Developmental expression of voltage-gated potassium channel β subunits

Developmental Brain Research 117 Ž1999. 71–80 www.elsevier.comrlocaterbres

Research report

Developmental expression of voltage-gated potassium channel b subunits Martha Downen ) , Stanley Belkowski 1, Heather Knowles, Marina Cardillo, Michael B. Prystowsky Department of Pathology, Albert Einstein College of Medicine, 1300 Morris Park AÕe., Bronx NY 10461, USA Accepted 13 July 1999

Abstract Expression of potassium channel beta subunits ŽKvb . was determined in the developing mouse CNS using an antiserum against an amino acid sequence present in the C-terminus of Kvb1, Kvb2, and Kvb3. Using the anti-Kvb antiserum, we determined that Kvb expression is restricted to the spinal cord and dorsal root ganglia in the embryonic CNS. At birth, Kvb expression is detected in brainstem and midbrain nuclei, but was not detected in the hippocampus, cerebellum or cerebral cortex. During the first postnatal week, Kvb expression is present in hippocampal and cortical pyramidal cells and in cerebellar Purkinje cells. Expression of Kvb subunits reaches adult levels by the third postnatal week in all of the brain regions examined. A rabbit antiserum directed against a unique peptide sequence in the N-terminus of the Kvb1 protein demonstrates that this subunit displays a novel expression pattern in the developing mouse brain. Kvb1 expression is high at birth in all brain regions examined and decreases with age. In contrast, Kvb2 expression is low at birth and increases with age to reach adult levels by the third postnatal week. These findings support the notion that the differential regulation of distinct potassium channel beta subunits, in the developing mouse nervous system, may confer the functional diversity required to mediate both neuronal survival and maturation. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Potassium channels; Auxiliary subunits; Kvb1; Kvb2; Mouse; Ion channels

1. Introduction Voltage-gated potassium ŽKv. channels comprise a functionally diverse group of membrane bound ion channels that are formed from four pore-forming a subunits consisting of six transmembrane domains whose function is modulated via association with auxiliary b subunits w13,24,28x. Four different Kvb subunits have been identified which are uniquely expressed in lymphocytes, brain and heart w13x. We reported the isolation of a novel cDNA clone, referred to as F5, from cloned T-lymphocytes stimulated with IL-2, and noted that the corresponding mRNA is expressed in the adult mouse central nervous system w6,29x. Analysis of the complete coding sequence of F5 identified the corresponding protein as Kvb2, one of the auxiliary subunits of the voltage-gated potassium channels w4x. In

)

Corresponding author. Fax: q 1-718-430-8867; e-mail: [email protected] 1 Current address: Fox Chase Cancer Center, 7701 Burholme Rd., Philadelphia PA 19111, USA.

previous studies using an anti-C-terminal peptide antiserum Žanti-Kvb ., the expression of Kvb in the adult brain was localized to the Purkinje cells in the cerebellum, the pyramidal and dentate granule cells of the hippocampus, layer III and V pyramidal neurons in the cerebral cortex and to the choroid plexus w1x. Kvb was localized to the cell body and dendrites of neurons while the protein was not detected in glial cells. While several studies have characterized the heterogeneity of functional Kq channels in vitro via electrophysiology performed on cells transfected with specific subunits, the in vivo expression patterns of specific subunits of voltage-gated Kq channels are not well defined w28x. The distribution of Kva and b subunits in the adult nervous system w25x as well as the distribution of a subunits in the developing nervous system w19x has been described. However, the developmental expression of Kvb in discrete regions of the CNS has not been determined. Our previous studies suggest that Kvb expression first appears late in postnatal development. Specifically, Northern blot analysis was used to show that in the cerebral cortex, the Kvb mRNA was first detected during the first

0165-3806r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 3 8 0 6 Ž 9 9 . 0 0 1 0 0 - 5

72

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

postnatal week and attained adult levels by postnatal day ŽP.21 w1,2x. Western blot analysis of whole brain isolated from mouse identified a specific 42 kDa protein that first appeared at P21 w1x. The distribution of Kvb mRNA was also determined in postnatal mouse brain using in situ hybridization w2x. Kvb mRNA was not detected at P2; weak staining was detected in large neurons in the neocortex, hippocampus, thalamus, hypothalamus, cerebellar cortex, brainstem and olfactory bulb at P9. At P22, the distribution and expression was similar to that of the adult. While a recent report identified Kvb1 mRNA expression in developing rat brain w5x, b1 and b2 protein expression in the developing CNS have not been described. Since our previous developmental studies were performed using whole brain homogenates, it is likely that low levels of expression in discrete neuronal populations would go undetected. The expression of a channels in the developing rodent hippocampus displays considerable regional heterogeneity w19x. Since channel properties are modulated by the association of the b subunits w28x, determining protein expression in the developing CNS may define critical time points during development in which channel kinetics are reflected by the expression of specific subunits. Also, given that functional ion channels may modulate processes regulating neuronal development, we were interested in determining the developmental expression of Kvb protein in the CNS. Our present studies determine the expression of Kvb subunits in discrete nuclei during development using immunocytochemical methods and quantitative Western blot analysis.

2. Experimental methods

was shown by Western blot analysis including competition by immunizing peptide, but not a Kvb2 peptide. AntiKvb1 peptide antibodies were enriched by affinity chromatography using the immunizing peptide conjugated to sepharose. 2.3. Immunocytochemistry Briefly, the CNS was harvested from mice at ages ranging from embryonic day ŽE. 16 through adult, and immersed in Bouin’s fixative overnight at room temperature followed by 30% sucrose in PBS, pH 7.4 at 48C. Four to eight micron sections of paraffin-embedded tissue were prepared and mounted on poly-lysine coated slides ŽSigma.. Sections were de-paraffinized, blocked in 3% normal goat serum in PBS, pH 7.4 for 1 h at room temperature, and incubated with the polyclonal rabbit anti-Kvb Ž1:100. or anti-Kvb1 Ž1:25. overnight at 48C. The sections were incubated for 2 h at room temperature with a goat anti-rabbit-horseradish peroxidase conjugate Ž1:500. followed by 30 min with the avidin–biotin complex ŽSigma.. Immune complexes were visualized following reaction with nickelenhanced diaminobenzidine, for 10 min. Specificity of the secondary antibody was confirmed by the absence of reaction product when the primary antibody was omitted. Specificity of the anti-Kvb antiserum has been previously reported w1x. Specificity of the anti-Kvb1 antiserum was determined by reduced staining in cell bodies in the adult brain sections following pre-absorption of the primary antibody with the immunizing peptide. We also used a dot blot to determine that the anti-Kvb1 antiserum recognized the immunizing peptide, a 20 amino acid sequence uniquely expressed in the Kvb1 N-terminus, and that it did not cross-react with a peptide uniquely expressed in the Kvb2 N-terminus.

2.1. Animals 2.4. Western blot analysis Timed pregnant C57BLr6 mice were obtained from Jackson Labs ŽBar Harbor, ME.. 2.2. Antiserum preparation Rabbit anti-Kvb antiserum prepared by Arai and Cohen w1x was generated by immunizing rabbits with an 18 amino acid sequence derived from the C-terminus of the deduced protein conjugated to keyhole limpet hemocyanin via a N-terminal cysteine residue as previously described. Antibody specificity was demonstrated through binding to recombinant protein expressed in COS cells w1x. A rabbit antiserum recognizing Kvb1 was prepared for selective analysis of the b1 subunit. A 20 amino acid sequence unique to the N-terminus of Kvb1 was synthesized ŽLMA, AECOM, Ruth Angeletti, Director. and conjugated to keyhole limpet hemocyanin. The rabbit antiserum was prepared commercially ŽAnaSpec, San Jose, CA.. Specificity

CNS was harvested from mice at P0 Žbirth., 2, 5, 9, 16, 21 and adult Ž8–12 weeks postnatal.. The brain was rapidly dissected on ice into selected regions including; brainstem, cerebellum, midbrain, hippocampus and fronto-parietal cortex. Tissue samples were snap frozen in liquid nitrogen and stored at y808C until use. The tissues were diluted 5% Žwrv. and homogenized in 0.05 M HEPES, 0.01 M EDTA, 0.01 M sodium fluoride, 0.01 M sodium vanadate, 0.03 M sodium pyrophosphate, 100 mgrml aprotinin, 20 mgrml PMSF, 1.0 mgrml leupeptin, and 1% Žvrv. Triton X-100. Protein content was determined using the Bradford method with bovine serum albumin as a standard. The samples were subjected to electrophoresis under denaturing conditions. Briefly, 10 mg of protein was diluted in 0.0625 M Tris–HCl, pH 6.8, 10% Žvrv. glycerol, 2% SDS, 5% b-mercaptoethanol, 0.00125% bromophenol blue Žwrv. and separated on a 12% polyacrylamide gel for 1.5

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

h at 90 V, transferred to a nylon membrane ŽImmobilon-P, Millipore. for 2 h at 100 V in a 25 mM Tris, 192 mM glycine, 20% MeOH Žvrv.. The membrane was blocked for 30 min at room temp in 5% non-fat milk ŽCarnation., incubated overnight at 48C with the anti-Kvb Ž1:100. or anti-Kvb1 Ž1:10. antiserum, washed in TBSrTween, and then incubated with a goat anti-rabbit-HRP conjugate Ž1:1000.. Immune complexes were visualized using ECL ŽPierce.. A standard consisting of the recombinant Kvb2 Ž0.05 mg. was included as a positive control on each of the gels. The bands were visualized by exposing the gels to film ŽFuji Rx. and the autoradiograms developed using a Kodak XOMAT film processor.

3. Results 3.1. Immunocytochemical detection of KÕb expression in the deÕeloping nerÕous system The developmental expression of Kvb protein in discrete regions of the mouse CNS was determined immunocytochemically using an anti-Kvb antiserum directed against a C-terminal peptide that could potentially recognize Kvb1, b2, and b3. The developmental appearance of Kvb reflected a caudal to rostral gradient with expression detected initially in more caudal regions followed by detection in more rostral regions of the CNS. At E16, Kvb expression was detected in the ventral motor neurons of the spinal cord ŽFig. 1A and B. and the DRG Žnot shown., but not in the hippocampus, cerebellum or the cerebral cortex. Neurons expressing Kvb in select brainstem nuclei were first observed at P0 Žnot shown.. At P9, Kvb positive cells in brainstem nuclei display a distribution similar to that found in the adult ŽFig. 1C and D.. The cerebellum displayed an interesting expression pattern with transient Kvb expression observed in the external granule cells at P9 that decreased to undetectable levels at P25, a point at which these cells have migrated to the internal granule layer ŽFig. 1E and F.. Purkinje cells display faint Kvb expression at P9 with immunoreactivity present in the cell body and apical dendrites ŽFig. 1E.. By P25, expression of Kvb approximated that of the adult and was limited to the Purkinje cells and deep cerebellar nuclei. In the adult cerebellum, robust immunoreactivity in the Purkinje cell bodies and faint staining of the apical dendrites was detected ŽFig. 1F.. Kvb expression in the hippocampus was not readily apparent until P21 Ždata not shown.. No specific staining was observed in the hippocampal formation at E16 or P0 ŽFig. 2A and B.. Kvb immunoreactivity was barely detectable in the hippocampal pyramidal neurons at P9 ŽFig. 2C. with robust expression evident in the adult ŽFig. 2D.. Kvb immunostaining was faint in Layer III pyramidal neurons in the cerebral cortex as early as P9 ŽFig. 2E. and the distribution was similar to that of the

73

adult ŽFig. 2F.. The pyramidal cell bodies in the adult cerebral cortex are robustly immunoreactive ŽFig. 2F.. 3.2. Western blot analysis detects multiple proteins that are deÕelopmentally regulated in rodent CNS The postnatal expression of Kvb protein was quantified using Western blot analysis with the anti-Kvb antiserum. Tissues were harvested from various regions of the CNS including the brainstem, the cerebellum, the midbrain, the hippocampus, and the cerebral cortex Žfronto-parietal region.. As predicted from the immunocytochemical data, the developmental appearance of the Kvb subunits follows a caudal to rostral gradient ŽFig. 3.. Kvb expression was detected early postnatally in the most caudal brain regions and attained adult levels earlier than that in the more rostral brain regions. In the brainstem and in the midbrain, a doublet was detected corresponding to previously reported bands that migrate at 39 and 42 kDa w2x. Kvb expression in these regions was readily detectable at birth, reaching adult levels by P16. In the cerebellum, Kvb was expressed at low levels from P0 through P16 and expression reached adult levels by P21. As predicted from the immunocytochemical findings, Kvb was low at birth through P5 in the hippocampus and the cerebral cortex. Readily detectable expression of Kvb was not apparent until P9. Expression in the hippocampus and the cerebral cortex increased appreciably at P16, reaching adult levels. Interestingly, a more slowly migrating species was found in the hippocampus, cerebral cortex, and cerebellum during the first 10 postnatal days. In the hippocampus and the cerebral cortex, this higher molecular weight species was present during the first 9 days postnatal, and thereafter, decreased with age. In the cerebellum, expression of this more slowly migrating species remained relatively constant throughout development. Analysis of overexposed blots ŽFig. 3, P9 lane. revealed that this more slowly migrating species was present in all brain regions examined. While the anti-Kvb antiserum directed against a Cterminal epitope common to Kvb1, Kvb2, and Kvb3, will potentially recognize all of those proteins, given differences in molecular size and localization of the beta subunits, it appears that our anti-Kvb antiserum predominantly recognizes Kvb2. The predicted molecular sizes of the Kvb subunits are 44.7 kDa for Kvb1, 41 kDa for Kvb2, and 45.2 kDa for Kvb3 w1,10,24,25x. Our data suggest that the anti-Kvb antiserum predominantly recognizes Kvb2 since the predominant species observed using this antiserum on a Western blot is smaller than b1 Žsee Figs. 3 and 4. and the predicted size for b3 w10x. In addition, the localization of Kvb3, which is notably absent in the hippocampus but present in the olfactory bulb w10x, differs from that identified with the anti-Kvb antiserum. Thus, the combination of molecular size and protein local-

74

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

Fig. 1. Immunocytochemistry reveals Kvb expression in the developing rodent brainstem and midbrain. Whole embryos were harvested at E16 and brains were harvested at birth through adult, immersion fixed in Bouin’s, paraffin-embedded and sectioned. Sagittal sections were processed for immunohistochemical identification of Kvb using an anti-C-terminal peptide antiserum. The anti-C-terminal antiserum immunolabels cells in the E16 spinal cord ŽA.. A higher magnification depicts ventral motor neurons in E16 spinal cord immunoreactive for Kvb ŽB.. Kvb immunoreactive cell bodies are localized in the brainstem and pons at P9. Large motor neurons in the brainstem, whose localization is compatible with the facial nucleus ŽC. and the trigeminal motor nucleus ŽD., are robustly immunopositive for Kvb at P9. Kvb immunoreactive cerebellar Purkinje cells are depicted at P9 ŽE.. Both the cell bodies and the apical dendrites of Purkinje cells in the adult cerebellum display Kvb immunoreactivity ŽF.. Scale bar s 50 mm.

ization detected using the anti-Kvb antiserum suggests that the major species recognized by the anti-Kvb antiserum is in fact Kvb2. 3.3. Expression of KÕb 1 is differentially regulated in discrete regions of the deÕeloping rodent CNS The antiserum used to detect Kvb is directed against a sequence in the C-terminus that is common to Kvb1,

Kvb2 and Kvb3 and in our in previous studies recognized a predominant 42 kDa band and a minor 39 kDa band w1x, corresponding to the molecular weight for Kvb2 reported by other investigators w25x. In the current study, we detected an additional, more slowly migrating band in the developing brain ŽP9. using the anti-Kvb antiserum following overexposure of a Western blot. We postulated that this band might represent Kvb1 or a posttranslationally modified Kvb2 since a previous report from our lab

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

75

Fig. 2. Immunocytochemical identification of Kvb expression in the developing hippocampus and cerebral cortex. Embryos or postnatal mouse brains were harvested and processed for immunocytochemical detection of Kvb as described in Section 2. Kvb immunoreactivity was not detected in the E16 ŽA. or P0 ŽB. hippocampal formation. Kvb immunoreactivity was barely detectable in the hippocampal pyramidal neurons at P9 ŽC.. In the adult hippocampus, Kvb expression was distributed uniformly in the hippocampal pyramidal cell layer extending from CA1 through CA4 ŽD.. At P9, Kvb protein expression was identified in the Layer III and V pyramidal neurons in the somatosensory cortex ŽE.. Kvb expression is localized to neuronal cell bodies and apical dendrites primarily in layers III and V in the adult neocortex ŽF.. Scale bar s 50 mm.

demonstrated that both Kvb1 and Kvb2 are present in brain w4x. In order to determine whether this band represents Kvb1, we generated rabbit antiserum recognizing a novel sequence in the N-terminus of the Kvb1 protein. Using affinity-purified antiserum that recognizes a unique sequence in the Kvb1 N-terminus, we examined the developmental expression of b1 in mouse brain regions ŽFig. 4.. As predicted from the Western blot data with the anti-Kvb

antiserum, the b1 subunit is differentially expressed during development in the brain regions examined. In the cerebral cortex, Kvb1 is robustly expressed at birth and at P9; a more slowly migrating species is also present in lower abundance in the adult cerebral cortex. In the hippocampus, the expression pattern for Kvb1 mimics that found in the cerebral cortex. The cerebellum displays high levels of a Kvb1 species that is expressed at birth, significantly

76

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

pocampus, but was apparent in P9 pyramidal cells. Kvb1 immunoreactivity was apparent in the P9 Purkinje cells. The expression of Kvb1 in the adult hippocampal pyramidal cells was heterogeneous, the CA1 pyramidal cells were faintly immunoreactive, while the CA2 and 3 pyramidal cells were robustly immunoreactive ŽFig. 5C and D.. As predicted from the Western blot data, immunoreactivity was detected in the early postnatal cerebral cortex where it was localized to neurons in the developing cortical plate

Fig. 3. Western blot analysis demonstrates Kvb expression is developmentally regulated in discrete regions of the early postnatal brain. Mouse brain was dissected into discrete regions at birth, P2, P5, P9, P16, P21 and adult Ž8 weeks. and processed for Western blot analysis as described in the Methods. Protein was extracted from the cerebral cortex ŽCx., hippocampus ŽHp., midbrain ŽMb., cerebellum ŽCbl. and brainstem ŽBrS.. As previously described w1x, a major band migrating at 42 kDa was detected and expression reached adult levels by 3 weeks postnatal in all brain regions examined. At birth, the expression of the 42 kDa species was greatest in the Mb and BrS, while its expression was significantly lower in the Cbl. In the Mb and BrS, expression of the 42 kDa protein increased steadily at P2, 5, and 9 to reach adult levels by P16. In the Cbl, the 42 kDa protein was barely detectable throughout the first 2 weeks postnatal, but displayed a dramatic increase at P21. The 42 kDa species was minimally expressed in the Cx and Hp at birth, but was readily detectable at P9 and continued to increase to adult levels by P21. Interestingly, a more slowly migrating species was also identified at birth in the Cx and Hp. The expression of this species decreased with age in these two brain regions. The slowly migrating species was also identified at birth in the Cbl, and is expressed at a constant low level throughout development and maturation.

decreases by P9, and is absent in the adult. In addition, a more slowly migrating species is present at birth and then increases at P9 and in the adult cerebellum. In the brainstem, expression of Kvb1 is high at birth, present at P9 and decreases in the adult. Similar to other brain regions, only the more slowly migrating species is found in the adult. This form of Kvb1 possibly corresponds to a posttranslationally modified Kvb1. The distribution of Kvb1 was determined in developing and adult mouse CNS ŽFig. 5.. As with the anti-C terminal antiserum, Kvb1 immunoreactivity was detected at E16 in the ventral motor neurons of the spinal cord ŽFig. 5A. and in the dorsal root ganglia ŽFig. 5B.. In the brainstem, Kvb1 immunoreactivity was faint at birth and more robust at P2 where it was localized to cell bodies of cranial nerve nuclei Ždata not shown.. In the adult, expression was localized to large output neurons, i.e., cerebellar Purkinje cells, and the hippocampal and cortical pyramidal cells. Kvb1 expression was not observed at birth in the hip-

Fig. 4. Western blot analysis demonstrates Kvb1 protein is developmentally regulated. Kvb1 expression was determined in mouse brain dissected from discrete brain regions at birth, P9 and in adults using Western blot analysis. Kvb1 expression was detected using a rabbit antiserum raised against a unique 20 amino acid sequence present in the N-terminus of the protein. Total protein Ž10 mg per lane. was separated on a 12% SDS-PAGE, transferred to a nylon membrane ŽImmobilon-P, Millipore., and then processed for immunochemical detection of the Kvb1 via ECL ŽPierce.. The arrow indicates the position of migration of the 46 kDa protein standard. The anti-b1 antiserum detected a predominant band migrating above the 46 kDa standard as well as a minor band that migrated more slowly. At birth, the more rapidly migrating species of Kvb1 was robustly expressed in all of the brain regions examined including; Cx, Hp, Cbl and BrS. At P9, the more rapidly migrating species of Kvb1 was present in the Cx, Hp, and BrS, but had decreased to barely detectable levels in the Cbl. In the adult, this species was dramatically decreased in all of the brain regions examined. However, there was a dramatic increase in expression of the more slowly migrating species of Kvb1. Expression of the slower migrating species was greatest in the BrS followed by the Cbl, the Hp and the Cx. Blots were subsequently re-probed using the rabbit antiserum raised against the 18 C-terminal residues that represent a sequence common to the Kvb1 and Kvb2 proteins Žanti-Kvb .. The more slowly migrating species recognized by the anti-Kvb antiserum migrates at the same position labeled by the anti-b1 antiserum. As illustrated in Fig. 3, this species is faintly detected by anti-Kvb antiserum and shows an expression pattern that is developmentally regulated. This band is not attributed to an artifact representing the presence of bound b1 antiserum following the wash procedure, since previous blots Žsee Fig. 3. probed only with F5 also displayed this band. The predominant 42 kDa band labeled by the anti-Kvb antiserum displays a dramatic increase from low levels at birth to high levels of expression and the adult in each of the four brain regions examined.

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

77

Fig. 5. Immunocytochemical identification of Kvb1 expression in the adult and developing rodent CNS. Kvb1 expression was localized to discrete regions of the adult and developing mouse CNS using an affinity-purified rabbit antiserum raised against a unique 20 amino acid sequence present in the N-terminus of this protein. Overall, the expression pattern is similar to that identified using the anti-C-terminal antiserum. Labeled cells display a neuronal morphology, glial cells were not immunoreactive. At E16, Kvb1 expression was identified in the ventral motor neurons of the developing spinal cord ŽA. and within the DRG ŽB.. Kvb1 immunoreactivity was heterogeneously distributed in pyramidal cells in the adult hippocampus. Pyramidal cells in the CA1 subfield were faintly positive ŽC., while pyramidal cells in the CA3 subfield were robustly immunoreactive ŽD.. Kvb1 expression was detected at P9 in the cell body Žarrows. and apical dendrites Žarrowheads. of pyramidal neurons in the cerebral cortex ŽE., but not in the pyramidal cells Žarrows. in the adult cortex ŽF.. The staining of myelinated fibers in the adult cerebral cortex was determined to be non-specific, since it could not be eliminated by pre-absorption with the immunizing peptide. Scale bar s 50 mm.

and deep cortical layers. The pyramidal cell bodies in the P9 cerebral cortex were weakly positive for Kvb1 ŽFig. 5E.. Consistent with the Western blot analysis demonstrating a decrease in Kvb1 protein in adult cerebral cortex, immunoreactivity was not detected in the pyramidal cell bodies in the adult cerebral cortex ŽFig. 5F.. Myelinated fibers in the cerebral cortex were immunoreactive. How-

ever, this staining was determined to be non-specific since pre-absorption of the primary antibody with the immunizing peptide failed to eliminate staining of these processes. In the adult cerebellum, Kvb1 immunoreactivity was localized to the Purkinje cell bodies and apical dendrites. As with the anti-C-terminal antiserum, the Purkinje cells displayed considerable heterogeneity in staining Žnot shown..

78

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

However, the external granule cell layer cells did not display the transient immunoreactivity as was detected with the anti-Kvb antiserum. In general, the distribution of Kvb1 in the adult mouse CNS was similar to that identified using the anti-Kvb antiserum with localization predominately to large output neurons. However, the frequency of Kvb1 immunoreactive neurons was much lower than that identified with the anti-C terminal antiserum. Thus, these data suggest that Kvb2 is the major subunit expressed in the adult mouse nervous system and that Kvb1 is more discretely localized.

4. Discussion Voltage-gated Kq channel properties and expression patterns are significantly affected by the association of distinct accessory b subunits with the channel forming a subunits w24,25,28,30,34x. Previous investigators have identified accessory subunits that associate with voltagegated Kq channels expressed in the heart, T-lymphocytes and in the CNS w12x. Three b subunits have been identified in the nervous system. The localization of specific channel forming subunits in discrete neuronal populations may underlie the functional diversity of the nervous system. While the Kvb1 subunits generally endow the functional channel with rapidly inactivating properties, the Kvb2 subunit effects are more subtle w28x. For example, the Kvb1.1, 1.2, 1.3 and 3.1 subunits are generally reported to promote rapid closure of Kq channels w24x, while the Kvb2 subunits are thought to modify the voltage dependence of Kv1 channel opening w11,35x. The b2 subunits are also reported to act as chaperones influencing channel assembly and membrane insertion w31x. Taken together with the known functional consequences resulting from the association of b and a subunits of the Kq channels, results reported here suggest novel ways in which the developmental regulation of b subunits may help shape functionality of the CNS. Given the functional diversity imparted by the combination of subunits available to form functional Kq channels it is critical that their distribution in situ be defined. We have previously identified Kvb expression in IL-2 stimulated T-lymphocytes and in neurons in adult CNS w2,6,29x and described the developmental appearance of Kvb2 mRNA using in situ hybridization w1,5x. Neurons were weakly positive at P9 and had attained the characteristic adult distribution at P21 w1,5x. As defined in the current manuscript, the appearance of Kvb immunoreactivity in the developing mouse CNS occurs in a caudal to rostral fashion. Kvb1 was first noted in the ventral motor neurons in the spinal cord at embryonic day 16 and was detected in various brainstem nuclei at birth, but did not reach detectable levels in the more rostral regions of the CNS until the first postnatal week. Kvb expression was first noted in the hippocampal pyramidal cells at P2, and reached adult

levels at P25, at which point the expression pattern mimicked that of the adult. The relatively late appearance of this subunit in the developing nervous system may confer channel properties that are significant in maintaining cell survival in the mature nervous system. The distribution of Kvb1 is unique in that it is expressed at high levels at birth and then its expression is down-regulated in the adult nervous system, in all brain regions examined. Interestingly, we also detected a more slowly migrating species in the adult CNS which possibly reflects an increase in Kvb1 phosphorylation or glycosylation in the mature nervous system. Investigations by Trimmer et al. have defined the distribution of the auxiliary Kvb subunits in adult brain, with regard to their association with Kv a subunits w3,25,26x. Kvb2 predominates in the rodent CNS and it is more widely distributed and associated with Kv1 a subunits than Kvb1 w26x. These investigators reported that Kvb1 and Kvb2 immunoreactivity was localized to the axons and terminal fields of striatal neurons, including the globus pallidus and the pars reticulata of the substantial nigra. In the cerebellar cortex, Kvb2 was found in the axon terminal plexuses of cerebellar basket cells surrounding the initial segment of Purkinje cell axons. Also, Kvb2 was localized in the juxtaparanodal region of myelinated axons in the cerebellar white matter. Kvb2 was not expressed at the basket cell terminals or nodes of Ranvier in the cerebellar white matter. The alpha subunit, Kv2.1, recently identified as the major component comprising the delayed rectifier current in rat hippocampal neurons w21x, does not associate with auxiliary b subunits in the rodent CNS w26x. While the expression of Kvb subunits has been well defined in the adult rodent brain, expression in the developing rodent nervous system is less well defined. Butler et al. w5x localized Kvb mRNA expression in developing mouse brain using in situ hybridization. While these authors report Kvb1 mRNA expression in E16 hippocampus, we did not detect Kvb1 protein expression at this point in development. Consistent with our protein data, they reported a heterogeneous distribution of Kvb1 in adult hippocampal pyramidal neurons. Interestingly, the heterogeneous distribution followed a uniform distribution of Kvb1 mRNA throughout CA field pyramidal cells in early postnatal brain. Our data extends the findings of Butler et al. w5x by defining the developmental expression of Kvb1 protein. A particularly intriguing expression pattern for Kvb has been described in the cerebellum in this manuscript as well as others. In the adult cerebellum, robust expression is apparent in the Purkinje cells and absent in the internal granule cells. The expression in the Purkinje cells was first noted at P9 and reached adult levels by P25. A transient expression of Kvb was detected in the migratory external granule cells at approximately 1 week postnatal. A transient expression of Kvb1 mRNA in the cerebellar granule cells was also described by Butler et al. w5x. These findings

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

are particularly interesting in light of reports describing a mutation in the G-protein coupled inwardly rectifying potassium channel in Weaver mice that alters the pore-forming domain of the protein w22x. These animals display a deficit in cerebellar granule cell migration and synaptogenesis. Previous reports describing the expression of Kv channels in the developing nervous system have primarily focused on describing the appearance of the alpha subunits. Specifically, Hallows and Tempel w9x have shown that Kv1.1 expression in the embryonic rodent CNS displays a complicated temporal pattern with two peaks during embryonic development, followed by an abrupt increase at 2 weeks postnatal that remains constant throughout adulthood. The expression of Kv1.1 in the embryonic hindbrain at particular rhombomeres suggests a potential role in mediating cell migration, and boundary formation w9x. Expression of the a subunit, Kv3.1, is increased in inferior colliculus neurons with the onset of hearing and then remains relatively constant w18x. Elevated potassium increases Kv3.1 mRNA levels and the amplitude of a high-threshold, noninactivating current prior to onset of hearing. This upregulation can be reduced with application of calcium channel blockers suggesting that depolarization and Ca2q influx may alter excitability of immature inferior colliculus neurons by selectively increasing the levels of a Kv3.1-like potassium current w18x. Neural–glial interactions may also mediate Kq channel composition and distribution. In the peripheral nervous system, Schwann cells reportedly mediate Kv distribution and Kq channels significantly affect conduction during remyelination w23x. The distribution of Kv1.1, 1.2 and b2 is reorganized following demyelination and during remyelination in peripheral nerves and inhibiting Schwann cell proliferation prevents the Kq channel redistribution and remyelination w23x. Oligodendrocytes express voltagedependent Kq channels and different channel properties have reportedly been correlated with different stages of oligodendroglial differentiation w32,33x. In a recent study, the Kv1.5 subunit was identified as a major component underlying the delayed rectifier in oligodendrocytes and these channels reportedly mediate oligodendroglial progenitor cell proliferation w3x. An interesting correlation between Kq channel expression and cell cycle regulation is suggested by these data and data reported in non-neural cells. For example, in the murine noncytolytic T lymphocyte clone, L2, voltage-gated potassium channels mediate proliferation in response to IL-2 w17x. The developmental regulation of Kq channel b subunit expression is significant in that it is likely to be involved in shaping electrophysiological properties of developing neurons. Previous work reports that the development of delayed-rectifier function, a process that occurs homogeneously in developing Xenopus spinal cord neurons, is reflected by the expression of different Kva subunits w27x. The developmental expression of b subunits, which exert

79

significant effects on Kq channel kinetics, may result in profound effects on channel functionality, and thus, affect processes governing neuronal development. Modulation of b subunit activity could underlie mechanisms leading to changes in Kq efflux and lead to downstream events including effects on neuronal membrane properties, cell survival, and synaptogenesis in the developing nervous system. Given the wealth of data describing effects of neuronal activity on developmental processes including synapse formation, formation of neural circuits in the visual and somatosensory systems, expression of neurotransmitter markers, as well as neuronal migration, survival and maturation w7,8,14–16,20,36,37x, a thorough understanding of the molecular heterogeneity underlying channel functioning is essential. Our current findings are significant in that they provide evidence that the developmental expression of Kq channel auxiliary subunits provide yet another mechanism for the regulation of channel diversity. Acknowledgements This work was supported by NIMH RO1 MH51327 ŽM.B.P.. and NINDS R29 NS37391 ŽM.D... We thank Sarah Friedman and Yaquing Wu for assistance with immunocytochemistry in the early phases of this project. References w1x M. Arai, J.A. Cohen, Characterization of the neuroimmune protein F5: localization to the dendrites and perikarya of mature neurons and the basal aspect of choroid plexus epithelial cells, J. Neurosci. Res. 36 Ž1993. 305–314. w2x M. Arai, M.B. Prystowsky, J.A. Cohen, Expression of the Tlymphocyte activation gene, F5, by mature neurons, J. Neurosci. Res. 33 Ž1992. 527–537. w3x B. Attali, N. Wang, A. Kolot, A. Sobko, V. Cherepanov, B. Soliven, Characterization of delayed rectifier Kv channels in oligodendrocytes and progenitor cells, J. Neurosci. 17 Ž1997. 8234–8245. w4x M.V. Autieri, S.M. Belkowski, C.S. Constantinescu, J.A. Cohen, M.B. Prystowsky, Lymphocyte-specific inducible expression of potassium channel beta subunits, J. Neuroimmunol. 77 Ž1997. 8–16. w5x D.M. Butler, J.K. Ono, T. Chang, R.E. McCaman, M.E. Barish, Mouse brain potassium channel b1 subunit mRNA: cloning and distribution during development, J. Neurobiol. 34 Ž1998. 135–150. w6x J.A. Cohen, M. Arai, E.L. Prak, S.A. Brooks, L.H. Young, M.B. Prystowsky, Characterization of a novel mRNA expressed by neurons in mature brain, J. Neurosci. Res. 31 Ž1992. 273–384. w7x L.M. Dahm, L.T. Landmesser, The regulation of synaptogenesis during normal development and following activity blockade, J. Neurosci. 11 Ž1991. 238–255. w8x X. Gu, N.C. Spitzer, Distinct aspects of neuronal differentiation encoded by frequency of spontaneous Ca2q transients, Nature 375 Ž1995. 784–787. w9x J.L. Hallows, B.L. Tempel, Expression of Kv1.1, a Shaker-like potassium channel, is temporally regulated in embryonic neurons and glia, J. Neurosci. 18 Ž1998. 5682–5691. w10x S.H. Heinemann, J. Rettig, F. Wunder, O. Pongs, Molecular and functional characterization of a rat brain K v b3 potassium channel subunit, FEBS Lett. 377 Ž1995. 383–389.

80

M. Downen et al.r DeÕelopmental Brain Research 117 (1999) 71–80

w11x S.J. Heinemann, J. Rettig, H.R. Graack, O. Pongs, Functional characterization of Kv channel beta-subunits from rat brain, J. Physiol. 493 Ž1996. 625–633. w12x B. Hille, Ionic Channels of Excitable Membranes, Sinauer, Sunderland, MA, 1992, w13x L.Y. Jan, Y.N. Jan, Cloned potassium channels from eukaryotes and prokaryotes, Annu. Rev. Neurosci. 20 Ž1997. 91–123. w14x S.M. Jones, A.B. Ribera, Overexpression of a potassium channel gene perturbs neural differentiation, J. Neurosci. 14 Ž1994. 2789– 2799. w15x L.C. Katz, C.J. Shatz, Synaptic activity and the construction of cortical circuits, Science 274 Ž1996. 1133–1138. w16x L. Landmesser, G. Pilar, Synaptic transmission and cell death during normal ganglionic development, J. Physiol. 241 Ž1974. 737–749. w17x S.C. Lee, D.E. Sabath, C. Deutsch, M.B. Prystowsky, Increased voltage-gated potassium conductance during interleukin 2-stimulated proliferation of a mouse helper T lymphocyte clone, J. Cell Biol. 102 Ž1986. 1200–1208. w18x A.-Q. Liu, L.K. Kaczmarek, Depolarization selectively increases the expression of the Kv3.1 potassium channel in developing inferior colliculus neurons, J. Neurosci. 18 Ž1998. 8758–8769. w19x M. Maletic-Savatic, N.J. Lenn, J.S. Trimmer, Differential spatiotemporal expression of Kq channel polypeptides in rat hippocampal neurons developing in situ and in vitro, J. Neurosci. 15 Ž1995. 3840–3851. w20x M. Maletic-Savatic, R. Malino, K. Svoboda, Rapid dendritic morphogenesis in CA1 hippocampal dendrites induced by synaptic activity, Science 283 Ž1999. 1923–1926. w21x H. Murakoshi, J.S. Trimmer, Identification of the Kv2.1 Kq channel as a major component of the delayed rectifier Kq current in rat hippocamal neurons, J. Neurosci. 19 Ž1999. 1728–1735. w22x N. Patil, D.R. Cox, D. Bhat, M. Faham, R.M. Myers, A.S. Peterson, A potassium channel mutation in weaver mice implicates membrane excitability in granule cell differentiation, Nat. Genet. 11 Ž1995. 126–129. w23x M.N. Rasband, J.S. Trimmer, T.L. Schwarz, S.R. Levinson, H. Ellisman, M. Schachner, P. Shrager, Potassium channel distribution, clustering, and functions in remyelinateing rat axons, J. Neurosci. 18 Ž1998. 36–47. w24x J. Rettig, S.H. Heinemann, F. Wunder, C. Lorra, D.N. Parcej, J.O. Dolly, O. Pongs, Inactivation properties of voltage-gated Kq channels altered by presence of b-subunit, Nature 369 Ž1994. 289–294. w25x K.J. Rhodes, M.M. Monaghan, N.X. Barrezueta, S. Nawoschik, Z. Bekele-Arcuri, M.F. Matos, K. Nakahira, L.E. Schechter, J.S. Trim-

w26x

w27x

w28x w29x

w30x

w31x

w32x

w33x

w34x w35x

w36x

w37x

mer, Voltage-gated Kq channel b subunits: expression and distribution of Kvb1 and Kvb2 in adult rat brain, J. Neurosci. 16 Ž1996. 4846–4860. K.J. Rhodes, B.W. Strassle, M.M. Monaghan, Z. Bekele-Arcuri, M.F. Matos, J.S. Trimmer, Association and colocalization of the Kvb1 and Kvb 2 b-subunits with Kv1 a-subunits in mammalian brain Kq channel complexes, J. Neurosci. 17 Ž1997. 8246–8258. A.B. Ribera, Homogeneous development of electrical excitability via heterogenous ion channel expression, J. Neurosci. 16 Ž1996. 1123– 1130. B. Robertson, The real life of voltage-gated Kq channels: more than model behavior, TIPS 18 Ž1997. 474–483. D.E. Sabath, P.L. Podolin, P.G. Comber, M.B. Prystowsky, cDNA cloning and characterization of interleukin 2-induced genes in a cloned T helper lymphocyte, J. Biol. Chem. 265 Ž1990. 12671– 12678. S. Sewing, J. Roeper, O. Pongs, Kvb1 subunit binding specific for Shaker-related potassium channel a subunits, Neuron 16 Ž1996. 455–463. G. Shi, K. Nakahira, S. Hammond, K.J. Rhodes, L.E. Schechter, J.S. Trimmer, b subunits promote Kq channel surface expression through effects early in biosynthesis, Neuron 16 Ž1996. 843–852. B. Soliven, S. Szuchet, B.G.W. Arnason, D.J. Nelson, Voltage-gated potassium currents in cultured ovine oligodendrocytes, J. Neurosci. 8 Ž1988. 2131–2141. H. Sontheimer, J. Trotter, M. Schachner, H. Kettenmann, Channel expression correlates with differentiation stage during the deveolopment of oligodendrocytes from their precursor cells in culture, Neuron 2 Ž1989. 1135–1145. J.S. Trimmer, Regulation of ion channel expression by cytoplasmic subunits, Curr. Opin. Neurobiol. 8 Ž1998. 370–374. V.N. Uebele, S.K. England, A. Chaudhary, M.M. Tamkun, D.J. Snyders, Functional differences in Kv1.5 currents expressed in mammalian cell lines are due to the presence of endogenous Kv beta 2.1 subunits, J. Biol. Chem. 271 Ž1996. 2406–2412. R.-L. Wu, D.M. Butler, M.E. Barish, Potassium current development and its linkage to membrane expansion during growth of cultured embryonic mouse hippocampal neurons: sensitivity to inhibitors of phosphatidylinositol 3-Kinase and other protein kinases, J. Neurosci. 18 Ž1998. 6261–6278. S.P. Yu, C.-H. Yehg, S.L. Sensi, B.J. Gwag, L.M.T. Canzoniero, Z.S. Farhangrazi, H.S. Ying, M. Tian, L.L. Dugan, D.W. Choi, Mediation of neuronal apoptosis by enhancement of outward potassium current, Science 278 Ž1997. 114–117.