothelial growth factor (VEGF) scFv (scFv V65) was selected in our laboratory from a human antibody phage-display library. This antibody was able to reduce tumor growth in mice by approximately 50%. Here, we employ a gene therapy strategy for sustained in vivo expression of scFv V65 and its derivatives. scFv fusion proteins containing parts of the constant IgG1 region were generated (minibody and scFv V65-Fc) to increase the serum half-life of the scFv V65. Systemic administration of recombinant adenovirus encoding scFv V65 resulted in substantial tumor inhibition. This effect could be improved by multiple virus injections. We found that the efficacy of different scFv V65 formats was dependent on the mode of administration: whereas scFv V65-Fc was the most efficient when expressed locally, scFv V65 was superior when delivered systemically. Our results show that therapeutic levels of scFv V65 can be obtained by systemic injection of recombinant adenoviruses. Therefore, therapeutic gene delivery of scFv is a feasible strategy that overcomes several limitations of conventional antibody therapy.

Signaling through the T1/ST2 molecule is not necessary for Th2 differentiation but is important for the regulation of type 1 responses in nonhealing Leishmania major infection.

T1/ST2 is a stable cell surface marker selectively expressed on type 2 T helper (Th2) effector cells. Since nonhealing Leishmania major infections in susceptible BALB/c mice have been ascribed to a polarized Th2 response, we used an anti-T1/ST2 monoclonal antibody (MAb) or a T1-Fc fusion protein to investigate the role of CD4+ T1/ST2(+) Th2 cells in experimental leishmaniasis. We show that interfering with T1/ST2 signaling had no effect on lesion development or parasite replication; however, it induced a significantly higher type 1 response and an enhanced capacity of CD4+ T cells to respond to interleukin 12 (IL-12). Surprisingly, even in the presence of an elevated Th1 response, the production of antigen-specific type 2 cytokines was not altered in the group of mice treated with the anti-T1/ST2 MAb or the T1-Fc fusion protein. To characterize further this Th2 response, we assessed the cytokine profile of CD4+ T cells and found that interfering with T1/ST2 signaling did not alter the cytokine profile of CD4+ T1/ST2(+) T cells. These results show that T1/ST2 signaling is not necessary for the differentiation of naive CD4+ T cells into antigen-specific CD4+ T1/ST2(+) Th2 cells. In addition to CD4+ T1/ST2(+) T cells, we detected another subpopulation of CD4+ Th2 cells, negative for the expression of T1/ST2, that could differentiate in vivo in response to L. major infection. Taken together, our results suggest that CD4+ T1/ST2(+) Th2 cells but not CD4+ T1/ST2(-) Th2 cells can downregulate the Th1 response during the course of a nonhealing L. major infection through a mechanism that is independent of IL-4 or IL-10.

Cytotoxic targeting of F9 teratocarcinoma tumours with anti-ED-B fibronectin scFv antibody modified liposomes.

We prepared small unilamellar liposomes derivatised with single chain antibody fragments specific for the ED-B domain of B-fibronectin. This extracellular matrix associated protein is expressed around newly forming blood vessels in the vicinity of many types of tumours. The single chain antibody fragments were functionalised by introduction of C-terminal cysteines and linked to liposomes via maleimide groups located at the terminal ends of poly(ethylene glycol) modified phospholipids. The properties of these anti-ED-B single chain antibody fragments-liposomes were analysed in vitro on ED-B fibronectin expressing Caco-2 cells and in vivo by studying their biodistribution and their therapeutic potential in mice bearing subcutanous F9 teratocarcinoma tumours. Radioactively labelled ((114m)Indium) single chain antibody fragments-liposomes accumulated in the tumours at 2-3-fold higher concentrations during the first 2 h after i.v. injection compared to unmodified liposomes. After 6-24 h both liposome types were found in similar amounts (8-10% injected dose g(-1)) in the tumours. Animals treated i.v. with single chain antibody fragments-liposomes containing the new cytotoxic agent 2'-deoxy-5-fluorouridylyl-N(4)-octadecyl-1-beta-D-arabinofuranosylcytosine (30 mg kg(-1) per dose, five times every 24 h) showed a reduction of tumour growth by 62-90% determined on days 5 and 8, respectively, compared to animals receiving control liposomes. Histological analysis revealed a marked reduction of F9 tumour cells and excessive deposition of fibronectin in the extracellular matrix after treatment with single chain antibody fragments-2-dioxy-5-fluorouridylyl-N(4)-octadecyl-1-beta-D-arabinofuranosylcytosine-liposomes. Single chain antibody fragments-liposomes targeted to ED-B fibronectin positive tumours therefore represent a promising and versatile novel drug delivery system for the treatment of tumours.

Signal-sequence trap in mammalian and yeast cells: a comparison.

Membrane associated and secreted proteins are translated as precursors containing a signal peptide that allows protein-insertion into the membrane of the endoplasmic reticulum and is co-translationally removed in the lumen. The ability of the signal peptide to direct a polypeptide into the secretory pathway is exploited in methods developed to select cDNAs encoding such proteins. Different strategies are known in which cDNA libraries can be screened for signal peptides by the ability of the latter to rescue the translocation of signal sequence-less proteins. In one method, a cDNA library is tested for interleukin 2 receptor alpha chain translocation to the membrane in COS cells, in another one for invertase secretion from yeast. In this work, we compared the two systems by testing six mouse signal peptides in COS and yeast cells. All of them were functional in the mammalian system, whereas only three of them in yeast. Two other sequences needed the 5' cDNA sequence flanking the ATG codon to be removed in order to enable protein translocation. Although the structure of signal sequences and the functioning of the secretory machinery are well conserved from prokaryotes to eukaryotes, it seems evident that not all signal peptides can be interchanged between different proteins and organisms. In particular, signal peptides that are functional in the mammalian system do not necessarily lead to protein translocation in yeast.

Regulation of ST2L expression on T helper (Th) type 2 cells.

T1/ST2L, an IL-1 receptor homologue, is selectively expressed on murine Th2 cells and specific anti-ST2L antibodies can profoundly modulate the Th1/Th2 balance in vivo. Naive CD4+ T cells do not express ST2L but do so on activation with specific antigen in the presence of IL-4 or when stimulated with low doses of antigen in the absence of exogenously added IL-4. Similarly enhanced ST2L expression occurred after stimulation of Th2 cells with antigen or the mitogen ConA in the presence of APC. Restimulation of Th2 cells in the presence of IFN-gamma led to a decreased expression of ST2L to below basal levels. Conversely, Th2 cells cultured with IL-4 led to increased ST2L expression. The reduced expression of ST2L in response to high doses of antigen is also reversed by the neutralization of IFN-gamma. Using an ST2L promoter/luciferase reporter gene construct, we show that the distal but not proximal ST2L promoter is responsible for specific gene expression in Th2 cells. IL-4 enhances, whereas IFN-gamma suppresses ST2L expression via direct modulation of the distal promoter of the ST2L gene. These data provide a mechanistic explanation for the selective expression of ST2L on Th2 cells.

Production of functionalized single-chain Fv antibody fragments binding to the ED-B domain of the B-isoform of fibronectin in Pichia pastoris.

The Pichia pastoris expression system was used to produce functionalized single-chain antibody fragments (scFv) directed against the ED-B domain of the B-fibronectin (B-Fn) isoform which was found to be present only in newly formed blood vessels during tumor angiogenesis. Therefore, scFv antibody fragments recognizing the ED-B domain are potential markers for angiogenesis. We constructed four functionalized scFv antibody fragments for direct labeling with radioactive molecules or toxins or for attachment to liposomes serving as carriers for cytotoxic or antiangiogenic compounds. The C-termini of the scFv antibody fragments contain 1-3 cysteine residues that are separated by a hydrophilic linker (GGSSGGSSGS) from the binding domain and are accessible for site-specific functionalization with thiol-reactive reagents. Plasmid expression, culture conditions, and purification were optimized in 1-L cultures. The scFv antibody fragments were purified by anion exchange chromatography. The yields were 5-20 mg/L culture medium. The large-scale production of one scFv antibody fragment in a 3.7-L fermenter gave a yield of 60 mg. The reactivity of the cyteines was demonstrated by labeling with the thiol-reactive fluorescent dye ABD-F. The four scFv antibody fragments bound specifically to ED-B-modified Sepharose and binding was further confirmed by immunofluorescence on cell cultures using ED-B-positive human Caco-2 tumor cells. Furthermore, we could demonstrate specific binding of scFv-modified liposomes to ED-B-positive tumor cells. Our results indicate that the P. pastoris expression system is useful for the large-scale production of cysteine-functionalized alpha-ED-B scFv antibody fragments.

The ALG-2/AIP-complex, a modulator at the interface between cell proliferation and cell death? A hypothesis.

During the development of an organism cell proliferation, differentiation and cell death are tightly balanced, and are controlled by a number of different regulators. Alterations in this balance are often observed in a variety of human diseases. The role of Ca(2+) as one of the key regulators of the cell is discussed with respect to two recently discovered proteins, ALG-2 and AIP, of which the former is a Ca(2+)-binding protein, and the latter is substrate to various kinases. The two proteins interact with each other in a Ca(2+)-dependent manner, and the role of the complex ALG-2/AIP as a possible modulator at the interface between cell proliferation and cell death is discussed.

Increased expression of H/T-cadherin in tumor-penetrating blood vessels.

Neovascularization is a prerequisite for tumor growth. Thus, selective destruction of the tumor vasculature should prevent tumor expansion. We have established a method to identify proteins that are specifically expressed on the surface of endothelial cells in tumors. CD31-positive endothelial cells were isolated from Lewis lung carcinoma lung metastases as well as from normal lung tissue. cDNAs derived from these cells were subjected to a subtractive hybridization procedure, and cDNAs overrepresented in tumor-derived endothelial cells were isolated; those encoding surface proteins were selected using a signal sequence trap assay. One isolated cDNA encoded H/T-cadherin. In this report, we show that mouse H/T-cadherin is overexpressed on endothelial cells of several tumors, whereas it is expressed only on a subset of endothelial cells in healthy organs. On the basis of the expression of H/T-cadherin in lung metastases of different tumors, we suggest that different tumors can have a differential influence on the expression of endothelial cell surface proteins.

Inhibition of tumor angiogenesis by a single-chain antibody directed against vascular endothelial growth factor.

Monoclonal antibody (Ab) directed against the vascular endothelial growth factor, one of the major inducers of angiogenesis, can inhibit tumor growth in mice. Treatment of cancer patients with monoclonal Ab requires large-scale production of the clean Ab and frequent application of the Ab. This might be improved by using single-chain Ab fragments (scFvs), which can be produced in large quantities in bacteria and are attractive for gene therapeutic approaches. Here we describe anti-vascular endothelial growth factor scFvs derived from a human phage-display library able to block the vascularization of the chorioallantoic membrane of chick embryos and reduce the growth of s.c. tumors in nude mice. This work opens the way to develop gene therapy-based strategies using a scFv to treat angiogenesis-dependent diseases.

T1-deficient and T1-Fc-transgenic mice develop a normal protective Th2-type immune response following infection with Nippostrongylus brasiliensis.

The IL-1 receptor-related protein T1 is expressed on the surface of Th2, but not Th1 cells. Studies with anti-T1 monoclonal antibodies have suggested that T1 is critical for development of normal Th2-type responses. To elucidate the role of T1 in vivo, we generated T1-deficient mice and a T1-transgenic strain which secretes soluble T1-Fc fusion protein into the serum. These were analyzed for the Th2 immune response induced by infection with the parasitic nematode Nippostrongylus brasiliensis. Although Th2 cytokine production by lymph node cells was similar in all groups of N. brasiliensis-infected mice, a decrease in IL-5 production by lung lymphocytes was detected in both T1-deficient and T1-Fc-transgenic mice compared to control littermates. This difference in IL-5 production did not influence blood eosinophilia, but recruitment of eosinophils into lung tissue, especially in T1-Fc-transgenic mice was slightly decreased. However, induction of all other immune parameters was normal and both T1-deficient and T1-Fc-transgenic mice were able to clear the parasite infection within 12 days with kinetics similar to those in control mice. Therefore, in contrast to previous suggestions, we conclude that the T1 protein is not obligatory for normal development of Th2 immune responses.

Attenuated expression of the serum responsive T1 gene in ras transformed fibroblasts due to the inhibition of c-fos gene activity.

The T1 gene encodes a protein, which shares homology with the IL-1 receptors. In fibroblasts, T1 is induced by growth factors and in response to the onset of oncogene expression. The c-fos gene is transiently activated in these situations and was shown to be the major mediator of T1 gene induction. In contrast, the sustained expression of a ras oncogene in NIH3T3 cells resulted in the downregulation of basal T1 gene activity and the attenuation of T1 gene induction in response to mitogenic signals. Likewise, the immediate early genes encoding c-Fos, FosB, and Fra-2 are repressed in these cells. T1 gene repression could be overcome by the forced expression of c-fos in ras transformed fibroblasts. Thus, the lack of c-fos gene expression is the likely cause for ras mediated T1 gene repression. Fra-1, in contrast to the other three members of the Fos family, is permanently synthesized in high amounts in ras transformed NIH3T3 fibroblasts. We show that AP-1, which is abundant in these cells throughout the whole cell cycle, consists predominantly of Fra-1/c-Jun and Fra1/JunD heterodimers. We provide evidence that Fra1/c-Jun heterodimers are responsible for the repression of c-fos gene induction following serum stimulation. The introduction of a dominant negative version of c-Jun into ras transformed fibroblasts was able to rescue c-fos gene induction in response to serum stimulation, further demonstrating that AP-1 is indeed involved in c-fos gene repression. We conclude that oncogenic ras mediates the activation of the fra-1 gene which results in elevated AP-1 activity throughout the cell cycle. Fra-1 containing AP-1 complexes repress the c-fos and possibly other immediate early genes thereby preventing the induction of certain delayed early genes such as the T1 gene in response to mitogenic stimulation.

The IL-1 receptor-related T1 antigen is expressed on immature and mature mast cells and on fetal blood mast cell progenitors.

Expression of the T1 gene, also known as ST2, DER4, and Fit-1, has been shown to be associated with cell proliferation. It gives rise to two different mRNAs that encode a receptor-like protein and a soluble molecule representing the ectodomain of the receptor form. Although T1 is a member of the IL-1R family, its biologic function is currently unknown. In this study, we have analyzed the expression of the T1 surface Ag in murine hemopoietic organs. Mast cells (MCs) were shown to be the only identifiable cell lineage that expressed T1 at high levels. T1 expression was found on cultured bone marrow-derived immature MCs. Similarly, freshly isolated connective tissue-type MCs from the i.p. cavity were also shown to express high levels of T1. Interestingly, the earliest detectable committed MC precursor isolated from fetal blood (FB) at day 15.5 of gestation, but not circulating hemopoietic stem cells in FB, also expresses high level of T1. Since FB promastocytes lack expression of the high affinity IgE receptor (Fc epsilonRI), T1 expression precedes expression of Fc epsilonRI in MC ontogeny. The finding that the T1 Ag is selectively expressed at several stages during development of the MC lineage suggests that this cell surface molecule, in combination with the well-established markers c-Kit and Fc epsilonRI, should be valuable for studying the MC lineage.

GATA-Dependent expression of the interleukin-1 receptor-related T1 gene in mast cells.

The murine delayed-early serum-responsive gene T1 encodes glycoproteins of the interleukin-1 receptor family. Transcriptional initiation in fibroblasts is regulated by c-Fos and gives rise to a rare 5-kb mRNA and an abundant 2.7-kb mRNA. These transcripts are translated into a receptor-like membrane-anchored protein and a secreted protein consisting only of the ectodomain. In mast cells, T1 gene transcription is initiated 10.5 kb further upstream than in fibroblasts and gives rise predominantly to the 5-kb transcript under normal growth conditions. Here we demonstrate that calcium ionophore stimulation of mast cells resulted in an upregulation of T1 gene expression and a switch from the long to the short T1 transcript. This was paralleled by the disappearance of the receptor-type T1 protein on the mast cell surface and the secretion of large amounts of the truncated T1 protein. c-Fos and a T1 enhancer, which have previously been identified to be essential for T1 expression in fibroblasts, were not required for calcium ionophore-mediated T1 gene upregulation. Overexpression of the transcription factor GATA-1 in mast cells caused elevated T1 synthesis. Three GATA elements were identified in the minimal GATA-responsive mast cell promoter. Mutational analysis revealed that all three GATA elements are involved in T1 gene expression. Point mutations within the middle GATA element eliminated promoter activity completely, while mutations of the distal and proximal GATA binding sites reduced promoter strength by factors of 2 and 5, respectively. Exogenous expression of GATA-1 was not sufficient to activate the mast cell-specific promoter in NIH 3T3 fibroblasts.

Expression analysis of the soluble and membrane-associated forms of the interleukin-1 receptor-related T1 protein in primary mast cells and fibroblasts.

The murine T1 gene encodes a membrane-bound glycoprotein (T1M) and a soluble variant (T1S) which represents the ectodomain of the receptor-type form. T1 is an orphan receptor belonging to the interleukin-1 receptor family. Its biological function is currently unknown. We analyze the expression of the two T1 proteins in mast cells and fibroblasts by using a set of monoclonal antibodies (MAb) that specifically recognize the extracellular portion of the T1 receptor. To generate anti-T1 MAbs, we immunized Lewis rats with a eukaryotically expressed chimeric protein consisting of the T1-receptor ectodomain fused to a human immunoglobulin domain. The two MAbs DJ4 and DJ8 were shown to specifically detect the murine T1M protein on the surface of primary IL-3-dependent bone marrow-derived mast cells as shown by flow cytometry and immunohistochemistry. Both antibodies were also capable of immunoprecipitating the membrane-associated 110-120 kDa T1M protein from mast cell lysates. In serum-stimulated but not in quiescent NIH3T3 fibroblasts, DJ4 and DJ8 MAbs detected both the soluble T1S protein as a 45-65 kDa band on SDS polyacrylamide gels as well as the membrane-bound 95 kDa T1M protein. The T1M protein in fibroblasts was less abundantly expressed and exhibited a lower molecular weight than the mast cell-produced T1M, probably as a consequence of different protein glycosylation. The MAbs described here represent highly specific reagents and valuable tools that should facilitate the establishment of the murine T1 protein expression pattern thus contributing to the solution of the question of its function.

Effects of ras transformation on the induction of the IL-1 receptor related T1 gene in response to mitogens, anisomycin, IL-1 and TNFalpha.

The T1 gene gives rise to two transcripts encoding a 62 kDa membrane-bound and a 37 kDa secreted protein with similarity to the type I IL-1 receptor. It is weakly expressed in proliferating but not in resting fibroblasts and is strongly induced during the entry of quiescent cells into the cell cycle. Here we show that the T1 gene is also transcriptionally activated in response to the treatment of fibroblasts with cycloheximide and anisomycin. These protein synthesis inhibitors are known to stimulate the JNK and p38/RK signal transduction pathways. We provide evidence that anisomycin triggers T1 gene induction through the stimulation of the p38/RK MAP kinase. This observation is in line with our finding that physiological activators of the p38/RK pathway, the proinflammatory cytokines IL-1 and TNFalpha, stimulate T1 gene expression efficiently. Growth factor mediated T1 gene induction is a delayed early event, requiring ongoing protein synthesis. In contrast, anisomycin induces T1 gene expression at concentrations which block translation completely. Thus, transcriptional induction of the T1 gene via the p38/RK pathway is an immediate early event not requiring de novo protein synthesis. The T1 gene is strongly induced by various mitogens in quiescent NIH3T3 fibroblasts but not in ras transformed NIH3T3 cells. In contrast, all of the three tested agent which activate the p38/RK pathway, IL-1, TNFalpha, and anisomycin led to strong T1 gene expression in normal and ras transformed NIH3T3 cells alike. Thus, the T1 gene can be induced through the activation of at least two MAP kinase pathways: signaling through the ERK pathway can occcur in normal but not in ras transformed NIH3T3 cells, whereas the signaling through the p38/RK pathway is not affected by ras transformation.

Abundance and location of the small heat shock proteins HSP25 and alphaB-crystallin in rat and human heart.

BACKGROUND: In the heart, there are high constitutive levels of the two related small heat shock proteins, HSP25 and alphaB-crystallin. To gain insight into their functional role, we have analyzed abundance and location of both proteins in rat and human hearts at different stages of development and in diseased state. METHODS AND RESULTS: Immunoblotting analysis of rat ventricular tissue at fetal, neonatal, and adult stages reveals the level of HSP25 to decline strongly during development, whereas the level of alphaB-crystallin remains nearly constant. In parallel, the portion of phosphorylated isoforms of HSP25 decreases as shown by two-dimensional polyacrylamide gel electrophoresis. HSP25 is detected in cardiomyocytes and endothelial and vascular smooth muscle cells, whereas alphaB-crystallin is detected in cardiomyocytes only by immunofluorescence and immunoelectron microscopy. Both proteins colocalize in the I-band and M-line region of myofibrils in cardiomyocytes. In diseased and transplanted adult human hearts, HSP25 and alphaB-crystallin levels are considerably elevated compared with fetal hearts. In failing adult human hearts, phosphorylated isoforms of HSP25 predominate, and cardiomyocytes with a partial dislocation of HSP25 and alphaB-crystallin are observed. CONCLUSIONS: Differential accumulation and location of HSP25 and alphaB-crystallin in heart tissue during development imply distinct functions of both proteins, which seem to be involved in organization of cytoskeletal structures. As judged by level, phosphorylation state, and location of both small heat shock proteins, diseased adult human hearts share features with fetal hearts.

Cooperative interactions of AP-1 and basic helix-loop-helix transcription factors regulate T1 gene expression.

The T1 gene is a murine, delayed early serum-responsive gene that encodes glycoproteins of the interleukin-1 receptor (IL-1R) family. Transcription of the T1 gene leads to production of two mRNAs that encode a transmembrane protein, which is highly similar to the type-1 IL-1R, and a secreted protein, which consists solely of the extracellular part. Fibroblasts, in contrast to mast cells, express predominantly the shorter form of the protein, and several mitogens cause strong, transient induction of the T1 gene in these cells. Here we describe the identification of a 148 bp enhancer element that is positioned 3.6 kb upstream of the transcription initiation site. A TPA-responsive element (TRE) and three identical E-boxes are located within this sequence. Introduced point mutations confirmed the necessity of these sites for full T1 promoter activity. The TRE and the distal E-box are absolutely indispensable for promoter function, whereas the two proximal E-boxes contribute less to promoter strength. In vitro the three E-boxes are bound by different protein complexes.

A hepatitis B virus mutant with a new hepatocyte nuclear factor 1 binding site emerging in transplant-transmitted fulminant hepatitis B.

Hepatitis B virus (HBV) DNA was cloned from serum of a heart transplant recipient who died from fulminant hepatitis B transmitted by the donor. Restriction enzyme analyses of the clones obtained by conventional cloning yielded six HBV variants: a major species (pF-1) representing 88% and five minor species (pF-2 to pF-6), each representing 2% to 4% of the clones. The complete nucleotide sequence of these six variants revealed that five of the six viral genomes, including pF-1, carried a novel 11 base pair (bp) insertion in the core promoter region as well as an 18 bp and an 108 bp in-frame deletion in the pre-S1 region not present in the donor. One genome was identical to the sequence of the donor. Functional analyses of HBV clones generated by in vitro mutagenesis and cassette exchange showed that the 11 bp insertion is a strong binding site for hepatocyte nuclear factor 1 (HNF-1). In transient transfection experiments, the novel HNF-1 sequence motif was shown to result in enhanced viral replication. Immunohistochemical analyses revealed high levels of cytoplasmic and nuclear hepatitis B core antigen (HBcAg) and only scattered hepatitis B surface antigen (HBsAg) expression in the liver. The data in our immunosuppressed patient showed that HBV variants can rapidly accumulate in severe hepatitis B and suggest that the novel HNF-1 binding site may have contributed to the fulminant clinical course, possibly via enhanced viral replication.

Examination of the molecular basis for the lack of alphaB-crystallin expression in L929 cells.

We have previously shown that murine L929 cells do not express the small heat shock protein alphaB-crystallin upon exposure to thermal stress (Mol Cell Biochem 155: 51-60, 1996). In these studies, we demonstrate that L929 cells also fail to express alphaB-crystallin upon exposure dexamethasone, whereas NIH 3T3 and Swiss 3T3 murine cells exhibit alphaB-crystallin expression under identical conditions. Mobility shift assays demonstrated heat-inducible binding, presumably by heat shock factor(s), to an alphaB-crystallin heat shock element (HSE) oligomeric sequence in total cellular extracts from L929 cells. Transient transfection of a plasmid containing the alphaB-crystallin promoter linked to a CAT reporter gene exhibited heat-inducible expression in L929 cells. In addition, L929 cells stably transfected with a plasmid containing the complete alphaB-crystallin gene showed expression of this gene following heat shock. The presence of the endogenous alphaB-crystallin gene was detected by Southern blot hybridization of genomic L929 DNA, and sequence analysis revealed identical nucleotide structure to published murine sequences throughout the entire promoter. Treatment of L929 cells with 5-azacytidine enabled heat-inducible expression of alphaB-crystallin from the endogenous gene, however, methylation of the putative heat shock element (HSE) and flanking promoter sequences of L929 cell genomic DNA was not detected. In vivo genomic footprinting demonstrated constitutive binding to the endogenous HSE of the alphaB-crystallin promoter in L929, L929/alphaB-crystallin transfectant cells, and Swiss 3T3 cells during unstressed and heat stressed conditions. Therefore, the genomic alphaB-crystallin HSE region in L929 cells appears to be available for binding of putative transcription factors, but methylation in other regions of the gene or genome repress the expression of alphaB-crystallin in L929 cells. In vitro culture of L929 cells appears to have rendered the alphaB-crystallin gene loci inactive through methylation, thus providing a unique system by which to study the function of transfected small heat shock proteins.

Glucocorticoids regulate the expression of the stressprotein alpha B-crystallin.

alpha B-crystallin is a major component of the eye lens but is also found in many extralenticular tissues. In established fibroblasts it is synthesized in response to stress such as hyperthermia. Here we report that the treatment of NIH3T3 fibroblasts with the synthetic glucocorticoid hormone dexamethasone resulted in the accumulation of substantial amounts of alpha B-crystallin, alpha B-crystallin mRNA accumulated slowly and over a period of many days in response to prolonged hormone treatment. alpha B-crystallin promoter-reporter constructs were hormone responsive. A putative glucocorticoid response element (GRE) within the analysed promoter region could bind the glucocorticoid receptor as revealed from in vitro footprint analysis but is not involved in the hormone-mediated gene activation. Deletions of 5' flanking regions to position -465 relative to the transcription start allowed for full hormone responsiveness. A deletion from -465 to -389 abolish hormone-mediated gene induction. No sequence element closely resembling a classical GRE is present within that hormone-responsive region.