| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Laboratory of Neurochemistry, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Harold Gainer, Ph.D., Chief, Laboratory of Neurochemistry, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Building 36, Room 4D20, Bethesda, Maryland 20892. E-mail: gainerh{at}ninds.nih.gov.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thus, although the OT and VP MCNs are similar in many respects, there are significant phenotypic differences between them, and one of our goals is to understand the molecular bases for these differences. The most well-established phenotypic difference is the expression of the OT or VP genes, and efforts to understand the regulation of this cell-specific expression are an ongoing effort in several laboratories (reviewed in Ref. 8). Additional phenotypic differences include the expression of other peptides that are differentially expressed in OT or VP cells. For example, galanin and dynorphin are preferentially expressed in VP cells, whereas corticotrophin-releasing factor and cholecystokinin are preferentially expressed in OT cells (8, 12). There are also physiological differences, such as in firing patterns between the OT and VP cells. OT cells display fast continuous firing, whereas phasic bursting activity is characteristic of VP cells, and OT and VP cells can be selectively stimulated by various physiological stimuli or by an array of pharmacologic agents (reviewed in Refs. 2 and 13).
Our goal was to isolate genes that are differentially expressed in OT- vs. VP-producing MCNs. In an effort to begin to understand the molecular basis for the phenotypic differences between OT and VP MCNs, we used differential hybridization screening of OT and VP single-cell cDNA libraries derived from identified MCNs in the supraoptic nucleus (SON) of the lactating rat. We chose to study the lactating rat because it is known that, under this natural physiological condition, there is a profound up-regulation of expression of both peptide genes (7, 8, 9, 11). This approach resulted in the isolation of 48 genes heretofore unknown to be present in the MCNs; and among these, 6 genes involved in disparate cell biological functions were differentially expressed in the OT and VP MCNs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
shows an overview of our differential hybridization screening strategy.
|
3 Prime, Inc., Boulder, CO), 324 U/ml RNAguard (Amersham Biosciences Corp.), and 0.5 mM each of deoxy (d)ATP, dCTP, dGTP, and deoxythymidine 5'-triphosphate] and used immediately for RT or stored at -80 C. The cells were lysed for 1 min at 65 C. One hundred units of Superscript II reverse transcriptase (Life Technologies, Inc.) was then added, and the reaction mixture was incubated for 15 min at 37 C. The reverse transcriptase was heat-inactivated for 10 min at 65 C, and poly(A) was added to the first-strand cDNA product by adding an equal volume of 200 mM potassium cacodylate (pH 7.2), 4 mM CoCl2, 0.4 mM dithiothreitol, and 200 µM dATP containing 10 U terminal transferase (Roche Molecular Biochemicals, Indianapolis, IN) for 15 min at 37 C. The reactants were heat-inactivated for 10 min at 65 C, and 0.5-µl aliquots from each tube were used to determine the phenotypes of the OT and/or VP cells by gene-specific PCR amplification.
Identification of OT and VP MCNs by PCR.
The amplification of the OT and VP cDNAs was performed as previously described (15), with a few modifications. A 50-µl hot-start PCR mixture containing 25 µl of a lower reaction mix composed of 1x PCR buffer F [Invitrogen (Carlsbad, CA) PCR optimization kit], 125 µM of each deoxynucleotide triphosphate, and 0.5 µM of each primer (OT-F, 5'-GAC GGT GGA TCT CGG ACT GAA-3'; OT-R, 5'-CGC CCC TAA AGG TAT CAT CAC AAA-3'; VP-F, 5'-CCT CAC CTC TGC CTG CTA CTT-3'; VP-R, 5'-GGG GGC GAT GGC TCA GTA GAC-3') was used. A wax bead was added to the reaction mix, melted for 5 min at 80 C, then allowed to harden at 4 C, forming a barrier on top of the lower reaction mix. The upper reaction mix contained 1x buffer F, single-cell cDNA template, and 2.5 U AmpliTaq DNA polymerase (Perkin-Elmer Corp.). Amplification was performed as follows: 40 cycles at 94 C for 45 sec, 62 C for 45 sec, and 72 C for 2 min. After cycling, the reaction was extended at 72 C for 7 min. Other amplification reactions were performed as follows: aliquot single-cell cDNA template was amplified in 20-µl reactions containing 1x PCR buffer (Perkin-Elmer Corp.), 125 µM of each deoxynucleotide triphosphate, 2.5 pM of each primer [glyceraldehyde 3-phosphate dehydrogenase (GAPDH)-F, 5'-CCT GCA CCA CCA ACT GCT TAGC-3'; GAPDH-R, 5'-GAG TTG CTG TTG AAG TCA CAGG-3'], and 2.5 U AmpliTaq Gold (Perkin-Elmer Corp.). Amplification was performed as follows: 1 cycle at 94 C for 10 min; 40 cycles at 94 C for 30 sec, 58 C for 30 sec, and 72 C for 2 min. After cycling, the reaction was extended at 72 C for 20 min. Five microliters from the total reactions was run on 1.5% agarose gels containing ethidium bromide (EtBr) and digitally photographed with an Eagle Eye camera (Stratagene, La Jolla, CA).
PCR amplification of cDNAs from single identified OT and VP MCNs.
The volume of each tube containing the single MCN cDNAs was brought to 100 µl with a PCR mixture composed of 10 mM Tris-HCl (pH 8.3); 50 mM KCl; 2.5 mM MgCl2; 100 µg/ml BSA; 0.05% Triton X-100; 1 mM dATP, dCTP, dGTP, and deoxythymidine 5'-triphosphate; 10 U AmpliTaq DNA polymerase, and 5 µg PCR primer AL1 (5'-ATT GGA TCC AGG CCG CTC TGG ACA AAA TAT GAA TTC (T)18-3'). PCR amplification was performed according to the following schedule: 94 C for 1 min, 42 C for 2 min, and 72 C for 6 min, with 10 sec extension per cycle, for 25 cycles. Five units of AmpliTaq DNA polymerase were then added, and the PCR amplification was continued for an additional 25 cycles. The success of the single-cell RT-PCR amplification procedure was determined by analyzing 5-µl aliquots of the reactions on a 1.5% agarose/EtBr gel. Successful amplifications were indicated by an intense smear of cDNA running between 400-2000 bp. The OT vs. VP phenotyping using the gene-specific primers (see above) was repeated using these amplified cDNAs to confirm the identities of the cell subtypes before proceeding to the next step.
cDNA library construction.
Before cDNA library construction, amplified cDNA was pooled from 10 identified OT neurons and separately from 10 identified VP neurons. Figure 2 illustrates typical PCR results for the whole-cell RT-PCR, using the AL1 primer, as well as typical results of cell phenotyping PCR, using VP-, OT-, and GAPDH-specific primers. The OT and VP cDNA pools were then digested with EcoRI and run on a 1.5% agarose/EtBr gel. cDNAs with sizes ranging from 400-1600 bp were isolated, purified, and then ligated into the EcoRI site of the pZErO-2 vector (Invitrogen). Escherichia coli DH10B ELECTROMAX cells (Life Technologies, Inc.) were transformed by electroporation with this ligation mix and plated at approximately 1000 colonies per 150-mm plate.
|
Hybridization was at 65 C, overnight, in hybridization solution containing 6x saline sodium citrate (SSC), 0.5% sodium dodecyl sulfate (SDS), 5x Denhardt’s reagent, 100 µg/ml denatured herring sperm DNA, and 5 x 106 cpm/ml of the amplified cDNA probe. The final washing conditions were: twice in 0.2x SSC and 0.1% SDS for 30 min at 65 C. Hybridization was detected by autoradiography after 3 d. Colonies that showed differential hybridization intensities were picked and grown. These clones were then checked for differential expression by slot blot hybridization analysis. Clones that seemed to be differentially expressed by slot blot analysis were further examined by Southern blot hybridization analysis using the same probes as those used for the differential hybridization screening. The slot blot and Southern blot hybridization analyses were performed according to standard procedures (17). The cDNA clones that seemed to be differentially expressed by Southern blot analysis were identified by DNA sequencing, performed by the NINDS DNA Sequencing Faculty (Bethesda, MD). Sequences were analyzed for similarity by using the basic local alignment search tool (BLAST) program accessed at http://www.ncbi.nlm.nih.gov/blast/.
In situ hybridization histochemistry (ISHH)
Two different ISHH methods were used in this study. In the first method, RNA probes from the differentially expressed clones were prepared as 35S-labeled probes using [
-35S]-uridine 5'-triphosphate (UTP) (NEN Life Science Products, Boston, MA). The plasmids were linearized with EcoRV or NotI for SP6 riboprobe synthesis or with HindIII, BamHI, or SpeI for T7 riboprobe synthesis using the MAXIscript in vitro transcription kit (Ambion, Inc., Austin, TX). Serial 12-µm brain and whole-embryo [embryonic d 21 (E21)] sections were cut on a cryostat microtome (2800 Frigocut E, Leica Corp., Deerfield, IL) and placed onto slides coated with 0.5% gelatin and 0.05% chromium potassium sulfate, dried on a slide warmer for 10–30 min at 40 C, and then stored at -80 C. Before hybridization, the sections were fixed in 4% formaldehyde solution for 15 min at room temperature, put into 0.1% triethanolamine-HCl (pH 8.0) containing 0.25% acetic anhydride for 10 min at room temperature, rinsed with 0.2x SSC buffer, transferred through graded ethanol, and then air-dried. Hybridization was carried out in 100 µl hybridization buffer (10 mM Tris-HCl (pH 7.4), 50% formamide, 100 µg/ml denatured salmon sperm DNA, 200 µg/ml yeast tRNA, 5x Denhardt’s reagent, 10% dextran sulfate, 0.6 M NaCl, 0.25% SDS, and 1 mM EDTA (pH 8.0) containing 106 cpm denatured 35S-labeled RNA probe. After overnight hybridization at 55 C, the sections were washed in 4x SSC three times and incubated with TNE buffer [10 mM Tris-HCl (pH 7.6), 0.5 M NaCl, and 1 mM EDTA] containing 20 µg/ml ribonuclease A for 30 min at 37 C, then washed in 2x, 1x, and 0.5x SSC, with two final washes in 0.1x SSC at 65 C. The sections were transferred through graded ethanol, then air-dried. The sections were placed apposed to BIOMAX MR film (Eastman Kodak Co., Rochester, NY) and exposed for 1–3 d at -80 C.
In the second method, digoxigenin (DIG)-labeled OT or VP antisense riboprobes were used for double-labeled ISHH (18). DIG labeling was done per the manufacturer’s instructions (Roche Molecular Biochemicals). For double-labeling experiments, 106 cpm 35S-labeled cDNA probe was mixed with 100 ng DIG-labeled riboprobe per 100 µl of hybridization solution. After overnight hybridization at 55 C, the sections were washed, processed for the DIG-labeled probe, and apposed to a Eastman Kodak Co. BIOMAX MR film overnight to estimate the hybridization intensity. The slides were then coated with Ilford K.5D nuclear emulsion (Ilford Scientific Product, Paramus, NJ) and developed after 1–5 d of exposure. The slides were mounted with Cytoseal 60 (Electron Microscopy Sciences, Fort Washington, PA) mount medium and coverslipped.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. These results are consistent with the numbers of OT, VP, and hybrid OT/VP cells previously found in the SONs of lactating rats by single-cell gene-specific RT-PCR, immunohistochemical, and ISHH analyses (15, 19). Because only 1–5% of the total cellular cDNA sample was necessary for an unequivocal assessment of their phenotypes, the remaining cDNA could be used to produce single-cell cDNA libraries for differential hybridization screening.
To generate enough material to construct and differentially screen cDNA libraries from identified OT and VP MCNs, we used a modified version of a previously described single-cell RT-PCR-based screening protocol (16). We generated microgram quantities of cDNA by PCR amplifying oligo-dT primed cDNA that was isolated from individual OT or VP MCNs. To average out the individual variations in peptide and other protein coexistences known to exist in the OT and VP MCNs (15), equal aliquots of amplified cDNA from 10 OT cells (and separately, from 10 VP cells) were pooled. Cellular RNAs that were from the OT/VP phenotype (e.g. cells 4 and 9 in Fig. 2) were systematically excluded from these libraries. The pooled cDNAs were then used to generate cDNA libraries in the plasmid vector pZErO-2.
To obtain OT cell-specific cDNA clones, the OT cell cDNA library was screened for colonies that hybridized with probes derived from the OT cells but did not, or only weakly, hybridized to probes derived from the VP cells. Duplicate filter sets of approximately 10,000 recombinant colonies from the OT cell cDNA library were hybridized with probes derived from VP and OT cell cDNAs, respectively. After autoradiography, the hybridized probe was stripped from the filters, and the filters were again hybridized using the opposite probes. Only those colonies that showed differential hybridization on both rounds were picked for further analysis. For the OT cell library, 319 colonies were picked. The same process was used to identify differentially hybridizing VP cell clones, resulting in 329 colonies being picked (Table 1, step 1). Because the assessment of differential hybridization from colony lifts can be unreliable, we also used slot blot hybridization to confirm differentially hybridizing clones (Table 1, step 2). Of the 319 OT colonies, only 133 could be confirmed to be differentially hybridized by slot blot analysis. Likewise, for the 329 VP colonies, only 139 were confirmed by differential hybridization on slot blots (Table 1, step 2).
|
, step 3). Similarly, for the VP clones, 53 VP clones are differentially expressed by Southern blot analysis, compared with 139 indicated by slot blot analysis (Table 1, step 3). These results emphasize the importance of using the slot blot and Southern blot hybridization analyses to confirm the original colony hybridization results before proceeding to DNA sequencing and the more labor-intensive procedures of ISHH.
Identification of OT and VP MCN cell-specific gene candidates by DNA sequencing
All of the cDNA clones that showed differential hybridization with the OT and VP MCN-specific probes by Southern blot analysis were then further characterized by DNA sequencing (Table 1, steps 4–6). Sequencing showed that of the 45 OT cell clones, 19 (42%) of them represented OT precursor protein mRNAs. The remaining 26 cDNA clones represented 18 other genes (Table 2). Of the 53 VP cell clones, 10 (19%) represented VP precursor protein mRNA, 12 (23%) represented neuronatin (Nnat) mRNA, and the remaining 31 clones represented 29 other genes (Table 3). It is very reassuring that no VP clones were found in the OT MCN cDNA library, and no OT clones were obtained from the VP MCN cDNA library (Table 1, steps 4 and 5), thereby providing evidence for the cell-specificity of the cDNA libraries.
|
|
) and 27 different VP cDNAs (Table 3), were selected for further analysis by ISHH.
Determination of gene expression patterns by ISHH
A preliminary single-label ISHH analysis was performed for each putative differentially expressed clone (other than the putative OT and VP precursor clones) to determine the level and the tissue specificity of expression exhibited by each. For each clone, [35S]-UTP-labeled riboprobes were synthesized from the entire insert, varying in size from 160–690 bp, in the pZErO-2 vector. Hybridization with the sense strand probe was used to control for nonspecific signals.
Several issues were considered in the ISHH analysis. The first was to determine which clones could be detected in the hypothalamus and how specific the expression was to the HNS. For this purpose, we used coronal hypothalamic frozen sections taken from lactating rats through the level of the HNS. Table 2 summarizes these data for the 45 OT cell clones, and Table 3 shows 53 VP cell clones that showed differential expression by Southern blot analysis. Most of theses clones could be detected in the HNS by ISHH, although the overall hybridization patterns in the other regions of the hypothalamus greatly varied between the clones (not illustrated, but see selected examples in Figs. 3 and 4). The second issue was whether any of these clones hybridized to mRNA in other regions of the brain and other tissues. For this assay, we used frozen parasagittal sections through a rat embryo at E21. Using this preparation, we could assay the hybridization of the clones over a wide region of the body in a single section. Virtually all but one clone (see Fig. 3B) also hybridized to other regions of the nervous system and other tissues in the body; however, the overall specific expression patterns greatly varied for each clone (not illustrated, but see selected examples in Figs. 3 and 4). The one clone that was expressed exclusively in the HNS, VP35S (Fig. 3B), turned out to be expressed selectively in VP MCNs, and its DNA sequence was 270 bp upstream of the transcription start site for 3'-phosphoadenosine 5'-phosphosulfate synthase 2 (PAPSS2) (see Table 4).
|
|
|
for four of the differentially expressed hybridizing clones listed in Table 4.
|
, step 7; and Table 2). Likewise, of the 53 VP cell clones that showed differential expression by Southern blot analysis, 27 (51%) clones were differentially expressed [10 VP, 12 Nnat, 2 metallothionein III/growth inhibitory factor (Mt3/GIF), and 3 other clones], whereas 9 (17%) clones were expressed equally in OT and VP neurons (Table 1, step 7; and Table 3). Overall, these results indicate that this differential screening strategy was effective in isolating differentially expressed cDNAs, albeit with a low efficiency.
Six clones differentially hybridized to OT or VP neurons, and these and their BLAST search identifications are summarized in Table 4. Among these, clone OT no. 129, identified as phosphofructokinase C (PFK-C), was the only differentially expressed clone (other than the OT precursor protein) to be isolated from the OT MCN cDNA library. PFK-C mRNA is highly expressed in the central nervous system (CNS) of the E21 rat embryo (Fig. 3A). In the adult rat brain, PFK-C mRNA is very intense in several nuclei of the hypothalamus, including the SON and the PVN (Fig. 3A). In the SON, double-label ISHH shows that PFK-C is highly expressed in OT-producing MCNs, with much lower levels of expression in VP-containing cells (Fig. 5, A and B).
Clone VP no. 35S, tentatively identified as PAPSS2-related mRNA, was not detected in any tissue in the rat embryo (Fig. 3B). In the adult rat brain, VP no. 35S is only found in the SON, PVN, and various accessory nuclei of the HNS (Fig. 3B). In the SON, VP no. 35S is primarily expressed in VP-containing MCNs, with very little expression in OT cells (Fig. 5, G and H). Clone VP no. 79, synaptotagmin V (Syt5), was primarily expressed in the CNS of the rat embryo (Fig. 3C). In the adult rat brain, Syt5 is heavily expressed in the hippocampus, SON, PVN, and some parts of the cortex (Fig. 3C). In the SON, Syt5 mRNA is primarily found in VP-containing MCNs, with very little expression in OT cells (not illustrated). Clone VP no. 121, Mt3/GIF, was primarily expressed in the CNS of the rat embryo but also exhibited other regions of lighter expression outside of the CNS, particularly in the liver (Fig. 3D). In the adult rat brain, Mt3/GIF is intensely expressed throughout the brain, with heavier expression in the hippocampus, SON, and the cortex (Fig. 3D). In the SON, Mt3/GIF is preferentially found in VP-containing MCNs, with very little expression in OT cells (not illustrated).
Clone VP no. 34, identified as paternally expressed gene (Peg)5/Nnat (20), was the most abundant clone isolated from the VP MCN cDNA library. Peg5/Nnat is heavily expressed in the CNS of the rat embryo, although it is also intensely expressed in various locations outside of the CNS (Fig. 4A). In the adult brain, it is very strongly expressed in the ventral part of the hypothalamus, obscuring the labeled HNS nuclei on the x-ray film (Fig. 4A). With higher resolution ISSH, the VP no. 34 is clearly observed in the SON primarily in VP-containing MCNs, with very little expression in OT cells (Fig. 5, C and D). Clone VP no. 130L, identified as Peg3, was expressed generally throughout the rat embryo (particularly intensely in the tongue), and with high levels of expression in the CNS (Fig. 4B). In the adult rat brain Peg3 is intense in the SON and PVN, with much lower levels of expression in the rest of the brain (Fig. 4B). In the SON, Peg3 is predominantly expressed in VP-containing MCNs, with much less expression in OT cells (Fig. 5, E and F).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
Using the protocol described in Fig. 1, we successfully constructed separate cDNA libraries for the OT and VP MCN phenotypes. Each library was derived from 10 identified OT or VP MCN cDNAs, which were pooled to better reflect the most stable features of the phenotypes (15). After completion of the differential screening of these cDNA libraries, we sequenced all the differentially expressed clones. Clones representing OT and VP mRNAs were obtained only from the OT or VP cDNA libraries, respectively (Table 1, steps 4 and 5), thereby validating the specificities of the libraries. The preliminary functional characterizations of the other clones that were obtained, based on Southern blot analysis and DNA sequences, are shown in Tables 2 and 3. Of these, we found that 6 clones were preferentially expressed in the OT or VP MCNs (Table 4).
Genes differentially expressed in OT and VP MCNs
None of the clones that we isolated in this differential screen would have been predicted a priori to be expressed, let alone differentially expressed in OT or VP cells. However, the fact that these genes are differentially expressed in the MCNs may provide clues to mechanisms in these cells that had not previously been considered. With this in mind, each of the six differentially expressed genes that we have isolated in this study is briefly discussed below.
PFK-C.
Our finding that PFK-C mRNA is preferentially found in OT-MCNs, as opposed to VP MCNs (Fig 5, A and B), suggests that the regulation of glucose metabolism might differ between these two cell-types. PFK catalyzes the ATP-dependent conversion of fructose 6-phosphate to fructose 1, 6-bisphosphate, which is the rate-limiting reaction in glycolysis. Three PFK subunit isozymes have been identified (PFK-C, PFK-L, and PFK-M), and each isozyme subunit has distinct catalytic properties and specific sensitivity to hormonal regulators (25, 26, 27, 28). The specific catalytic properties of the holoenzyme depend on the specific isozyme subunit composition of the tetramer (29, 30). PFK-C is predominantly located in the brain and in neuroendocrine tissues (27). PFK-M and PFK-L are also located in the brain (28), resulting in a potentially complex mixture of homo- and heterotetramers. In view of the complexity of these enzymes’ regulation (27, 28, 31), additional studies on the expression patterns of the other isozyme subunits present in the MCNs will be necessary to consider the possible tetrameric arrangements that might be present in these cells, and their potential metabolic consequences.
PAPSS2-related mRNA.
In this study, we isolated a cDNA, VP no. 35S, whose mRNA is completely restricted to the supraoptic, paraventricular, and accessory nuclei of the HNS (Fig. 3B). No other region in the CNS and no other tissue in the embryo expressed this gene (Fig. 3B). In addition, VP no. 35S mRNA is preferentially expressed in VP MCNs vs. OT MCNs in the SON (Table 4; and Fig. 5, G and H). VP no. 35S has significant nucleotide sequence similarity to the 5' upstream region of the mouse PAPSS2 gene. PAPS synthetase 2(PAPSS2), catalyzes the sulfation of various proteins, carbohydrates, and lipids that are involved in a variety of biological processes (32). The identification of mutations in the PAPSS2 gene as the underlying cause of orthologous inherited chondrodysplasias (33, 34, 35) and brachymorphisms (36) in mice has recently highlighted the critical role of the PAPSS2 isoform in the tissue-specific regulation of sulfation in vivo. Similar skeletal abnormalities were observed in human spondyloepimetaphyseal dysplasia, a dwarfing condition caused by a stop codon in the human PAPSS2 gene (37).
Although there is significant sequence similarity between VP35S and PAPSS2 mRNA, this identification is not consistent with the published distribution pattern of PAPSS2 mRNA and that which we observed for VP35S (Fig. 3B). Mouse PAPSS2 mRNA is detected in liver and brain (36), and human PAPSS2 mRNA is highly expressed in liver, but is lower in brain (38). This discrepancy could be explained by assuming differentially expressed isoforms of PAPSS2 that are generated by alternative splicing (36), and that the isoform homologous to the VP35S sequence is specifically expressed in the HNS, whereas the other isoforms are more generally distributed (38, 39, 40, 41, 42). Closer examination of the published PAPSS2 mRNA sequence raises additional questions. Though a BLAST search with the VP35S sequence against the nonredundant GenBank database resulted in an 87% identity match with a mouse PAPSS2 gene (AF172857), the VP35S sequence represents genomic sequence 582–270 nucleotides upstream of the PAPSS2 translation start site. A complete cDNA sequence has been reported for PAPSS2, which starts 130 nucleotides upstream of the translation start site (36). Because there is no overlap between the VP35S sequence and the reported PAPSS2 cDNA sequence, it is possible that VP35S is not part of PAPSS2 mRNA, even though it lies very close to the start of PAPSS2 transcription. The VP35S sequence lies right in the middle of the proposed PAPSS2 promoter sequence (40); and though it is possible that VP35S represents an alternatively spliced isoform of PAPSS2, it also could be a completely independent transcript. To distinguish between these possibilities, it will be necessary to clone the complete cDNA of VP35S by 3' and 5' rapid amplification of cDNA ends.
Syt V.
Syts are part of a large gene family of proteins that are characterized by a transmembrane domain and two C2 domains (43). The C2 domains are believed to mediate Ca2+-dependent and Ca2+-independent interactions with target molecules, and Syts have been implicated in the Ca2+-dependent regulation of secretory vesicle fusion with the plasma membrane. In nerve cells, they are abundant synaptic vesicle transmembrane proteins that may also function as Ca2+ sensors. Syt V mRNA expression is largely brain-specific (44) and is in the same Ca2+ binding class as Syts I and II (45, 46), which have been suggested as requiring high calcium concentrations for regulated secretion processes (43). Antiserum specific for the Syt V isoform recognizes a protein of about 50 kDa that is about 6-fold more abundant in brain synaptic vesicles than in whole brain membranes but is also found in nonneuronal tissues (47).
We show that Syt V mRNA is preferentially expressed in VP cells of the SON. The differential expression of Syt isoforms in VP neurons vs. OT neurons is interesting, in that it suggests that there may be alternative molecular mechanisms regulating membrane fusion of synaptic vesicles in these two closely related peptidergic cell types.
Mt3/GIF.
Mt3/GIF, a brain-specific member of the metallothionein protein family, is a small, heat-stable protein, which may be an important regulator of copper and zinc in the nervous system (48, 49, 50). Mt3/GIF is drastically reduced and down-regulated in Alzheimer’s disease brains, and its absence has been implicated in the development of Alzheimer’s disease. Additionally, Mt3/GIF may play an important role in tissue repair after adult brain injury and in preventing neuronal sprouting (51, 52, 53, 54). Various studies suggest that Mt3/GIF plays an important role in protecting neurons from ischemic insult and seizure activity by reducing neurotoxic zinc levels (55, 56, 57). Vasak and Hasler (50), in their recent general review on metallothioneins, conclude that, despite extensive study, their primary functions remain enigmatic, and the significance of the preferential expression of Mt3/GIF mRNA in VP MCNs over OT MCNs is equally so.
Peg5/Nnat.
Peg5/Nnat is a proteolipid-like protein that is homologous to PMP-1, an H+-ATPase subunit, and phospholamban, a Ca2+ ATPase subunit, and it has been suggested as being involved in the regulation of ions during brain development (58, 59). Nnat was first identified as a gene strongly expressed in neonatal rat brain, with a subsequent decreased expression in juveniles and adults (60). The gene structure is very conserved, consisting of three exons giving rise to primarily two alternatively spliced mRNAs and proteins of 54 and 81 kDa (20, 61, 62, 63, 64). Peg5/Nnat is a paternally expressed, imprinted gene that maps to distal chromosome 2 in an imprinting region where mutations can lead to developmental abnormalities (20, 65, 66). The early expression of Peg5/Nnat in the brain suggests that it is involved in early development (60, 64, 65). An extensive in situ hybridization study (67) in the mouse showed that there was strong segmental expression of Nnat in the embryonic hindbrain (E8.5) and in Rathkes pouch and the primordial anterior pituitary, and these authors suggest that this protein could be involved in the differentiation of anterior pituitary cells. In this regard, it is interesting that Nnat is also expressed in other endocrine cells, such as pancreatic ß-cells (68) and in thyroid tumors (58), where its transcription is up-regulated by thyroid hormone. Nnat is also abundant in PC-12 cells when undifferentiated, but it is greatly down-regulated when NGF is applied to induce differentiation.
The role of Peg5/Nnat in the adult brain has not been actively studied, and very little is known about it. Although reputed to be less expressed than at earlier stages, it is strongly expressed in the ventral aspect of the hypothalamus (see Fig. 4A), and specifically in the VP-MCNs in the SON (see Fig. 5, C and D). The biological significance of the selective expression of Peg5/Nnat in VP-MCNs is not known at present, but it is intriguing to note that the VP and OT genes, like the Peg5/Nnat gene, are localized to chromosome 2 in mice.
Peg3.
Peg3 is a paternally expressed, imprinted gene encoding a Kruppel-type zinc finger-containing protein that may be involved in DNA-binding and transcriptional regulation. High levels of Peg3 are detected in the brain of humans and rodents, suggesting an important and conserved role for this protein in neuronal cells (69, 70). Transgenic mice carrying a null mutation in the Peg3 gene resulted in growth retardation, as well as striking impairment of maternal behavior that frequently resulted in death of the offspring (71, 72). This result may be partly attributable to defective neuronal connectivity, as well as to a reduced number of OT neurons in the hypothalamus, because mutant mothers were deficient in milk ejection. Interestingly, we found that Peg3 mRNA is preferentially expressed in VP- as opposed to OT-MCNs (Fig. 5, E and F). Unfortunately, the biochemical and physiological status of VP neurons in the Peg3 mutant mice has not been studied.
Other experiments show that Peg3 (also referred to as Pw1) is involved with the TNF-NF
B signal transduction pathway (73). TNF signaling can mediate a wide range of biological activities, including cell proliferation and differentiation and programmed cell death. For example, TNF can enhance cell survival through a Peg3-TNF receptor-associated factor-mediated activation of NFkB, but TNF can also trigger cell death through a Peg3-c-myc mediated apoptosis pathway. Recently, Relaix et al. (74) have shown that Peg3 cooperates with another gene, Siah 1a, to induce p53-mediated apoptosis, and that inhibiting Peg3 activity blocks p53-induced apoptosis. Peg3 also seems to have a tumor suppression function, because its overexpression in glioma cells inhibits their tumorigenicity in nude mice (75), and down-regulation of Peg3 is found in glioma cells (76).
Conclusions
In recent years, studies of gene expression profiles in single neurons under various biological conditions (77) have provided new opportunities to examine the phenotypic diversity of cells in a complex organ, such as the brain (78). In a number of cases, cDNA libraries have been constructed from single neurons (16, 79, 80), and differential screening of cDNA libraries constructed from single olfactory neurons has led to the identification of pheromone receptors in mammals (16). The magnocellular OT and VP neurons are exceptional candidates for such a differential analysis because of their large size, their compact localization in the CNS, and their high rates of transcription.
In this study, we used a single-cell differential hybridization strategy to isolate genes that are preferentially expressed in VP cells vs. OT cells. The motivation for these experiments was to provide a starting point for the analysis of the molecular basis for the differences between VP and OT MCNs. We used PCR amplification of the mRNA from single cells to generate sufficient cDNA to produce cDNA libraries. However, PCR amplification is known to significantly distort the representations of the mRNAs that are present in the cell; and hence, this represents a significant problem for the differential hybridization screen. Although this differential screening strategy was relatively successful in the identification of previously unsuspected and differentially expressed genes in the single identified OT and VP MCNs, it is clearly inefficient. Many already-known differentially coexpressed neuropeptide genes (8, 9, 12, 23) were missed. One possible way to avoid the distortion of PCR amplification in differential screens and cDNA microarray analyses would be to develop and apply a less biased, linear method (77, 81, 82) for the amplification of mRNAs isolated from the single MCNs.
The physiological significance of the six novel, cell-specific expressed genes in the MCNs that we identified by this differential screen (summarized in Table 4) remains to be determined. All of these genes have been described in the database; however, none of this information, at present, is useful in understanding why VP- or OT-MCNs should express them so preferentially. Several of these genes are intriguing candidates for further study. Two of these, Peg3 and Peg5, are paternally expressed, imprinted genes. Genomic imprinting refers to the epigenetic marking of genes that results in the exclusive expression of the maternal or paternal allele (83, 84). Imprinted genes are intricately involved in fetal and behavioral development, and knockout experiments suggest their involvement in reproductive and maternal behavior. In view of the known influences of OT and VP on maternal and other reproduction-related behaviors, it is therefore interesting that two paternally imprinted genes, Peg3 and Peg5, are preferentially expressed in the VP cell. Peg3 is a particularly interesting candidate for further study in the HNS because it is a Kruppel-type zinc finger protein and, hence, might be involved in regulating selective gene expression in VP-MCNs, possibly that of the VP-gene itself.
VP35S is unique among all of the clones that we isolated because it is exclusively expressed in the MCNs (Fig. 3B). We have tentatively associated VP35S with its nearby downstream encoded gene, PAPSS2, but note that the available data does not conclusively demonstrate that the VP35S sequence is actually part of the PAPSS2 transcript. It is possible that this sequence is part of another transcript and that its exclusive cell-specific expression in VP MCNs suggests a fundamental role in the maintenance of the VP MCN phenotype.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Present address: Department of Neurobiology and Pharmacology, Northeastern Ohio Universities College of Medicine, Rootstown, Ohio 44272.
3 Present address: Department Psychiatry and Behavioral Neurosciences, McMaster University, HSC-4N78, 1200 Main Street West, Hamilton, Ontario, L8N-3Z5, Canada.
Abbreviations: BLAST, Basic local alignment search tool; CNS, central nervous system; d, deoxy; DIG, digoxigenin; E21, embryonic d 21; EtBr, ethidium bromide; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; HNS, hypothalamo-neurohypophysial system; MCN, magnocellular neuron; Mt3/GIF, metallothionein III/growth inhibitory factor; ISHH, in situ hybridization histochemistry; Nnat, neuronatin; OT, oxytocin; PAPSS2, 3'-phosphoadenosine 5'-phosphosulfate synthase 2; Peg, paternally expressed gene; PFK-C, phosphofructokinase C; SDS, sodium dodecyl sulfate; SON, supraoptic nucleus; SSC, saline sodium citrate; Syt, synaptotagmin; UTP, uridine 5'-triphosphate; VP, vasopressin.
Received May 14, 2002.
Accepted for publication July 17, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and ß-domains of growth inhibitory factor (GIF) on neuron injuries caused by oxygen free radicals. Biochim Biophys Acta 1568:129–134[Medline]
