Heat shock transcription factors and the hsp70 induction response in brain and kidney of the hyperthermic rat during postnatal development.

Heat shock transcription factor (HSF) 1 levels increase in brain regions and decline in kidney during postnatal rat development. In both neonatal and adult rats, levels of HSF1 protein in brain and kidney are proportional to the levels of HSF DNA-binding activity and the magnitude of heat shock protein hsp70 induction after thermal stress. There appears to be more HSF1 protein in adult brain than is needed for induction of hsp70 after thermal stress, suggesting that HSF1 may have other functions in addition to its role as a stress-inducible activator of heat shock genes. HSF2 protein levels decline during postnatal rat development in brain regions and kidney. Gel mobility shift analysis shows that HSF2 is not in a DNA-binding form in the neonatal brain and kidney, suggesting that HSF2 may not be involved in the constitutive expression of hsps in early postnatal development. There is no apparent relationship between levels of HSF2 protein and basal levels of hsp90, hsp70, heat shock cognate protein hsc70, and hsp60.

Localization of the heat-shock protein Hsp70 to the synapse following hyperthermic stress in the brain.

Heat-shock proteins are induced in response to cellular stress. Although heat-shock proteins are known to function in repair and protective mechanisms, their relationship to critical neural processes, such as synaptic function, has received little attention. Here we investigate whether the major heat-shock protein Hsp70 localizes to the synapse following a physiologically relevant increase in temperature in the mammalian nervous system. Our results indicate that hyperthermia-induced Hsp70 is associated with pre- and postsynaptic elements, including the postsynaptic density. The positioning of Hsp70 at the synapse could facilitate the repair of stress-induced damage to synaptic proteins and also contribute to neuroprotective events at the synapse.

Cellular localization of the heat shock transcription factors HSF1 and HSF2 in the rat brain during postnatal development and following hyperthermia.

The heat shock transcription factor HSF1 mediates the induction of heat shock genes in response to temperature elevation and other traumatic events. The induced hsps play roles in cellular repair and protective mechanisms. Immunocytochemistry revealed that in the unstressed rat, HSF1 was already prepositioned in the nucleus at abundant levels in both neuronal and glial cell types. Following a fever-like temperature, glial cells rapidly induced hsp70 whereas populations of large neurons did not. The lack of hsp70 induction in these neurons in vivo did not appear to be due to deficiencies in levels of nuclear HSF1. During postnatal development of the cerebellum, levels of HSF1 increased progressively from day 1 to 30. Members of the hsp gene set are also constitutively expressed in the unstressed animal and play roles as molecular chaperones. HSF2, which has been proposed as a developmental regulator of constitutive heat shock gene expression, demonstrated a developmental alteration in cellular localization, namely a nuclear distribution in neurons at postnatal day 2 and a cytoplasmic localization at day 30. During postnatal development the overall levels of neural HSF2 declined. This profile showed no obvious correlation with previously observed levels of constitutive hsp expression during postnatal neural development.

Effect of hyperthermia on the transcription rate of heat-shock genes in the rabbit cerebellum and retina assayed by nuclear run-ons.

The induction of heat-shock protein 70 (hsp70) mRNA in the hyperthermic rabbit brain has been examined previously by using Northern blotting and in situ hybridization procedures that measure steady-state levels of mRNA, which may be influenced by transcript stability and transcription rate. In the present investigation, the in vivo transcription rate of hsp70 has been examined by using run-on transcription assays on isolated brain nuclei. A major up-regulation in the transcription rate of hsp70 was observed between 0.75 and 1.50 hours after hyperthermia in the cerebellum and the retina. Gel-mobility shift assays revealed that the time course of conversion of heat-shock transcription factor (HSF1) to a DNA-binding form paralleled the transcriptional induction profile of hsp70. The transcription rates of several nonheat-shock genes were also studied in the hyperthermic brain, and little change was noted relative to the induction of hsp70. Thus, a physiologically relevant increase in temperature of 2.5 degrees C induces a major up-regulation in the in vivo transcription rate of hsp70 in the nervous system with little affect on the transcription rates of other genes.

Tissue-specific differences in heat shock protein hsc70 and hsp70 in the control and hyperthermic rabbit.

The ability to resolve protein members of the hsp70 multigene family by two-dimensional Western blotting permitted the characterization of antibodies which were specific in discriminating constitutively expressed hsc70 isoforms from stress-inducible hsp70 isoforms. This antibody characterization demonstrated that basal levels of hsp70 isoforms were present in the cerebellum of the control rabbit and that these were elevated following hyperthermia, whereas levels of hsc70 were similar in control and hyperthermic tissue. Multiple isoforms of hsp70 were detected but tissue-specific differences were not apparent in various organs of the rabbit. However, species differences were observed as fewer hsp70 isoforms were noted in rat and mouse. In the control rabbit, higher levels of hsc70 protein were present in neural tissues compared to non-neural tissues. Following physiologically relevant hyperthermia, induction of hsp70 was greatest in non-neural tissues such as liver, heart, muscle, spleen, and kidney compared to regions of the nervous system. These studies suggest that the amount of preexisting constitutive hsc70 protein may influence the level of induction of hsp70 in the stress response. Given this observation, caution is required in the employment of hsp70 induction as an index of cellular stress since endogenous levels of hsc70, and perhaps hsp70, may modulate the level of induction.

In vivo activation of neural heat shock transcription factor HSF1 by a physiologically relevant increase in body temperature.

Molecular mechanisms which underlie the heat shock response have commonly been analyzed using tissue culture systems, with less investigation of the intact mammal. In tissue culture, a temperature elevation of 5 degrees C is required to activate mammalian heat shock transcription factor 1 (HSF1) to the DNA-binding form. We demonstrate that a physiologically relevant increase in body temperature of 2.5 +/- 0.2 degrees C, similar to that attained during fever reactions, is sufficient to activate HSF1 in the rabbit nervous system. Maximal HSF activation, as measured by gel mobility shift assay, was attained at 1 hr with the cerebellum showing the strongest signal. Supershift experiments with antibodies specific to HSF1 and HSF2 demonstrated that the signal reflected activation of HSF1. Western blot analysis showed that cerebellum exhibited high levels of HSF1 protein.

Expression of mRNA species encoding heat shock protein 90 (hsp90) in control and hyperthermic rabbit brain.

Northern blot and in situ hybridization were employed to investigate regional and cell type differences in the expression of hsp90 mRNA species in control and hyperthermic rabbit brain. Riboprobes specific to hsp90 alpha and beta mRNA species were utilized in time-course Northern blot studies on cerebral hemispheres and the cerebellum. Following hyperthermia, levels of hsp90 alpha and beta mRNA were elevated in both brain regions; however, the magnitude of induction was more robust in the cerebellum than in cerebral hemispheres. The pattern of expression of hsp90 genes in rabbit brain was analyzed by in situ hybridization. These studies revealed that hsp90 genes are preferentially expressed in neuronal cell populations in the unstressed mammalian brain. The distribution of hsp90 alpha and beta mRNA was similar, though the signal for the latter was stronger. Following hyperthermia, changes were not detected in the pattern of hsp90 beta mRNA expression in the hippocampus. In the cerebellum, a rapid induction of hsp90 beta mRNA was apparent in the neuron-enriched granule cell layer, followed by a delayed accumulation in Purkinje neurons. Unlike hsp70, induction of hsp90 was not detected in glial cells of hyperthermic rabbit brain. The localization of hsp90 to neurons suggests that this heat shock protein plays an important role in neuronal function.

Localization of constitutive and hyperthermia-inducible heat shock mRNAs (hsc70 and hsp70) in the rabbit cerebellum and brainstem by non-radioactive in situ hybridization.

Neural expression of constitutive hsc70 mRNA and hyperthermia-inducible hsp70 mRNA is examined using radioactive and non-radioactive in situ hybridization procedures. A strong induction of hsp70 mRNA was noted in cell populations in cerebellar layers and in the brainstem which demonstrated expression of mRNA encoding proteolipid protein, an oligodendrocyte marker. The non-radioactive in situ hybridization procedure using digoxigenin (DIG)-UTP-labeled riboprobes permitted improved signal localization, and stress-inducible hsp70 mRNA was detected at the cytoplasmic cap areas of individual oligodendrocytes. Cell types which express constitutive members of the hsc/hsp70 multigene family were also identified. Neurons in the brainstem and in the deep white matter and molecular layer of the cerebellum showed expression of hsc70 mRNA while signal was not detected in adjacent glial cells. A neuron-specific enolase riboprobe aided in the identification of neuronal cell types. The non-radioactive DIG riboprobe revealed that hsc70 mRNA was highly localized to the cytoplasm of individual neurons. High constitutive levels of hsc70 in certain neurons may dampen hsp70 induction after hyperthermia in these cell populations.

Temporal and spatial distribution of heat shock mRNA and protein (hsp70) in the rabbit cerebellum in response to hyperthermia.

We have previously investigated the expression of hsp70 genes in the hyperthermic rabbit brain at the mRNA level by Northern blot and in situ hybridization procedures. Our studies have now been extended to the protein level utilizing Western blot and immunocytochemistry. Using an antibody which is specific to inducible hsp70, a prominent induction of hsp70 protein in glial cells of hyperthermic animals was noted. In particular, Bergmann glial cells in the cerebellum are strongly immunoreactive while adjacent Purkinje neurons are immunonegative. Extension of our in situ hybridization studies to a time course analysis revealed that the initial glial induction events were followed by a delayed accumulation of inducible hsp70 mRNA in Purkinje neurons at 10 hr post-heat shock. In control animals, high levels of constitutively expressed hsc70 mRNA and protein were observed in Purkinje neurons. Similar hsc70 and hsp70 mRNA observations were also made in neurons of the deep cerebellar nuclei and in motor neurons of the spinal cord. Our results suggest that these neuronal cell types accumulate hsp70 mRNA in response to hyperthermic treatment; however, the response is delayed when compared to the rapid response seen in glial cells. The high constitutive levels of hsc70 in certain neuronal cell types may play a role in the initial dampening of the hsp70 induction response in these cells.

Molecular cloning of a novel mRNA using an antibody directed against synaptic glycoproteins.

It has been suggested by a number of investigators that glycoproteins may play a role in the development and/or maintenance of synapses in the mammalian CNS. For many synaptic glycoproteins, however, little precise structural or functional information is available. In an effort to isolate probes specific to individual glycoproteins, we have screened a rat brain cDNA expression library with a mixed polyclonal antibody directed against concanavalin A-binding synaptic junctional glycoproteins. Using this approach, we have previously reported the cloning of SC1, a putative extracellular matrix glycoprotein found in adult brain (Johnston et al., Neuron 2:165-176, 1990). We now report the cloning and characterization of a second novel cDNA, which has been designated SC2. Northern blots show that this cDNA recognizes a 1.2-kb mRNA that is present throughout postnatal development in the rat. It is expressed at high levels in brain and is also found at lower levels in several other tissues. In situ hybridization suggests that the SC2 mRNA is strongly expressed by many types of neurons. Sequence data reveals a single open reading frame in the cDNA, encoding a putative hydrophobic protein with a calculated molecular weight of 36.1 kDa. Sequence analysis reveals some similarity between SC2 and 5 alpha-reductase, a microsomal membrane protein important in testosterone metabolism.

Induction of a heat shock gene (hsp70) in rabbit retinal ganglion cells detected by in situ hybridization with plastic-embedded tissue.

Elevation of body temperature by 2-3 degrees C induces a 2.7 kilobase hsp70 mRNA species in the rabbit retina within 1 hr. In situ hybridization with thin sections derived from plastic-embedded tissue permitted a higher level of resolution of retinal cell types compared to procedures which involved the use of frozen tissue sections. A prominent induction of hsp70 mRNA in retinal ganglion cells was observed when an hsp70 riboprobe was utilized for in situ hybridization. These results indicate that this neuronal cell type responds rapidly to fever-like body temperatures by inducing one of the major heat shock genes.

Expression of heat shock genes (hsp70) in the mammalian brain: distinguishing constitutively expressed and hyperthermia-inducible mRNA species.

The mammalian genome contains both constitutively expressed and heat-shock-inducible members of the hsp70 gene family. Riboprobes derived from members of these two classes of heat shock genes were utilized in Northern blot studies to analyze brain mRNA isolated from control rabbits and rats and from animals subjected to hyperthermic treatment. A riboprobe derived from a constitutively expressed hsp70 gene detected a 2.5 kilobase (kb) mRNA in brain tissue from control rabbits and a 2.3 kb mRNA species in control rat brain. These brain mRNAs showed little change in abundance in animals which were subjected to hyperthermic treatment. A riboprobe derived from a heat-shock-inducible hsp70 gene detected an abundant 2.7 kb brain transcript in hyperthermic rabbits which was not apparent in control animals. A time course study revealed that the induction of this mRNA species was transient and paralleled the rise and fall in body temperature. Peak induction was observed at 1 hr. The level of this message had greatly decreased by 5 hr and only trace levels were present at 10 and 24 hr. In the rat brain the induced hsp70 mRNA species was slightly larger than that observed in rabbit (2.9 kb vs. 2.7 kb). The riboprobe which detected the hyperthermia-inducible mRNA species was highly specific and did not cross react to the constitutively expressed mRNA species under the conditions employed in the Northern blot studies.

Induction of a 'stress' protein in intact mammalian organs after the intravenous administration of sodium arsenite.

The profile of nascent proteins synthesized in various rabbit organs after the intravenous injection of sodium arsenite was analyzed by the cell-free translation of purified polysomes. Examination of the translation products of polysomes isolated 1 hr after injection of sodium arsenite revealed a marked induction of synthesis of a protein of molecular weight 74,000 (74K) in the kidney, heart and liver which was similar to a 'heat shock' protein which was induced in these organs after elevation of body temperature by 2.5 to 3 degrees C. Synthesis of the 74K protein was not detected in the translation products of brain polysomes isolated 1 hr after sodium arsenite injection.