Insights into bunyavirus architecture from electron cryotomography of Uukuniemi virus.

Bunyaviridae is a large family of viruses that have gained attention as "emerging viruses" because many members cause serious disease in humans, with an increasing number of outbreaks. These negative-strand RNA viruses possess a membrane envelope covered by glycoproteins. The virions are pleiomorphic and thus have not been amenable to structural characterization using common techniques that involve averaging of electron microscopic images. Here, we determined the three-dimensional structure of a member of the Bunyaviridae family by using electron cryotomography. The genome, incorporated as a complex with the nucleoprotein inside the virions, was seen as a thread-like structure partially interacting with the viral membrane. Although no ordered nucleocapsid was observed, lateral interactions between the two membrane glycoproteins determine the structure of the viral particles. In the most regular particles, the glycoprotein protrusions, or "spikes," were seen to be arranged on an icosahedral lattice, with T = 12 triangulation. This arrangement has not yet been proven for a virus. Two distinctly different spike conformations were observed, which were shown to depend on pH. This finding is reminiscent of the fusion proteins of alpha-, flavi-, and influenza viruses, in which conformational changes occur in the low pH of the endosome to facilitate fusion of the viral and host membrane during viral entry.

Formation and intracellular transport of a heterodimeric viral spike protein complex.

We have analyzed the heterodimerization and intracellular transport from the ER to the Golgi complex (GC) of two membrane glycoproteins of a bunyavirus (Uukuniemi virus) that matures by a budding process in the GC. The glycoproteins G1 and G2, which form the viral spikes, are cotranslationally cleaved in the ER from a 110,000-D precursor. Newly synthesized G1 was transported to the GC and incorporated into virus particles about 30-45 min faster than newly synthesized G2. Analysis of the kinetics of intrachain disulfide bond formation showed that G1 acquired its mature form within 10 min, while completion of disulfide bond formation of G2 required a considerably longer time (up to 60 min). During the maturation process, G2 was transiently associated with the IgG heavy chain binding protein for a longer time than G1. Protein disulfide isomerase also coprecipitated with antibodies against G1 and G2. In virus particles, G1 and G2 were present exclusively as heterodimers. Immunoprecipitation with monoclonal antibodies showed that heterodimerization occurred rapidly, probably in the ER, between newly made G1 and mature, dimerization competent G2. Taken together, our results show that these two viral glycoproteins have different maturation kinetics in the ER. We conclude that the apparent different kinetics of ER to GC transport of G1 and G2 is due to the different rates by which these proteins fold and become competent to enter into heterodimeric complexes prior to exit from the ER.

Prominent expression of acidic fibroblast growth factor in motor and sensory neurons.

Several growth factors originally characterized and named for their action on a variety of cells have more recently been suggested to be importantly involved in the development and maintenance of the nervous system. Acidic fibroblast growth factor (aFGF) is a member of a family of seven structurally related polypeptide growth factors. The cells responsible for expression of aFGF in the nervous system of adult rats have been identified using an affinity-purified antibody to aFGF in immunohistochemical studies and synthetic oligonucleotide probes for in situ hybridization studies. High levels of aFGF expression were observed in motoneurons, primary sensory neurons, and retinal ganglion neurons. Glial cells did not express detectable amounts of aFGF. Confocal and electron microscopic analysis suggested that a large portion of aFGF immunoreactivity was associated with the cytoplasmic face of neuronal membranes, consistent with the hypothesis that aFGF is a sequestered growth factor.

Uukuniemi virus maturation: accumulation of virus particles and viral antigens in the Golgi complex.

We studied the maturation of Uukuniemi virus and the localization of the viral surface glycoproteins and nucleocapsid protein in infected cells by electron microscopy, indirect immunofluorescence, and immunoelectron microscopy with specific antisera prepared in rabbits against the two glycoproteins G1 and G2 and the nucleocapsid protein N. Electron microscopy of thin sections from infected cells showed virus particles maturing at smooth-surfaced membranes close to the nucleus. Localization of the G1/G2 and N proteins by indirect immunofluorescence at different stages after infection showed the antigens to be present throughout the cell interior but concentrated in the juxtanuclear region. The G1/G2 antiserum also appeared to stain the nuclear and plasma membranes. Double staining with tetramethylrhodamine isothiocyanate-conjugated wheat germ agglutinin, which preferentially stains the Golgi complex, and fluorescein isothiocyanate-conjugated anti-rabbit immunoglobulin G, which stained the G1/G2 or N proteins, showed that the staining of the juxtanuclear region coincided. Similarly, double staining for thiamine pyrophosphatase, an enzyme activity specific for the Golgi complex, showed the fluorescence and the cytochemical stain to coincide in the juxtanuclear region. Immunoperoxidase electron microscopy of cells permeabilized with saponin revealed that the viral glycoproteins were present in the rough endoplasmic reticulum and the nuclear and Golgi membranes; the latter was heavily stained. With this method, the N protein was localized to the cytoplasm, especially around smooth-surfaced vesicles in the Golgi region. Taken together, the results indicate that Uukuniemi virus and its structural proteins accumulate in the Golgi complex, supporting the idea that this compartment rather than the plasma membrane is the site of virus maturation. This raises the interesting possibility that deficient transport of the glycoproteins to the plasma membrane and hence their accumulation in the Golgi complex determines the site of virus maturation.

Genes for epidermal growth factor receptor, transforming growth factor alpha, and epidermal growth factor and their expression in human gliomas in vivo.

Anomalies of the epidermal growth factor receptor (EGFR) gene, including amplification, rearrangement, and overexpression, have been reported in malignant human gliomas in vivo. In vitro glioma cell lines coexpress EGFR and at least one of its ligands, transforming growth factor alpha, suggesting the existence of an autocrine growth stimulatory loop. We have studied the tumor tissue from 62 human glioma patients and examined the structure and quantity of the EGFR gene and its transcripts, as well as the quantity of the receptor protein. In addition we have examined the genes and transcripts coding for the pre-pro forms of epidermal growth factor and transforming growth factor alpha, the two endogenous EGFR ligands. EGFR gene amplification was detected in 16 of the 32 malignancy grade IV gliomas (glioblastoma) studied (50%), but only in 1 of 30 gliomas of lesser malignancy grade (I-III). All tumors with an amplified gene overexpressed EGFR mRNA. More than one-half (62.5%) of the glioblastomas with amplified EGFR genes also showed coamplification of rearranged EGFR genes and concomitant expression of aberrant mRNA species. Overexpression, without gene amplification, was observed in some of the low grade gliomas, and aberrant EGFR transcripts were also seen in some cases without gene amplification or detected gene rearrangements. mRNA expression for one or both of the pre-pro forms of the ligands was detected in every tumor studied. Thus, several mechanisms for the activation of the EGFR-mediated growth stimulating pathway are possible in human gliomas in vivo: expression of a structurally altered receptor that may have escaped normal control mechanisms; and/or auto-, juxta-, or paracrine stimulating mechanisms involving coexpression of receptor and ligands, with or without overexpression of the receptor.

Translation of p15.5INK4B, an N-terminally extended and fully active form of p15INK4B, is initiated from an upstream GUG codon.

The expression of the cyclin-dependent kinase inhibitor p15INK4B in normal cells after induction with TGF-beta1, or following overexpression from an adenovirus-encoded cDNA, appears on an SDS-polyacrylamide gel as a doublet. Here, the underlying mechanism behind the synthesis of the two species has been studied. By expressing cDNAs truncated at their 5' end, we found that the synthesis of the more slowly migrating form, called p15.5INK4B, is dependent on a sequence upstream of the first AUG codon thought to initiate translation of p15INK4B. Two potential, in frame, alternative upstream initiation codons, ACG and GUG, were individually changed to GCA encoding alanine. Analysis by in vitro translation, or immunoblotting of lysates from transfected 293 cells, showed that translation of p15.5INK4B is initiated at the GUG located 13 codons upstream of the first AUG. When this AUG was mutated, p15INK4B was no longer made. Instead, a shorter form, initiated at an in frame AUG located seven codons downstream, was synthesized. Finally, when both these AUGs were mutated, only p15.5INK4B was generated. Both p15INK4B and p15.5INK4B bound to CDK4 and CDK6, inhibited DNA synthesis, and caused replicative senescence of a human glioma cell line. We thus conclude that p15INK4B and p15.5INK4B, encoded by the CDKN2B gene, are functionally indistinguishable as based on these assays.

The coxsackie-adenovirus receptor--a new receptor in the immunoglobulin family involved in cell adhesion.

The physiological and cell biological aspects of the Coxsackie-Adenovirus Receptor (CAR) is discussed in this review. The receptor obviously recognizes the group C adenoviruses in vivo, but also fibers from other groups except group B in vitro. The latter viruses seem to utilize a different receptor. The receptor accumulates at, or close to, the tight junction in polarized epithelial cells and probably functions as a cell-cell adhesion molecule. The cytoplasmic tail of the receptor is not required for virus attachment and uptake. Although there is a correlation between CAR and uptake of adenoviruses in several human tumor cells, evidence of an absolute requirement for integrins has not been forthcoming. The implication of these findings for adenovirus gene therapy is discussed.

Chemical synthesis and molecular cloning of a STOP oligonucleotide encoding an UGA translation terminator in all three reading frames.

We have chemically synthesized an oligonucleotide 5'd(TGATTGATTGA)3' 3'd(ACTAACTAACT)5' that encodes the translation termination codon TGA in all three reading frames. After ligation of appropriate restriction endonuclease linkers to the ends, the double-stranded oligonucleotide (STOP-oligonucleotide) was joined to the plasmid pBR322 between the EcoRI and BamHI, or HindIII and BamHI sites, and the hybrid plasmids were transformed into Escherichia coli HB101. Four different constructions were obtained: (i) EcoRI-STOP-BamHI (STOP-oligonucleotide flanked by EcoRI and BamHI linkers; pKTH606), (ii) HindIII-STOP-BamHI (pKTH601), (iii) BamHI-STOP-HindIII (pKTH604), and (iv) HindIII-STOP-POTS-BamHI (two STOP-oligonucleotides in opposite orientation; pKTH605). The inserts in pKTH606 and pKTH601 were excised and transferred to a modified plasmid constructed previously for the expression and secretion of foreign gene products from Bacillus subtilis. The resulting secretion plasmids now contain the promoter/signal sequence region of the alpha-amylase gene from Bacillus amyloliquefaciens joined to the STOP-oligonucleotide by EcoRI or HindIII linkers. Foreign genes can be cloned into these sites. The plasmids can be used to express foreign genes truncated at their C-terminal end and therefore lacking their own translation termination codon. One such plasmid has been successfully used to express the Semliki Forest virus (SFV) membrane protein E1 truncated at its C-terminus.

Nucleotide sequence of the promoter and NH2-terminal signal peptide region of the alpha-amylase gene from Bacillus amyloliquefaciens.

We have isolated and partially sequenced the gene coding for alpha-amylase (EC 3.2.1.1) from Bacillus amyloliquefaciens by molecular cloning in the plasmid pUB110 using Bacillus subtilis as a host. The nucleotide sequence of the NH2-terminal region of the cloned gene was determined and found to contain a 31-residue-long stretch of amino acids preceding the NH2-terminal sequence of the extracellular alpha-amylase. Within this sequence there is a 15-residue-long stretch of uncharged amino acids similar to that found at the NH2 terminus of other precursors to exported proteins. This "signal sequence" is probably removed in conjunction with the translocation of alpha-amylase through the cytoplasmic membrane. In vitro labeling of alpha-amylase with radioactive amino acids in a coupled transcription-translation system followed by partial sequencing established the exact location of the NH2 terminus of the alpha-amylase gene. The nucleotide sequence preceding the NH2 terminus has properties resembling the RNA-polymerase- and ribosome-binding sites found at the 5' terminus of many prokaryotic genes.

Temporal and spatial increase of astroglial basic fibroblast growth factor synthesis after 6-hydroxydopamine-induced degeneration of the nigrostriatal dopamine neurons.

The present study investigates the temporal and spatial changes of the cellular expression of basic fibroblast growth factor messenger RNA and immunoreactivity after a 6-hydroxydopamine-induced lesion in the nigrostriatal dopamine system. In situ hybridization revealed a sustained (from 4 h to two weeks) and strong (300-400% of control, at the peak intervals) increase of basic fibroblast growth factor messenger RNA in the pars compacta of the substantia nigra and the ventral tegmental area ipsilateral to the lesion. A short-lasting increase of basic fibroblast growth factor messenger RNA was observed in he ipsilateral pars reticulata of the substantia nigra (from 4-24 h, 300% of control) and neostriatum (24 h, 180% of control) as well as in the ipsilateral and contralateral hippocampus and neocortex (by 4 h, 200% of control). Brightfield microscopy showed an increased number of putative glial cells expressing the basic fibroblast growth factor messenger RNA signal. Basic fibroblast growth factor immunohistochemistry revealed on control brains the protein in the nuclei of glial cells throughout the forebrain and the midbrain and in the nuclei of neurons of the layer II of the retrosplenial granular cortex, the CA2 region of the hippocampus and the fasciola cinereum as well as in the nuclei of ependymal cells. The injection of 6-hydroxydopamine increased basic fibroblast growth factor immunoreactivity in the nuclei of astrocytes only within the ipsilateral substantia nigra and ventral tegmental area. By 2 h after the drug injection, the density of glial basic fibroblast growth factor-immunoreactive profiles was increased in the pars compacta of the substantia nigra and the ventral tegmental area. The density, size and intensity of the astroglial basic fibroblast growth factor immunoreactive nuclei were increased in the entire substantia nigra and the ventral tegmental area at 72 h, and peaked one week after the 6-hydroxydopamine injection. The saline injection promoted a time-dependent increase in the density of the glial basic fibroblast growth factor immunoreactivity but only in the ipsilateral pars compacta of the substantia nigra. In conclusion, the dopamine cell degeneration may give rise to extracellular signals activating the surrounding astroglia, leading to a sustained increased synthesis of astroglial basic fibroblast growth factor, which may exert neuroprotective action and increase repair on the nigrostriatal dopamine system.

Brain-derived neurotrophic factor (BDNF) peptide antibodies: characterization using a Vaccinia virus expression system.

We describe and characterize a series of polyclonal antibodies, generated against amino acid sequences unique to various regions within pro- and mature brain-derived neurotrophic factor (BDNF), a member of the highly conserved nerve growth factor (NGF) family of neurotrophins. Synthetic peptides were coupled to carrier proteins in the presence of glutaraldehyde to restrict the host animals' immune response to epitopes that are compatible with aldehyde fixation. Initial screenings of the reactivity of the antisera were made on brain sections processed for immunohistochemistry after peptide injections into brain parenchyma. As a means of further characterizing these peptide antisera, we have evaluated the reactivity and specificity of the peptide antibodies in BHK cells expressing recombinant pro- and mature BDNF protein from a T7 RNA polymerase-driven Vaccinia virus system. Several of the antibodies strongly stained components of cells transfected with the BDNF gene but did not label wild-type cells nor cells containing only the expression vector. It has also been possible to detect differential compartmentalization of the BDNF protein at various stages of processing in the BHK cells, as well as in situ in cryostat sections of brain tissue, with antisera to the pro- and mature protein. We conclude that several of our antisera recognize not only the specific peptide immunogens but also what appears to be the corresponding protein native to neurons.

The membrane glycoprotein G1 of Uukuniemi virus contains a signal for localization to the Golgi complex.

Members of the Bunyaviridae family acquire their envelopes by budding into the Golgi complex (GC). The accumulation of the membrane glycoproteins G1 and G2 in the GC probably determines the site of maturation. Here we have studied the intracellular transport and targeting to the GC of G1 and G2 of Uukuniemi virus, a member of the Phlebovirus genus, and report on their expression from cloned cDNAs either together or separately by using a T7 RNA polymerase-driven vaccinia virus expression system. When G1 and G2 were expressed together from a full-length cDNA as the p110 precursor, both proteins were localized to the Golgi complex, as evidenced by colocalization with the Golgi marker enzyme mannosidase II. Immunofluorescent staining indicated that G1 expressed alone also localized to the GC. However, pulse-chase experiments showed that G1 remained endoglycosidase H sensitive. G2 expressed alone remained associated with the endoplasmic reticulum (ER). G2 could be rescued from the ER and transported to the GC by coexpression with G1 from separate mRNAs. Coexpression also increased the efficiency of G1 transport to the GC. With none of the constructs could the glycoproteins be observed on the cell surface. These results show that efficient export of G1 and G2 from the ER requires coexpression of both proteins, in conformity with our previous results showing that G1 and G2 form heterodimeric complexes in the ER. Since G1 expressed alone is retained in the GC, we conclude that G1 contains a retention signal for localization to the GC. G2 might thus become associated with the GC indirectly via its interaction with G1.

Secretion of Semliki Forest virus membrane glycoprotein E1 from Bacillus subtilis.

The gene coding for the Semliki Forest virus (SFV) membrane protein E1 was joined to a secretion vector containing the promoter and signal sequence regions of the alpha-amylase gene from Bacillus amyloliquefaciens. To facilitate secretion, the regions coding for the N-terminal signal peptide (the 6K protein) and the C-terminal hydrophobic transmembrane domain of the E1 gene were deleted. After transformation into B. Subtilis, E1 was shown by immunoblotting to be expressed at a low level (about 0.5-1 mg/1). Contrary to what was expected, most of the E1 remained cell-associated. Deletion of a residual 7 C-terminal amino acids from the 6K region neither increased the level of expression nor significantly improved the secretion. Immunofluorescence microscopy of protoplasts prepared from B. subtilis cells expressing E1 suggested that the cell-associated E1 was located at the outer surface of the bacterial membrane. Addition of protease inhibitors to the culture medium somewhat increased the amount of extracellular E1, suggesting that proteolytic degradation of the foreign gene product may be one reason for the low level of expression. This conclusion was also supported by experiments carried out in Bacillus minicells, which indicated that the expression of the E1 gene in the absence of synthesis of bacterial proteases was about the same as that of alpha-amylase expressed from the cloned gene using the same promoter and signal sequence.

Computer-assisted mapping of basic fibroblast growth factor immunoreactive nerve cell populations in the rat brain.

We have performed a mapping of basic fibroblast growth factor (bFGF) immunoreactive (ir) glial and nerve cell populations in the male rat brain using a rabbit antibody raised against a synthetic peptide of bovine bFGF. Regional morphometric and microdensitometric analysis of the bFGF ir neuronal profiles in coronal brain sections was carried out by means of an automatic image analyser. The density and intensity of the bFGF ir glial profiles were subjectively evaluated. The bFGF immunoreactivity (IR) was detected within the cytoplasm of neurons, except within the pyramidal neurons of hippocampal CA2 region, the fasciola cinerea and the indusium griseum, where bFGF IR was present in the nucleus. In contrast, in glial cells bFGF IR was always found in the nucleus. Neuronal and glial IR was no longer observed after absorption of the bFGF antiserum with recombinant bFGF. Basic FGF IR was found in neuronal and glial cell populations throughout the brain as well as in the choroid plexus and in the ependymal cells lining the ventricles. Basic FGF ir nerve cells were found in all layers of both the neocortex and allocortex. Within the caudate putamen and the nucleus accumbens a low density of weak bFGF ir neuronal profiles was detected. The majority of the thalamic nuclei showed medium to high densities of moderate to strong bFGF ir neuronal profiles. All the hypothalamic nuclei, with the exception of the anterior and lateral hypothalamic area and of the ventral hypothalamic nucleus, contained a high density of bFGF ir profiles. The pons and the medulla oblongata were characterized by the presence of a large number of nuclei containing moderate to high densities of strong bFGF ir profiles. The Purkinje cell layer of the cerebellar cortex contained a high density of moderately bFGF ir profiles. A moderate density of strong bFGF ir nerve cell profiles was observed within all the laminae of the spinal cord, except within the II and III laminae where a high density of strongly ir profiles was found. Histogram analysis of total immunoreactivity showed that the distribution of bFGF ir profiles within the telencephalon and mesencephalon tend to be similar with regard to the central tendency and spread. Using Kendall's tau, a significant correlation between intensity and density values was obtained only in the diencephalon. The cytoplasmic bFGF IR found in distinct nerve cell populations all over the rat brain and spinal cord may represent forms of bFGF which can be released from the nerve cells via non-exocytotic mechanisms in view of the absence of an intracellular signal peptide in bFGF. The presence of nuclear bFGF IR within the glial cells all over the central nervous system (CNS) suggests an intracellular function of bFGF, such as the promotion of mitogenesis and/or participation in the transcriptional regulation of various genes.

Differential expression of acidic and basic FGF in the rat substantia nigra during development.

Both acidic (aFGF) and basic (bFGF) fibroblast growth factors have been shown to be present in the adult rat ventral mesencephalon and to exert effects on cultured mesencephalic cells. In the present study we have examined the expression of aFGF and bFGF in the rat ventral mesencephalon at various stages of development. bFGF was present at all ages examined [embryonic day 16 (E16) to postnatal day 90 (P90)]. In contrast, aFGF was not detectable at embryonic and early postnatal ages, but was observed at later (P20, P60, P90) postnatal stages. These data suggest that aFGF and bFGF may have functions in mesencephalic dopamine neurones in different stages of development.

Corticosterone increases FGF-2 (bFGF) immunoreactivity in the substantia nigra of the rat.

The effects of acute and subchronic (7 days) administrations of the adrenocortical hormone corticosterone on basic fibroblast growth factor (bFGF, FGF-2) immunoreactivity were studied in the substantia nigra of the rat by semiquantitative immunocytochemistry coupled with image analysis. Corticosterone was able to increase FGF-2 immunoreactivity in different nigral subregions and cell types (astrocytes and neurones) depending on the duration of the treatment. These results open up the possibility that stress hormones can modulate the trophic state of the substantia nigra through an action on FGF-2.

Basic FGF is present in dopaminergic neurons of the ventral midbrain of the rat.

Immunocytochemistry procedures and 6-hydroxy-dopamine-induced degeneration of dopamine nerve cells provided evidence that practically all tyrosine hydroxylase immunoreactive (IR) neurons of the substantia nigra and ventral tegmental area contain cytoplasmic basic fibroblast growth factor immunoreactivity (bFGF IR), while many astroglial cells in the neostriatum and substantia nigra contain nuclear bFGF IR. RNA blot analysis demonstrated strong expression of a major 3.7 kb bFGF mRNA in the substantia nigra and ventral tegmental area. These results indicate that bFGF may be a significant growth factor in the DA neurons.

aFGF, bFGF and NGF differentially regulate neuropeptide expression in dorsal root ganglia after axotomy and induce autotomy.

Using immunohistochemistry and in situ hybridization the in vivo effects of acidic and basic fibroblast growth factor (aFGF, bFGF), and of nerve growth factor (NGF) on the expression of galanin, neuropeptide Y (NPY) and substance P in axotomized dorsal root ganglia (DRGs) were examined. Self-mutilation (autotomy), a supposed pain-related behavior, was investigated after growth factor treatment. One microgram of aFGF, bFGF or NGF was applied directly to the transected sciatic nerve via a capsule. In normal rats 3.2%, 0% and 17.5% of the neuron profiles in the DRGs contained galanin-, NPY- and substance P-like immunoreactivity (LI), respectively. Sciatic nerve transection induced a distinct increase in galanin- and NPY-LIs, but a downregulation of substance P-LI. Thus three days after axotomy 23.5%, 26.9% and 9.8% of the DRG neuron profiles showed immunoreactivity for galanin-, NPY- and substance P-LI, respectively. In vivo administration of aFGF counteracted the axotomy-induced increase in galanin and NPY, whereas bFGF only suppressed NPY upregulation. NGF reversed in the injury-induced decrease in substance P-LI, but had no significant effect on galanin- and NPY-LIs. These results were confirmed by monitoring the mRNA levels for these neuropeptides. Moreover, aFGF was found to induce autotomy in 60% of the rats 3 days after axotomy. NGF produced autotomy in about 30% of the rats. Taken together, the present results suggest (1) that aFGF, bFGF and NGF differentially regulate neuropeptide expression in vivo; (2) that FGFs can inhibit neuropeptide upregulation of some peptides after nerve injury; and (3) that aFGF and NGF may induce pain-related behavior.

Photochemically induced focal cerebral ischemia in rat: time dependent and global increase in expression of basic fibroblast growth factor mRNA.

Induction of basic fibroblast growth factor (bFGF) mRNA expression was studied in a Rose bengal induced focal cerebral ischemia during a time course of 2, 4, 24, 72 h and 7 days. Focal cerebral ischemia induced by Rose bengal resulted in a global upregulation in bFGF gene expression at the 24 h time-interval. This upregulation in bFGF gene expression was due to an upregulation in glial bFGF expression in most of the areas studied as seen by means of non-radioactive in situ hybridization in combination with immunocytochemistry for glial fibrillary acidic protein. However, in the piriform cortex a putative neuronal upregulation of bFGF could be detected by combination of non-radioactive in situ hybridization, immunohistochemistry for glial fibrillary acidic protein and nuclear staining with Neutral red. Semiquantitative data concerning bFGF mRNA expression were obtained by use of computer-assisted microdensitometry and revealed substantial increases in bFGF mRNA expression in the cingulate cortex, the neostriatum, a 1 mm marginal zone close to the external capsule and the olfactory tubercle at bregma levels 1 to 2 mm rostral to the lesion. No changes in bFGF gene expression were seen in field CA1 of Ammon's horn on the lesioned side and in dentate gyrus at bregma levels between -2.12 to -3.30 mm. We observed significant changes in bFGF upregulation in the caudate putamen, the piriform cortex and the amygdaloid region and the frontoparietal cortex at bregma levels -2.12 to -3.30 mm. These data indicate that photochemically induced focal cerebral ischemia leads to an early and global response in bFGF gene expression, which is due to an upregulation mainly in astrocytes. The observed widespread upregulation of the bFGF gene transcription rostral and caudal to the lesion is suggested to be due in part to neuronal glutaminergic connections between the areas investigated and in part due to increases in extracellular fluid signals (volume transmission).