|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online July 12, 2005 as doi:10.1096/fj.05-3743fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||








Sections of
* Psychiatry and
Alcohol and Drug Dependence Research, Department of Clinical Neuroscience, Karolinska Institute, Stockholm, Sweden
2 Correspondence: Psychiatry Section, R5:00, Department of Clinical Neuroscience, Karolinska Hospital, Stockholm 17176, Sweden. E-mail: Yasmin.hurd{at}cns.ki.se
SPECIFIC AIMS
Dynorphins are endogenous opioid peptides that are critically involved in addiction, pain, neurodegenerative, and psychiatric disorders. As a first step in the elucidation of the role of the prodynorphin gene regulation in these diseases we aimed to characterize PDYN mRNA and protein products in the adult human brain and functions of protein products of this gene in model cell lines.
PRINCIPAL FINDINGS
1. Six novel PDYN mRNAs identified in the adult human brain
The canonical form of PDYN pre-mRNA consisting of 3 introns and 4 exons coding for the full-length protein (FL1-PDYN mRNA) has previously been identified. Other than FL1, little is known about PDYN transcripts, their promoters and protein products in the adult human brain. Two approaches were used to characterize PDYN transcripts: poly(dA) based rapid amplification of cDNA ends (RACE) which detects all mRNAs; and RNA-ligase mediated RACE (RLM-RACE) which selectively detects 5'-capped mRNAs. Apart from the known mRNA coding for FL1, a second FL form (FL2) with a novel first exon (2') was detected as well as 3 new 5'-truncated transcripts T1T3, a splice variant Sp1 in the coding region of exon 4, which contains the dynorphin-encoding sequences, and splice variant Sp2, which lacks exons 2 and 3 (Fig. 1
).
|
2. Distinct regional expression of full-length (FL1 and FL2) PDYN mRNAs
PDYN mRNA expression was studied in the adult human brain using in situ hybridization histochemistry with riboprobes specific for the 5'-end of the FL1- and FL2-PDYN mRNAs and for exon 4, which should recognize both FL- and T-PDYN mRNAs. The FL1-mRNA showed a classic PDYN mRNA expression pattern with labeling predominantly in limbic-related structures such as the ventral striatum (nucleus accumbens), dorsal striatum patch compartment, accessory basal and cortical amygdala, dentate gyrus of the hippocampus, and entorhinal cortex (Fig. 2
; left panel). In contrast, FL2-mRNA was more limited in its distribution with selective expression predominantly in, for example, the claustrum and supraoptic hypothalamus (Fig. 2
; center panel). Even when present in the same brain structure (e.g., basal nucleus of Meynert and hypothalamus) the two PDYN mRNA forms were differentially expressed in discrete subnuclei (Fig. 2)
. The exon 4 probe resulted in a hybridization pattern that was generally a combination of the FL1 and FL2 expression. The FLs are also expressed in the midgestational human fetal brain where they are processed into dynorphins, suggesting a role of dynorphins in synaptic transmission at early stages of development of the central nervous system.
|
3. Full-length PDYN protein
PDYN-derived opioid peptides have been extensively characterized, but the presence and properties of the FL-PDYN have not yet been reported in the human or animal brain. To detect PDYN, rabbit polyclonal antibodies against human PDYN C-terminal fragment (CTF241-254) were generated and characterized by Western blotting. The antibodies detected a single band in COS1 cells transfected with pCMV-h-FL-PDYN plasmid. The apparent 28 kDa protein molecular size coincided to the calculated size of FL-PDYN. High levels of the FL-PDYN were detected in the nucleus accumbens, whereas the signal was weaker in the amygdala and virtually absent in the cerebellum. A substantial fraction, 5070%, of FL-PDYN was present in brain tissues in Triton x100-insoluble fraction. Thus, PDYN is not immediately processed into opioid peptides, but is apparently stored in or transported through the secretory pathways as the FL protein.
4. Translation of FL1 and FL2 PDYN mRNAs and processing of their protein products
To test whether FL1- and FL2-mRNAs are functional, their peptide products were measured in the caudate nucleus and claustrum. PDYN mRNA (in situ hybridzation), PDYN protein (Western blot analysis) and its processing product dynorphin B (radioimmunoassay, RIA) were detected in both structures with 2-fold higher levels in the caudate nucleus compared with the claustrum. Thus, both FL1 and FL2 mRNAs are translated into PDYN, which is apparently processed into opioid peptides in both brain regions.
5. Regulatory functions of the first FL1 and FL2 exons
To test whether the first FL exon (exon 1 for FL1 and exon 2' for FL2 (Fig. 1)
are involved in the regulation of the PDYN gene expression, their sequences were inserted into a pEGFPN1 plasmid between a CMV promoter and the EGFP protein coding sequence. The plasmids were transfected into neuroblastoma SH-SY5Y cells and the expression of the EGFP reporter gene was analyzed. Western blot analysis demonstrated that insertion of the FL1 exon 1 results in a 2-fold increase in GFP expression. Strong, 10-fold inhibition of expression was observed when the FL2 exon 2' fragment was incorporated. Thus expression of FL1- and FL2-mRNAs may be regulated through exons 1 and 2'.
6. Translation of Sp1 PDYN mRNA and processing of its protein product
To test whether the splice variant Sp1 mRNA is translated and the Sp1 protein is processed into peptides, Sp1 cDNA under control of the CMV promoter was transfected into rat insulinoma RINm-5F cells. Protein and peptide products were analyzed by Western blot and by RIA for dynorphin B. These experiments demonstrated that Sp1-mRNA is translated into a protein product that is processed into dynorphin B and dynorphin A and that these peptides are secreted into the culture medium. The efficiency of opioid peptide generation from Sp1 protein was, however,
4-fold lower than that of FL-PDYN, suggesting a critical role of the deleted segment in the processing of FL1 precursor. Low levels of PDYN of 16 kDa that coincided with the calculated size of Sp1-PDYN were identified by Western blot analysis, suggesting rapid processing of this protein into dynorphins.
7. Truncated PDYN proteins
To test whether truncated PDYN mRNA can be translated, T1- and T2-PDYN cDNAs under control of the CMV promoter were transfected into COS1 cells. Peptide fractions isolated from cell extracts and enriched on C18-SEP-PAK columns were analyzed by Western blot with anti-h-PDYN CTF antibodies. Bands with apparent molecular size of 12 and 6 kDa, respectively, coinciding with the calculated molecular size of T1- and T2-PDYN, were detected in cells transfected with expression plasmids but not with vector plasmid. The T1- and T2-PDYN levels were much lower than that of FL-PDYN expressed from the same CMV promoter, suggesting low efficiency of translation of T1- and T2-mRNAs or rapid degradation of the mRNAs or protein products. COS1 cells transfected with CMV-h-FL-, CMV-h-T1- and CMV-h-T2-PDYN plasmids were labeled with anti-h-PDYN CTF antibodies to characterize PDYN localization in cells. FL-PDYN demonstrated a pattern characteristic for localization in the endoplasmic reticulum. T1-PDYN was primarily localized in the cell nuclei, whereas the T2 protein was seen exclusively in the cytoplasm of transfected cells. This indicates that the nuclear localization signal is situated between the Met146 and Met198 residues in the N-terminal segment of the T1-PDYN that is absent in T2-PDYN.
8. Promoter activity of exon 4
Transcription of the T1-T3 PDYN mRNAs may be initiated by an intragenic promoter located in exon 4 and adjacent sequences. Promoter activity of exon 4 was tested with pEx4-EGFPN1 plasmid containing exon 4 located upstream of the EGFP gene in a promoterless (CMV) construct. Approximately 10% of the cells demonstrated bright EGFP fluorescence when COS1 cells were transfected with pEGFPN1 (+CMV) plasmid as a positive control. Bright fluorescent cells were observed in cultures transfected with pEx4-EGFPN1 plasmid, but not with promoterless pEGFPN1(CMV) plasmid. Thus, exon 4 has promoter activity that may contribute to the intragenic initiation of transcription of truncated PDYN mRNA.
CONCLUSIONS AND SIGNIFICANCE
Our study demonstrates that the PDYN gene has a complex pattern of transcription in the adult human brain where seven mRNA types with different 5'-ends are detected. The findings strongly imply that there is a high level of plasticity of PDYN expression on cellular and regional levels of organization. The unexpected observations that T1-PDYN lacks a signal peptide and is located in the cell nucleus suggest non-opioid functions of this protein. The marked difference throughout the human brain in the regional distribution pattern of the FL-PDYN transcripts implies distinct involvement of the FL1 and FL2 transcripts in reward, memory, motor, and endocrine functions. The differential anatomical expression pattern of the FL mRNAs is even evident in discrete neuronal populations within the same brain structure. For example, in the hypothalamus the FL1-mRNA was present in the lateral nucleus, suggesting that the FL1-PDYN-derived peptides would be involved in, for example, feeding and the regulation of sleep. The FL2-transcript was instead expressed in the supraoptic nucleus where the dendritic release of dynorphins is known to regulate neurosecretion of oxytocin and vasopressin. The brain region-specific distribution of FL1- and FL2-PDYN mRNAs suggests that their expression is controlled by different promoters. Our findings reveal that the first exon is indeed critical for differential regulation of these transcripts and thus may control regional specificity of their expression.
In contrast to the human PDYN gene, only one or two PDYN transcripts which code for the FL proteins have been identified in other species such as the mouse, rat, guinea pig, and amphibia. Most exons are strongly conserved in the human, mouse, and rat genomes. However, exons that are only included in minor alternative splice forms (as opposed to the constitutive or major transcript form) are frequently characterized by their recent creation. The expression of minor human PDYN mRNA forms that are either spliced or contain new exons and that apparently are absent in other species, supports the notion of an association of alternative splicing in the human genome with recent evolutionary changes. Impairment of the complex PDYN regulatory mechanisms at the levels of gene transcription and translation may ultimately define human pathological conditions such as chronic pain, substance dependence, depression, and epilepsy. It remains to be determined whether the differential expression and distribution of the PDYN transcripts are specifically altered in these pathological conditions.
FOOTNOTES
1 These authors contributed equally to this work. ![]()
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-3743fje; doi: 10.1096/fj.05-3743fje
This article has been cited by other articles:
![]() |
A. Nikoshkov, K. Drakenberg, X. Wang, M. Cs. Horvath, E. Keller, and Y. L. Hurd Opioid neuropeptide genotypes in relation to heroin abuse: Dopamine tone contributes to reversed mesolimbic proenkephalin expression PNAS, January 15, 2008; 105(2): 786 - 791. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Drakenberg, A. Nikoshkov, M. C. Horvath, P. Fagergren, A. Gharibyan, K. Saarelainen, S. Rahman, I. Nylander, G. Bakalkin, J. Rajs, et al. From the Cover: {micro} Opioid receptor A118G polymorphism in association with striatal opioid neuropeptide gene expression in heroin abusers PNAS, May 16, 2006; 103(20): 7883 - 7888. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |