Elsevier

Brain Research Reviews

Volume 32, Issue 1, 24 March 2000, Pages 45-56
Brain Research Reviews

Short review
Connexin43 null mice reveal that astrocytes express multiple connexins

https://doi.org/10.1016/S0165-0173(99)00067-3Get rights and content

Abstract

The gap junction protein connexin43 (Cx43) is the primary component of intercellular channels in cardiac tissue and in astrocytes, the most abundant type of glial cells in the brain. Mice in which the gene for Cx43 is deleted by homologous recombination die at birth, due to profound hypertrophy of the ventricular outflow tract and stenosis of the pulmonary artery. Despite this significant cardiovascular abnormality, brains of connexin43 null [Cx43 (−/−)] animals are shown to be macroscopically normal and to display a pattern of cortical lamination that is not detectably different from wildtype siblings. Presence of Cx40 and Cx45 in brains and astrocytes cultured from both Cx43 (−/−) mice and wildtype littermates was confirmed by RT-PCR, Northern blot analyses and by immunostaining; Cx46 was detected by RT-PCR and Northern blot analyses. Presence of Cx26 in astrocyte cultures was indicated by RT-PCR and by Western blot analysis, although we were unable to resolve whether it was contributed by contaminating cells; Cx30 mRNA was detected by Northern blot in long term (2 weeks) but not fresh cultures of astrocytes. These studies thus reveal that astrocyte gap junctions may be formed of multiple connexins. Presumably, the metabolic and ionic coupling provided by these diverse gap junction types may functionally compensate for the absence of the major astrocyte gap junction protein in Cx43 (−/−) mice, providing whatever intercellular signaling is necessary for brain development and cortical lamination.

Introduction

Astrocytes are increasingly appreciated as active participants in neural function, providing not only efficient dissipation of ionic concentration gradients and neurotransmitter uptake during neuronal activity, but also delivering crucial trophic and metabolic signals to neurons 11, 19. Glial cells form communication compartments in the nervous system, in which intracellular ions and small molecules are exchanged directly as a result of electrotonic and metabolic coupling mediated through gap junction channels 10, 11. Although coupling is variably strong among astrocytes from different brain regions (e.g., Refs. 1, 22), it is likely that the network of coupled glia may extend virtually throughout all areas of the central nervous system 32, 33. Astrocytes possess uptake mechanisms for K+ as well as for various neurotransmitters, and these cells also display a remarkable capacity for volume regulation 20, 35. The functions served by the gap junctions connecting this glial syncytium are thus hypothesized to include providing a conduit for removal of extracellular K+, taking up neurotransmitters (e.g., glutamate) from the local environment of neurons [10], serving as a route for delivery of metabolites throughout the network [15]and increasing buffer capacity for dilution of toxic or metabolically active compounds [3].

The predominant gap junction protein expressed by astrocytes has been believed to be connexin43 (Cx43), as evaluated by Northern and Western blot analyses, by immunostaining, and using electrophysiological criteria 8, 12, 14, 25, 40, 28. However, the number of known connexins has greatly expanded since astrocytic connexin expression and distribution in brain were initially explored [12], raising the possibility that additional connexins might be expressed in these cells. Indeed, recent studies using RT-PCR, immunostaining, in situ hybridization and Northern and Western blot analyses have revealed Cx45 and its mRNA in astrocytes in the intact brain, suggesting that Cx45 gap junction channels might also play a role in intercellular communication among astrocytes [7].

Mice have now been genetically engineered in which Cx43 expression has been deleted through the use of homologous recombination techniques [34]. The Cx43 (−/−) mice exhibit a pronounced cardiovascular malformation, characterized by pulmonary stenosis and hypertrophied right ventricle, and these animals die at birth, presumably as a consequence of severe cardiopulmonary insufficiency when the newborn mouse is deprived of placental circulation. As is demonstrated in this report, however, there are no gross anatomical abnormalities in the brain [31], despite the high degree of Cx43 expression in this organ in adult wildtype animals and during brain development 12, 26, 40.

More than a dozen members of the connexin gene family are now known in rodents [37], and it is plausible that the lack of gross anatomical abnormalities in brains of Cx43 (−/−) mice might result from rescue of function due to either upregulated or constitutive expression of additional gap junction proteins. Moreover, these animals offer the opportunity to explore whether other connexins might be normally expressed in astrocytes, their functional presence possibly having gone undetected in the high Cx43 background that is present in wildtype brain. Our findings, in agreement with those of Perez-Velazquez et al. [31], indicate that astrocytes from Cx43 (−/−) and wildtype mice are similarly differentiated, as evidenced by expression of the phenotypic marker glial fibrillary associated protein (GFAP). Northern and Western blot analyses and immunocytochemical studies here identify Cx40 and Cx45 in normal and Cx43 (−/−) astrocytes, strengthening the evidence that Cx40 and Cx45 are additional functional gap junction channels in astrocytes. In addition, RT-PCR analyses and Northern blot demonstrate the expression of mRNA encoding Cx46 in both wildtype and Cx43 (−/−) astrocytes.

This analysis of Cx43 (−/−) mice has thus revealed the presence of additional gap junction proteins in astrocytes from these animals as well as their hetero- and homozygous wildtype littermates. Furthermore, our finding that Cx43 expression is not required for expression of the astrocyte phenotype (as evaluated by staining for GFAP) or for macroscopically normal brain development indicates that the strong coupling that characterizes the glial compartment in adult brain is not necessary for either early differentiation of these cells or the establishment of cortical lamination. Presumably, the low level of coupling provided by expression of a variety of supplemental gap junction proteins in the astrocytes of Cx43 (−/−) mice is sufficient for the exchange of morphogens during critical steps in early neural ontogeny.

Section snippets

Cx43 (−/−) animals

Mating pairs of Cx43 hemizygous (+/−) mice (CGJA1M1) were obtained from Jackson Laboratories (Bar Harbor, ME) and maintained in our AALC-accredited animal facilities. Results described below were representative of measurements obtained on numerous platings of astrocytes (passages 0 to 3) isolated from more than 20 litters of matings of more than fifteen different parents and on brain tissue obtained from an additional eight litters.

Genotyping progeny

In order to determine the genotype of the animals from which

Determination of progeny genotypes

At the time of birth, tail tissue was obtained from each animal. PCR analyses for a representative litter of animals is illustrated in Fig. 1, using primers designed to amplify short segments of wildtype Cx43 and knockout, neo-containing alleles (Table 1; amplified loci indicated in top portion of Fig. 1). Mice were considered to be homozygous for the knockout allele if PCR detected only the 101 bp amplicon, homozygous wildtype if only the 269 bp amplicon was revealed, and heterozygous if both

Discussion

It has previously been reported that astrocyte morphology is quite similar in brain slices from Cx43 (−/−) and wildtype mice [31]and that growth rate is slower for Cx43 (−/−) astrocytes than wildtype in both organotypic [31]and dissociated cell cultures [29]. We found that brains of Cx43 (−/−) mice were morphologically normal with regard to overall sizes and gross cytoarchitecture of cortical lamination. Moreover, astrocytes from Cx43 (−/−) mice were well differentiated, as evidenced by

Acknowledgements

We gratefully acknowledge the assistance of Mr. Joseph Zavilowitz in maintaining and breeding colony of these mice and the secretarial and editorial assistance of Ms. Frances Andrade in the preparation of this manuscript. Major support for these studies was provided by NIH Program Project grant NS07512 to MVLB, with projects directed by MVLB and DCS, and by a grant from the Deutsche Forschungsgemeinschaft (SPP Glia) to RD. Additional support was derived from NIH grant NS34931 and a Grant-in-Aid

References (40)

  • R. Dermietzel, Molecular diversity and plasticity of gap junctions in the nervous system, in: D.C. Spray, R. Dermietzel...
  • R. Dermietzel et al.

    Gap junctions between cultured astrocytes: immunocytochemical, molecular, and electrophysiological analysis

    J. Neurosci.

    (1991)
  • R. Dermietzel et al.

    Oligodendrocytes express gap junction proteins connexin32 and connexin45

    Glia

    (1997)
  • R. Dermietzel et al.

    Gap junctions in the brain: where, what type, how many, and why?

    Trends Neurosci.

    (1993)
  • R. Dermietzel et al.

    From glue (`nervenkitt') to glia: a prologue

    Glia

    (1998)
  • R. Dermietzel et al.

    Differential expression of three gap junction proteins in developing and mature brain tissues

    Proc. Natl. Acad. Sci. U.S.A.

    (1989)
  • M.M. Froes et al.

    Gap junctional coupling between neurons and astrocytes in primary CNS cultures

    Proc. Natl. Acad. Sci.

    (1999)
  • C. Giaume et al.

    Metabolic trafficking through astrocytic gap junctions

    Glia

    (1997)
  • P.A. Guerrero et al.

    Slow ventricular conduction in mice heterozygous for a connexin43 null mutation

    J. Clin. Invest.

    (1997)
  • A. Hofer et al.

    C-erb2/neu transfection induces gap junctional communication incompetence in glial cells

    J. Neurosci.

    (1996)
  • Cited by (187)

    View all citing articles on Scopus
    View full text