Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on October 14, 2003

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 12/23/3181    most recent ddg345v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (29) Request Permissions Google Scholar Articles by Rezniczek, G. A. Articles by Wiche, G. Search for Related Content PubMed PubMed Citation Articles by Rezniczek, G. A. Articles by Wiche, G. Social Bookmarking          What's this?

Human Molecular Genetics, 2003, Vol. 12, No. 23 3181-3194
DOI: 10.1093/hmg/ddg345
© 2003 Oxford University Press

Plectin 5'-transcript diversity: short alternative sequences determine stability of gene products, initiation of translation and subcellular localization of isoforms

Günther A. Rezniczek, Christina Abrahamsberg, Peter Fuchs, Daniel Spazierer and Gerhard Wiche^*

Institute of Biochemistry and Molecular Cell Biology, University of Vienna, Vienna Biocenter, Dr. Bohrgasse 9, A-1030 Vienna, Austria

Received August 29, 2003; Accepted October 6, 2003

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Plectin is a large cytoskeletal linker protein expressed as several different isoforms from a highly complex gene. This transcript diversity is mainly caused by short 5'-sequences contained in alternative first exons. To elucidate the influence of these sequence differences and to determine potential differential functionality of the resulting protein forms, we conducted a systematic investigation of plectin isoforms on transcript and protein levels. Isoform expression was highly dependent on the different 5' ends, largely due to effects of the 5'-untranslated regions. Initiation of translation downstream of the expected start site led to loss of actin- and integrin ß4-binding in some isoforms. The small alternative N-terminal sequences (5–180 residues) profoundly affected the subcelluar localization of this >500 kDa protein. Specifically, plectin 1f was concentrated at focal adhesion contacts and plectin 1b was exclusively targeted to mitochondria, providing a connection of these organelles to intermediate filaments. Thus, with plectin as a model, we demonstrate a role for 5'-untranslated regions and alternative 5'-splicing as an important regulatory mechanism of protein expression and protein function.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
The ability of eukaryotic cells to adopt a variety of shapes and to carry out coordinated and directed movement depends on the cytoskeleton, a complex network of protein filaments and associated proteins that extends throughout the cytoplasm. It is responsible for such diverse tasks as the positioning and trafficking of vesicles, organelles and proteins, the alignment and segregation of chromosomes during mitosis, providing mechanical strength as well as dynamic properties to cells, including migration, muscle contraction and change of shape. In recent years, the cooperation of the different cytoskeletal networks, actin, intermediate filaments (IFs) and microtubules, to perform many of their tasks has become apparent. This interplay relies on an important family of structurally and, in part, functionally related proteins referred to as plakins or cytolinkers (1–3), capable of interlinking different elements of the cytoskeleton. Plectin mediates network formation of IFs, the interlinking of IFs with microtubules and microfilaments, as well as the anchoring of IFs to the plasma and nuclear membranes. A novel important role of plectin as regulator of actin cytoskeleton dynamics has recently been demonstrated (4). There is growing evidence for an essential role of plectin in epithelial, muscle and nervous tissues. Plectin has been localized at Z-lines of striated muscle, dense plaques of smooth muscle and intercalated discs of cardiac muscle, at the basal cell surface membranes of stratified epithelia and in peripheral areas of cells in all cell layers (5). Direct interactions of plectin with integrin ß4 in hemidesmosomes (6,7) and with desmoplakin (8) have been demonstrated. Plectin knock-out mice exhibit severe skin blistering combined with a reduction in the number of hemidesmosomes, as well as abnormalities in heart and skeletal muscles (9). Similar disorders were found in patients suffering from epidermolysis bullosa simplex (EBS) with muscular dystrophy, an autosomal recessive disease caused by mutations in the plectin gene (reviewed in 10). A dominant form of the disease, EBS-Ogna, leads to skin changes in the absence of muscular symptoms (11). Analysis of the genomic exon–intron organization of the murine plectin gene, which contains well over 40 exons and spans over 32 kb on chromosome 15, revealed an unusual 5' transcript complexity of plectin isoforms (12). In total, 16 alternatively spliced exons were identified, 11 of them (1–1j) directly splicing into a common exon 2, which is the first exon encoding a highly conserved actin binding domain (ABD), three (-1, 0a, 0) into exon 1c and two additional exons (2 and 3) are optionally spliced within the exons encoding the ABD. In human and rat, thus far four alternatively spliced first exons have been described (13,14). Furthermore, isoforms lacking exon 31 (encoding the rod domain) were identified in rat (14) and mouse (unpublished data). A putative rodless human plectin variant, as judged by molecular size, was identified on the protein level (15).

In this work, we present a systematic investigation of the role of the alternatively spliced first plectin exons using cDNAs (full-length and truncated forms) resembling all identified isoforms. Transcript and protein products were assessed regarding their expression levels and localization. Distinct from previous studies conducted so far, we have included the 5'-untranslated regions (5'-UTRs) of plectin in our analysis and investigated their influence on isoform expression. We show that all identified variants of plectin transcripts gave rise to proteins of expected sizes, and that in the case of isoforms with non-coding first exons, translation initiation unexpectedly did not occur at the first in-frame ATG, but at a further downstream ATG contained in exon 6. Quantitative comparison of isoform expression on protein and transcript levels in fibroblasts and keratinocytes proposed a role for the different 5'-ends in cell-type specific regulation of expression and protein stability. Finally, fluorescence microscopy of transfected cells revealed distinct subcellular localization of plectin isoforms.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Features of their 5'-UTRs divide alternative plectin transcripts into three classes
To assess the influence of the alternative first plectin exons with their, in some cases, highly complex 5'-UTRs on isoform expression, we cloned full-length and truncated plectin cDNAs, based on a previous detailed analysis of the murine plectin gene (12). Detailed analyses of the 5'-ends of plectin isoform transcripts are shown in Figure 1. The contexts of plectin start ATGs (sATGs) and upstream ATGs (uATGs) were analyzed using software tools predicting translation initiation sites, NetStart (16) and AUG^pr (17), and by comparison to the Kozak consensus sequence. Based upon their 5'-UTRs, plectin mRNAs can be divided into three classes. (i) Translation starts with the first ATG contained in the sequence. This sATG is compatible with the plectin reading frame, and no uATGs are present. The majority of plectin isoforms falls into this category [Ple 1, 1a, (0)1c, 1d(2), 1e, 1f and 1g]. (ii) Several uATGs precede the sATG [Ple 1b and (-1/0a)1c]. As in class (i), translation starts with the first ATG compatible with the plectin reading frame. (iii) This class comprises the isoforms with non-coding first exons [Ple 1h, 1i, 1j(23)]. Here, translation initiation does not occur at the first plectin-compatible ATG (exon 4), but at an ATG in exon 6 (see below). Due to this downstream shift of initiation, the 5'-UTRs of this class of transcripts are long and complex.

View larger version (37K): [in this window] [in a new window]   Figure 1. Schematics and analysis of plectin isoform 5'-regions of cDNA sequences are represented by light gray bars, exon boundaries are indicated by short vertical lines above the bars and exon numbers are displayed on top; exons numbered in bold contain start codons compatible with the plectin reading frame. Positions of ATGs are indicated by vertical lines extending downward from the light gray bars and the respective ORFs are represented by dark gray bars (black bars for the plectin ORF). Asterisks mark the ATGs actually used for translation initiation. 1)Length of the first exons, including 5'-UTRs and, in the case of plectin 1c, also non-coding exons upstream of exon 1c (exon 0, 107 nt; exon -1, 26 nt; exon 0a, 816 nt). 2)Number of nucleotides upstream of the start ATG. In the case of non-coding first exons (1h–j) the 5'-UTR includes sequences up to the start codon in exon 6. 3)G/C content of the 5'-UTR. 4)Number of ATGs upstream of the actually used start codon. 5)Deduced molecular masses of full-length (f.l.) plectin isoforms, excluding any tags. 6)Molecular masses of plectin ABD isoform truncations, including EGFP (29.0 kDa). 7)Consecutive numbering of ATGs, starting with the one closest to the 5' end of the sequences. The ATG actually used for initiation is shown in bold letters. 8)Positions are relative to the 5' end of the cDNA sequences (12). 9)ORFs are numbered consecutively. Overlapping ORFs in the same reading frame have the same numbers; the plectin ORF is indicated by the letter ‘P’. 10)These columns contain values representing probabilities for translation initiation at the respective ATGs and were calculated using different methods (see text for more details). Higher numbers mean higher probability. In the case of NetStart, a plus sign indicates that this ATG is predicted to serve as initiation site.

 
Unexpected translation initiation at a downstream ATG in the case of isoform transcripts with non-coding first exons
Generally, when uATGs were present, ATGs in the context of the plectin reading frame always had higher predicted initiation scores and the actual sATG in first coding exons always had a score higher than that of the ATG in exon 4. In the case of the isoforms with non-coding first exons, predictions were strongly in favor of the exon 4-ATG with NetStart scores of 0.683–0.776 compared to 0.576 for the first ATG in exon 6 (Fig. 1). However, when Ple 1h-8, Ple 1i-8, or Ple 1j(23)-8 were expressed in mouse fibroblasts, keratinocytes or N2A cells, a single protein species of molecular mass smaller than expected (42 kDa instead of the predicted 49.3 kDa for the EGFP-fusion protein starting with the ATG in exon 4) was observed. Further, to exclude the possibility that initiation occurred at the second ATG in exon 6, we constructed two mutants where either the first (ATG1) or the second ATG (ATG2) in exon 6 was eliminated (Fig. 2A). While the wild-type and the ATG2 mutant expressed normally in transfected mouse fibroblasts, the mutant lacking the first exon 6-ATG failed to express (Fig. 2B). This was in agreement with the NetStart prediction, where this ATG had a score of 0.403 and was not predicted to serve as an initiation site.

View larger version (38K): [in this window] [in a new window]   Figure 2. Plectin isoforms with non-coding first exons lack the first CH domain and lose integrin ß4-binding. (A) Part of the plectin 1h transcript is shown as a schematic (see legend to Fig. 1). The deduced amino acid sequence for exon 6 is shown in detail below this schematic, and the corresponding protein part is shown above. Initiation of translation in exon 6 completely eliminates the first of the two CH domains in plectin's ABD; start in exon 4 would preserve half of it (indicated by levels of gray). The methionine (M) to valine (V) amino acid changes (A to G nucleotide exchanges introduced by splice overlap extension PCR) which eliminated one or the other putative start methionine are emphasized (dark gray boxes and ellipsoids). (B) Wild-type and mutant EGFP-fusion protein expression constructs were transfected into mouse fibroblasts. After 24 h, cells were lysed in SDS-sample buffer and lysates subjected to SDS–15%-PAGE and subsequent immunoblotting using a monoclonal anti-EGFP antibody. No expressed protein was detectable when the first exon 6-ATG was mutated (ATG1) while expression was not affected when the second ATG (ATG2) was mutated. Molecular masses are indicated. (C,D) Binding of plectin to integrin ß4 was assessed using an in vitro binding assay. (C) Eu3+-labeled his-tagged integrin ß4 cytoplasmic tail (ß4-F1,2LF3,4C) was overlaid onto microtiter plate-immobilized MBP-fusions (100 nM) of plectin ABD fragments (MBP-E2-8, closed circles; MBP-E4-8, open circles; MBP-E6-8, closed triangles) or MBP alone (negative control, open triangles). (D) Eu3+-labeled MBP (negative control, open circles) or MBP-E2-8 (closed circles) were overlaid onto integrin ß4 (100 nM). Scatchard transformation of the binding data is shown in the inset (apparent Kd=0.17±0.05 µM). All values shown represent mean±SD of triplicate determinations.

 
Loss of integrin ß4-binding in isoforms with non-coding first exons
The complete second and parts of the first CH domain have been shown to be necessary for binding of plectin to actin (12). We assessed the binding of plectin's ABD to integrin ß4 in an in vitro overlay assay. MBP-fusions of the complete (MBP-E2-8) and truncated ABD versions (MBP-E4-8, MPB-E6-8) were coated onto microtiter plates, along with MBP alone (negative control) and overlaid with Eu³⁺-labeled, his-tagged cytoplasmic domain of integrin ß4 (ß4-F_1,2LF_3,4C). ß4-F_1,2LF_3,4C specifically bound equally well to MBP-E2-8 and MPB-E4-8, while binding to MBP-E6-8 was comparable with that of the negative control (Fig. 2C). Binding did not show saturation, due to the previously described self-interaction of integrin ß4 (6). When ß4-F_1,2LF_3,4C was coated and overlaid with soluble Eu³⁺-labeled plectin, MBP-E2-8 bound to coated integrin ß4 in a saturable manner (Fig. 2D). Scatchard transformation of the binding data gave an apparent K_d of 0.17±0.05 µM (r²>0.95). This high affinity binding was in the same range as determined for plectin-binding to actin (0.32±0.06 µM) (4). These results mirrored those obtained for the plectin ABD–actin interaction. In both cases, complete lack of the first CH domain entirely abolished binding, while removal of only half of CH1 did not affect the binding characteristics. In fact, even enhanced integrin ß4-binding of an ABD version containing the extra five amino acids encoded by exon 2 was observed (data not shown). Further in vitro binding studies and cotransfection experiments in mammalian cells revealed that binding of integrin ß4 and actin to plectin's ABD was mutually exclusive (data not shown).

5' sequences determine stability of plectin mRNAs and proteins
Plectin isoform transcripts are differentially expressed in different tissues (12). To assess the influence of the alternative first exons and their 5'-UTRs on protein expression, we coexpressed short EGFP-tagged plectin versions together with EGFP alone in mouse fibroblasts and keratinocytes. This experimental system facilitated quantification and normalization of the data obtained by densitometry of immunoblots since plectin ABD isoforms and normalization control could be detected simultaneously. In fibroblasts, a significant 5'-UTR influence was only detected in the case of Ple 1c-8 and isoforms with non-coding first exons (Fig. 3A). While exon 0 increased the expression of Ple 1c-8, exons -1/0a lowered the expression level. A reduction to about half was also observed in case of Ple 1h-8, 1i-8, and 1j(23)-8, compared to levels of expression from a construct lacking 5'-UTRs (Ple 4-8, see also below). Similar observations were made in keratinocytes (Fig. 3B), except that here Ple 1f-8 and 1g-8 were most prominently expressed, while in fibroblasts Ple (0)1c-8 showed highest levels. Quantification of endogenous plectin isoforms (12) suggested that plectin (0)1c was predominantly expressed in brain, while plectin (-1/0a)1c was specific for skin. In line with this, we found that the exon 0-form was expressed 3.5-fold compared to the -1/0a-form in mouse neuroblastoma (N2A) cells, while only about 2-fold in epithelial-derived cell lines (PtK2, mouse keratinocytes) or mouse fibroblasts (Fig. 3C). Since the effects of mRNA stability, translation efficiency and protein stability were superimposed in this approach (transcription was equalized due to the same promoter used in all expression vectors), we quantified mRNA levels next. The presence of endogenous plectin mRNAs even in plectin (-/-) cells (4 and unpublished data) disallowed simultaneous detection of ectopically expressed transcripts and the normalization control, as was possible on the protein level. Therefore, we used a vector with two independent expression cassettes, one of them expressing myc-tagged plectin constructs, the other EGFP. Transcripts were detected simultaneously using probes specific for the myc-tag or EGFP (the EGFP-signal was used for data normalization). In fibroblasts, little variability was detected. Compared to the other isoforms, Ple (-1/0a)1c-8 levels were found reduced, and Ple 1f-8 levels increased (Fig. 3D). In keratinocytes, variability was somewhat greater (Fig. 3E). Ple (-1/0a)1c-8 levels were also low, indicating that exons -1/0a might destabilize these transcripts in both cell types. Stability of Ple 1e-8 and of transcripts containing the non-coding first exons also appeared to be negatively affected in keratinocytes. When protein levels were adjusted for transcript amounts (eliminating effects of mRNA stability), cell-type dependent differences became more apparent (Fig. 3H and I). Interestingly, the results obtained for plectin 1c implied that translation of this particular isoform was equally efficient from both transcripts, despite the extremely long 5'-UTR of Ple (-1/0a)1c-8. Ple 1b-8 as well as the isoforms with non-coding first exons seemed to suffer from low translation efficiency, possibly due to their 5'-UTRs containing several uATGs. However, this could also be due to lower protein stability. Further transformation of the data allowed direct per-isoform comparison of expression levels (Fig. 3J and K). Interestingly, it appeared that although more Ple 1a-8, Ple 1b-8, Ple (0)1c-8, Ple (-1/0a)1c-8 and Ple 1d(2)-8 transcripts were found in keratinocytes, these cells contained less protein. Thus, these proteins were more stable or more efficiently translated in fibroblasts. The opposite was the case for Ple 1e-8 and a relative increase of Ple 1g-8 protein was found in keratinocytes.

View larger version (63K): [in this window] [in a new window]   Figure 3. Quantification of plectin isoform expression. (A,B) Expression plasmids encoding plectin ABD isoforms or truncations with an EGFP-tag at their C-termini, either lacking (gray bars) or including (black bars) the 5'-UTR, were co-transfected into mouse fibroblasts (A) or mouse keratinocytes (B). In the case of exon 1 g, no expression construct lacking the 5'-UTR was available. Ple 4–8 served as a reference for isoforms with non-coding first exons (1h, 1i, 1j). Note that different first exons affect expression levels and/or stability of the resulting proteins in a particular cell type, with the 5'-UTR having a considerable influence. (C) EGFP-fusions of Ple (0)1c-8 (black bars) and Ple (-1/0a)1c-8 (gray bars) were expressed in mouse neuroblastoma (N2A), PtK2, mouse fibroblast (mFib) and mouse keratinocyte (mKer) cells. Stronger protein expression of 1.7 to 3.5-fold was observed for Ple (0)1c-8. For quantitative measurement, cell lysates were subjected to SDS–12%-PAGE and immunoblot analysis using anti-GFP antibodies. Blots were evaluated densitometrically, and data normalized using signals obtained from a cotransfected control (pEGFP-N2). Data shown represent means±SD of at least three transfection experiments. (D,E) myc-tagged plectin ABD isoforms were expressed in mouse fibroblasts and mouse keratinocytes from plasmids containing plectin cDNAs including 5'-UTRs. EGFP was expressed from an independent expression cassette on the same plasmids and was used for normalization. Quantitative data were obtained by RNase protection assays using EGFP- and myc-specific probes. (F,G) Protein data from (A) and (B) is repeated here to facilitate comparison with mRNA data. (H,I) Approximate measure of protein stability based on normalization of protein expression for transcript content. (J,K) Comparison of expression levels of plectin ABD isoforms in different cell types. Expression data of plectin ABD isoforms in mouse fibroblasts (gray bars) and mouse keratinocytes (black bars) were normalized based on EGFP-control expression levels and scaled to 100% for fibroblasts, allowing for a direct comparison of individual isoforms between the two cell types on protein (J) and mRNA (K) levels.

 
5'-UTRs enhance fidelity of translation
Exon 1 is unique among the first coding exon in that it is the longest such exon and also contains several ATGs in addition to the sATG (Fig. 4A). While AUG^pr-prediction favored the first ATG, the other ATGs in the plectin frame were not ruled out by NetStart, and some had excellent Kozak-scores. When Ple 1-8 was expressed in mouse fibroblasts, a single protein species of 76 kDa was expressed from a construct containing the 5'-UTR, while with absence of this sequence several smaller bands of varying intensities were additionally observed (Fig. 4B). The double-band just below 66 kDa was probably due to initiation at exon 1-ATGs 9 and 11, and the very faint band at 42 kDa due to initiation within exon 6. A similarly improved expression was observed for other first coding exons, e.g. Ple (0)1c-8 and Ple 1f-8 (Fig. 4C).

View larger version (54K): [in this window] [in a new window]   Figure 4. Immunoblot analysis of EGFP-tagged plectin proteins expressed in mouse fibroblasts. (A–C) Enhanced expression from cDNAs containing 5'-UTRs. (A) Analysis of ATGs contained in exon 1. The contexts of the 11 ATGs contained in exon 1 were analyzed and predictive scores for their likelihood to serve as translation initiation sites were calculated (see legend to Fig. 1). 1)Sequential numbering of ATGs starting at the 5' end. The ATG actually used for initiation is shown in bold letters. 2)Positions are relative to the 5' end of the sequence. 3)ORFs are numbered consecutively; the plectin ORF is indicated by the letter ‘P’. 4)These columns contain values representing probabilities for translation initiation at the respective ATGs and were calculated using different methods (see text for more details). Higher numbers mean higher probability. In case of NetStart, a plus sign indicates that this ATG is predicted to serve as initiation site. 5)Molecular masses of expected proteins, including the EGFP-tag (29.0 kDa). (B,C) Expression from plasmids lacking (left lanes) or including the 5'-UTR (right lanes). Plectin ABD isoforms were separated by SDS–12%-PAGE, and their C-terminal EGFP-tags were detected using GFP-specific antibodies. In most cases, the 5'-UTR affected both quantity and quality of expression. Enhanced quality was especially evident for the cases shown [Ple 1-8, 1c(0)-8, and 1f-8]. In the case of Ple 1-8, the presence of the 5'-UTR completely eliminated smaller sized proteins, which were probably caused by unspecific initiation evens at downstream ATGs (B). In the cases of other overexpressed plectin ABD isoforms, some initiation in exon 6 was observed, resulting in a 42 kDa band; in the presence of 5'-UTRs, this was minimized, while total protein expression was increased (C). (D,E) Analysis of lysates of mouse fibroblasts transfected with EGFP-fusion protein expression plasmids using anti-GFP monoclonal antibodies (SDS–12%-PAGE). (D) In all cases, translation is initiated exclusively (Ple 4-8) or partially (Ple 2-8, Ple 6-8) downstream of the first ATG in the expression constructs (in Ple 6-8, an additional ATG was introduced through cloning, causing the double-band). (E) Plectin proteins encoded by expression plasmids containing exons 2–8 preceded by all first exons thus far identified (with non-coding exons 1h–1j the same protein is expressed, so only Ple 1h-8 is shown) were expressed with the expected sizes; Ple 1h-8 exhibited a molecular mass compatible with translation initiation within exon 6. (F) Expression of intact full-length plectin isoforms in mouse fibroblasts. Lysates of plectin-deficient mouse fibroblasts transfected with plasmids encoding all eleven full-length plectin isoforms as EGFP-fusion proteins were separated by SDS–5%-PAGE and immunoblots analyzed using anti-serum to plectin (no. 9) or anti-GFP antibodies (not shown). Plectin from a rat glioma C6 cell IF preparation served as positive control and size marker (IF). A lysate from untransfected fibroblasts was used as a negative control (C). Proteins of expected molecular masses were expressed from all isoform plasmids, including those with non-coding first exons (Ple 1h, 1i, 1j), albeit considerably weaker. Molecular masses are indicated (B–E); brackets in (F) indicate a molecular mass range of 500–562 kDa.

 
Subcellular localization of plectin is dependent upon the short alternative N-terminal sequences
Initially, we expressed short versions of plectin isoforms to assess the effect of the alternative sequences on actin-binding and subcellular localization of these proteins (Fig. 5A–I). Ple 1-8 was primarily localized in the nucleus and weakly throughout the cells with faint staining of actin stress fibers (SF) (Fig. 5A). Exon 1 contains a putative nuclear localization signal and N-terminal plectin fragments up to a certain length including the exon 1-encoded sequence exhibit nuclear localization (unpublished data). Ple 1a-8, 1c-8, 1d-8(2), 1e-8, and 1g-8 decorated SFs along their entire lengths (Fig. 5B, D–F and H). SF-association was less pronounced in the case of Ple 1d-8(2), which appeared more diffuse (Fig. 5E). Ple 1b-8 was not associated with SF at all, but was concentrated in the perinuclear region (Fig. 5C), encasing some type of cellular structure identified as mitochondria (see below). While some SF-association remained in the case of Ple 1f-8, it appeared concentrated at SF-ends and the cell periphery, particularly in areas resembling filopodia or lammellipodia (Figs 5G and 8). Unexpectedly, when the ABD alone (Ple 2-8, with an exogenous ATG introduced with the vector) was expressed, it was mostly diffuse and only weakly SF-associated (Fig. 5J). Ple 1h-8 and Ple 4-8 (originally thought to resemble the product of Ple 1h-8-expression) were diffusely distributed, without any SF-association (Fig. 5I and K). The reason for this was that the ATG in exon 6 was strongly favored for translation initiation, even in a construct (Ple 6-8) with an exogenous ATG preceding exon 6 (Fig. 4D). Correct expression of plectin ABD isoforms as single protein species of expected molecular masses was confirmed by immunoblotting (Fig. 4E). Expression of full-length plectin cDNAs starting with all 11 alternatively spliced first exons demonstrated that all isoforms were expressed as full-length proteins with molecular masses ranging from 498–533 kDa (527–562 kDa including the 29 kDa EGFP-tag) (Fig. 4F). Plectin antiserum no. 9 (13) was used here because it allowed better detection than the GFP antibody, and, more importantly, it was ensured that the N termini of the expressed proteins were largely intact (the serum was raised against a fragment encoded by exons 9–12). Full-length proteins expressed in mouse keratinocytes were visualized via their EGFP-tags (Fig. 5L–T). Plectin 1 was distributed throughout the cytosol with a diffuse-dotty appearance and faint indications of filamentous arrangement at the cell periphery, but not in the nucleus (Fig. 5L). Plectin 1a was concentrated in patchy structures (Fig. 5M, arrowheads), probably hemidesmosomes, which are formed by these cells in culture (13). The localization of plectin 1b was similar to the one observed for the short version (Fig. 5N). Plectin 1c, 1e and 1g showed a pronounced filamentous arrangement (Fig. 5O, Q and S), which was much less the case for plectin 1d(2) and plectin 1f (Fig. 5P and R). Plectin 1h was granular-diffuse, unlike all other isoforms (Fig. 5T).

View larger version (56K): [in this window] [in a new window]   Figure 5. Expression of EGFP-tagged truncated and full-length plectin isoforms in cultured cells. (A–I) Plectin ABD versions with the eight first coding exons 1-1g (A–H) and the non-coding first exon 1h (I) were expressed in mouse fibroblasts. Additionally, versions lacking a first exon, Ple 2-8 (J) or a truncated ABD, Ple 4-8 (K) were expressed. Most expressed proteins (Ple 1a-8, 1c-8, 1d-8, 1e-8, 1f-8, 1g-8) associated with actin stress fibers, with Ple 1f being more prominent at fiber ends and at the cell periphery. Ple 1-8 was concentrated in the nucleus, Ple 1b-8 was localized perinuclear in dotty aggregates (identified as mitochondria; see Fig. 7) and Ple 1h-8 (as well as Ple 1i-8 and 1j-8, not shown) was diffuse, showing no distinct localization (the same was observed for Ple 4-8 and partly for Ple 2-8). (L–T) Full-length plectin isoforms were expressed in mouse keratinocytes. Ple 1 showed a rather diffuse granular localization throughout the cell (L), while the localization of Ple 1a appeared patchy, indicating an association with hemidesmosome-like structures (M, arrowheads). Ple 1b was concentrated in the perinuclear region and appeared to be associated with mitochondria (N). Ple 1c, 1e and 1g had a pronounced filamentous phenotype (O,Q,S), which was much weaker in case of Ple 1d(2), 1f and 1h (P,R,T). Bars, 20 µm.

 

View larger version (88K): [in this window] [in a new window]   Figure 8. Localization of plectin isoform 1f at stress fiber ends. Plectin-deficient mouse fibroblasts were transfected with EGFP-fusion expression plasmids encoding Ple 1a-8 (A–D) and Ple 1f-8 (E–H), or full-length lectin 1f (IJ). After fixation, cells were immunolabeled with antibodies to actin (B,F) or vinculin (D,H,J). Ple 1f-8 appeared concentrated at the cell periphery and in distinct dots/streaks (E,G, arrowheads), while Ple 1a-8 mainly decorated actin stress fibers more central within the cells and uniformly along their entire length (A,C). Also, in contrast to Ple 1a-8, which appeared somewhat weaker at FACs (CD, arrowheads), Ple 1f-8 and full-length plectin 1f appeared to be specifically decorating the ends of actin stress fibers at vinculin-positive FACs (EF, GH, IJ, arrowheads). Bars, 20 µm.

 
Plectin 1h colocalizes with vimentin IFs, but not with microtubules
Plectin isoforms expressed from transcripts with non-coding first exons in principle cannot be selectively and directly detected among other isoforms by immunofluorescence microscopy (only a subtractive procedure is conceivable, but would be tremendously difficult). Expression of tagged versions therefore represents the only sensible alternative to study these isoforms. We expressed plectin 1h in fibroblasts and performed antibody costaining of vimentin (Fig. 6A–D) or tubulin (Fig. 6E). Plectin 1h was found associated with vimentin IFs, in some cases leading to a collapse of the filaments to the perinuclear area, as has previously been observed when plectin fragments containing the C-terminal IF-binding site were overexpressed (6,18). No colocalization with tubulin or the microtubule network was observed and microtubules appeared unaffected by IF collapse.

View larger version (100K): [in this window] [in a new window]   Figure 6. Ectopic expression of EGFP-tagged full-length plectin 1h in plectin-deficient mouse fibroblasts. After fixation, cells were immunolabeled using primary antibodies to vimentin (B,D) or tubulin (E) and TexasRed-labeled secondary antibodies. Overexpressed plectin 1h (A,C) clearly colocalized with vimentin filaments in extended networks (C,D, arrowheads in insets), or in perinuclear aggregates after filament collapse (AB). No colocalization with tubulin was observed (E). Bar, 20 µm.

 
Plectin 1b links the vimentin IF system to mitochondria
The exon 1b encoded sequence had a remarkable effect on localization, as it completely abolished SF-association, and exclusively targeted the expressed protein to mitochondria. Ple 1b-8 appeared to literally engulf these organelles, as revealed by immunofluorescence microscopy of transfected PtK2 and mouse fibroblast cells using antibodies directed against the mitochondria-specific -subunit of oxidative phosphorylation system complex V (ATPase) (Fig. 7AB and CD). Full-length plectin 1b, too, was associated with mitochondria, as confirmed by treating transfected fibroblasts with a mitochondria-selective dye (Fig. 7EF; arrowheads in enlargements). Plectin 1b codistributed with mitochondria, while also being associated with vimentin IFs (Fig. 7GH), likely forming a bridge between the filament network and the organelles.

View larger version (115K): [in this window] [in a new window]   Figure 7. Co-localization of plectin 1b with mitochondria. C-terminally EGFP-tagged Ple 1b-8 was expressed in PtK2 (AB) and mouse fibroblast cell lines (CD). Full-length Ple-1b was expressed in mouse fibroblasts (EF, GH). After fixation, cells were immunolabeled using primary antibodies to the -subunit of mitochondrial ATPase (A, C) or to vimentin (G) and TexasRed-conjugated secondary antibodies, or incubated with a TexasRed-conjugated mitochondria-specific dye (MitoTracker) prior to fixation (E). Plectin 1b colocalized with mitochondria (AB, CD and EF, arrowheads in enlarged insets), while the full-length protein also colocalized with vimentin filaments (GH). Bars, 20 µm.

 
Plectin 1f is concentrated in vinculin-positive structures at actin stress fiber ends
A distinct localization was also found for Ple 1f-8 when ectopically expressed in cells. To directly compare the localization of Ple 1f-8 with that of Ple 1a-8, a form typically associating with actin SFs, both proteins were co-expressed in PtK2 cells using two fluorescent tags, EGFP and DsRed1, respectively (data not shown). While Ple 1a-8 decorated parallel arrangements of actin SFs running throughout the main cell body, Ple 1f-8 only weakly associated with these fibers but instead was concentrated at their ends and at the cell periphery. We further analyzed this by immunofluorescence microscopy of plectin-deficient mouse fibroblasts expressing EGFP-versions of Ple 1a-8 and Ple 1f-8 using antibodies to actin and vinculin (Fig. 8). Plectin-deficient fibroblasts were chosen due to their increase in actin SFs and focal adhesion contacts (FACs), although this phenotype was expected to be reversed to some degree by the ABD contained in the ectopically expressed plectin constructs (4). Confirming the earlier observations, SFs were decorated along their entire lengths by Ple 1a-8 (Fig. 8A and C), with an apparently weaker association with SF-ends at vinculin-positive FACs (Fig. 8CD). Ple 1f-8-fluorescence, however, was more accentuated at the endings of SFs (Fig. 8EF) at vinculin-positive FACs (Fig. 8GH). Interestingly, in cells overexpressing Ple 1f-8, more actin seemed to be recruited to FACs (Fig. 8F) as compared to untransfected cells or to cells transfected with Ple 1a-8 (Fig. 8B). Overexpressed full-length plectin 1f was also found concentrated at vinculin-positive FACs (Fig. 8IJ), confirming that this behavior was not limited to the short form.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
In addition to plectin's already established great versatility as cytolinker protein, the unusual multiplicity of tissue and cell type-specific plectin variants distinguished by their different N termini unfolded new perspectives regarding both, possible functions of expressed isoforms and regulation of gene expression. Ectopic expression of recombinant proteins is a widely used experimental approach in which usually only the coding sequences are considered for cloning. Our results, however, indicate that 5'-transcript ends have an important part in translation, and may also improve the quality of expression. This should be considered whenever designing cDNA constructs for expression in eukaryotic cells, especially in light of the recent challenging (19) of the general rule that translation initiation in vertebrate gene transcripts takes place at sites fulfilling certain position requirements and context dependencies (20), usually the ATGs closest to the 5'-mRNA ends and closely resembling the Kozak sequence CCA/GCCaugG (21,22).

The influence of 5'-UTRs has been recognized as an important factor in regulation of eukaryotic protein synthesis. Stalling of the translational machinery by means of short upstream ORFs (uORFs) within the 5'-UTR of genes is a known mechanisms for controlling translation (reviewed in 23), and potential defects of this control point have even been observed in relation to a human disease associated with a mutation affecting the 5'-UTR of the serotonin receptor mRNA (24). In the case of plectin transcripts with non-coding first exons, skipping of the exon-4 ATG as translation initiation site intrigued 2-fold. Firstly, this indicated reinitiation of translation at the ATG in exon 6, probably after translation of an uORF that masks the ATG in exon 4. In plectin 1j(23), through insertion of the exon 3 sequence, the uORFs do not extend over this ATG, so in this case, initiation at the exon 4-ATG might be possible. However, with our short constructs, we always observed initiation at the ATG in exon 6. The second interesting consequence manifests itself on the protein level. Initiation of translation in exon 6 instead of exon 4 leads to the complete loss of the first calponin homology (CH) domain, and consequently also to the loss of actin- and integrin ß4-binding. Positive binding of recombinant forms starting with the exon 4-ATG to both interaction partners showed that the N-terminal ABD sequence up to, but not including, the second actin binding site identified by peptide binding studies (25) was dispensable for these interactions. At least regarding integrin ß4, this was inconsistent with reported yeast-2-hybrid binding data suggesting that the complete ABD was necessary for binding to integrin (26). Due to the close relationship between the two binding sites it was not surprising to find that binding to actin or integrin ß4 was mutually exclusive. In a recent study, this was convincingly explained by a two-step induced fit mechanism involving binding and subsequent domain rearrangement (27). Interestingly, in microtubule-actin crosslinking factor isoform 3 (28), an alternative sequence is spliced into a site corresponding to the plectin exon 5/6 boundary, also causing the loss of the first of two CH domains. Similarly, truncation of BPAG1, leading to BPAG1n, an isoform lacking the first CH domain of the ABD, was suggested to activate a cryptic microtubule-binding site located within the plakin domain (29). However, in our plectin 1h-overexpression studies no indication of microtubule association was observed. In this context it should be noted that the analysis (and even verification of expression) of endogenous plectin 1h, 1i or 1j probably will be very difficult. They lack sequences discriminating them from other plectin isoforms, and transcript levels (12) as well as translation (this study) are low. So far, their, in view of the very low expression presumably highly specific, function is unknown.

Our quantitative analyses of expression levels revealed considerable variability among the different plectin isoforms. The long and complex 5'-UTRs of plectin (-1/0a)1c, 1h, 1i and 1j(23) apparently destabilized the transcripts, but surprisingly the normalization of isoform expression data based on mRNA levels showed that translation efficiency of plectin isoform 1c from its alternative transcripts was independent of whether they contained a short (113 nt) or long (848 nt) 5'-UTR. Previous data (12) had indicated that plectin (0)1c(23) was brain-specific, while plectin (-1/0a)1c showing a broader tissue distribution was mainly expressed in skin. Besides harboring tissue-specific promoters, different 5'-UTRs may contain other (structural) elements that ensure preferential expression of certain plectin isoform. This is supported by our finding that translation of exon 0 containing transcripts in a neuronal derived cell-line was favored by 75% compared to several epithelial cell lines.

Analysis of subcellular localization of overexpressed truncated and full-length plectin isoforms revealed that the alternative N termini had the potential to profoundly influence isoform localization. Full-length plectin 1a, but not the short N-terminal fragments, localized to hemidesmosomes in basal keratinocytes and were capable of rescuing the hemidesmosomal phenotype of plectin-deficient cells (13). Thus, it seems unlikely, that the 37 amino acid sequence encoded by exon 1a directly affects the ABD or acts as a hemidesmosomal targeting signal. It may instead affect functions of plectin that involve other molecular domains not necessarily in its close vicinity. Influence on more remote regions could be exerted through folding of the polypeptide chain, or intermolecular interaction in plectin oligomers formed by the lateral association of coiled-coil dimers in an antiparallel fashion (30).

We show here that plectin 1b specifically colocalizes with mitochondria and possibly connects them to the IF system. The involvement of the cytoskeleton in moving and positioning of mitochondria had been suggested at numerous occasions (reviewed in 31), and a role for plectin in linking myofibrils to mitochondria via desmin IFs was suggested by immunogold electron microscopy of striated muscle (32). In addition to regulating mitochondrial positioning (33), cytoskeleton-mitochondria interactions were implicated in modulation of mitochondrial function through changes of mitochondrial shape (34) or signaling via outer membrane proteins. Desmin-deficient mice displayed abnormal mitochondrial distribution in muscle fibers and dramatic changes in mitochondrial function (35) and mitochondria were disrupted in plectin-deficient muscle (32). Assuming that the various isoforms of plectin expressed in muscle (12) have different functions, it would make sense that only one particular variant, in this case plectin 1b, would act as the linker between desmin IFs and mitochondria.

For the isoform containing the exon 1f-sequence, we found a diminished decoration of SFs, and instead a shift of its localization to FACs and filopodia/lammellipodia. The 1f sequence might by itself, or through modulation of the plectin ABD, favor interaction with FAC-associated proteins. Thus, plectin 1f may have a role in sequestering regulatory factors and cytoskeletal subcomponents in their soluble, not yet fully polymerized form, and thereby initiate network formation at such sites. One candidate protein would be vimentin, that, in its unassembled form, has been shown to bind to the ABDs of both fimbrin (36) and plectin (unpublished data). Once filament formation has been established, plectin may then bind and stabilize the filaments via its C-terminal IF-binding site. Further experiments will be necessary to obtain evidence supporting (or refuting) this hypothesis.

In summary, the experimental approach of ectopic expression and quantification of plectin isoform fragments combined with expression of truncated as well as full-length protein versions proved useful in characterization of plectin isoform function. In addition, analysis of 5'-UTRs revealed the somewhat unexpected influence of these sequence stretches on protein expression. The alternative splicing of plectin's first exons may make this well characterized gene an interesting system for more general studies of 5'-UTR functions and mechanisms. Our observations should also raise the appreciation of the role of 5'-UTRs and alternative splicing (specifically in the 5' region of a gene) as an important regulatory mechanism of protein expression and protein function—especially now, where whole genome sequences have become available and, on their bases, several bioinformatics studies have already revealed that the frequency of alternative splicing is much higher than originally anticipated (37,38).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Plectin cDNA constructs
All cDNAs were engineered based on published sequences (12) with ends containing suitable restriction sites. For constructs not including 5'-UTRs, 5'-PCR primers were flanked with EcoRI and NdeI sites (GAATTCCATatg...), the latter including the start codon. For constructs including 5'-UTRs, the 5'-primers included EcoRI immediately upstream of the exon-specific sequence. 3'-primers were flanked by in-frame EcoRI-sites (i.e. adding the amino acids EF to the expressed proteins) located in exon 8 (12) or after the last codon in exon 32 (13). cDNAs identified in 5'-RACE experiments, and some genomic clones, were used as PCR templates. All PCR-generated cDNAs were verified by sequencing. Final cDNA constructs were cloned into the expression vectors pEGFP-N2 (BD Clontech, Palo Alto, CA; adding a C-terminal EGFP) or pGR274, a derivative of pCMS-EGFP (Clontech), where the multiple cloning site was modified to accommodate EcoRI-flanked inserts and a downstream myc-epitope tag-encoding sequence was added via EcoRI/XbaI sites. The myc-tag was excised from pAD29 (18). Mutagenesis of the ATGs in exon 6 was achieved by SOE-PCR (first ATG) or simple PCR using a mutated primer (second ATG). In both cases, BglII/StuI-fragments were used to replace the wild-type sequence of Ple 1h-8.

Cell culture, transfection and immunofluorescence microscopy
Rat kangaroo PtK2 (CCL 56; Amer. Type Culture Collection, Rockville, MD), immortalized (p53-deficient) plectin (-/-) and plectin wild-type mouse fibroblasts (4), and N2A neuroblastoma cells (ATCC CCL-131) were cultured in DMEM, supplemented with 50 U/ml penicillin, 50 µg/ml streptomycin, and 10% FCS, at 37°C and 5% CO₂. Immortalized (p53-deficient) mouse keratinocytes (13) were cultured in KGM (Bio Whittakar CC-4131) supplemented with 2% FCS. For fluorescence micro-scopy, cells were grown on glass coverslips. For transfection, FuGENE6 (Roche Diagnostics) or Lipofectamine 2000 (Invitrogen, Carlsbad, CA) was used as recommended by the manufacturers. After 16–24 h, transfected cells were processed for mRNA isolation, immunoblotting or immunofluorescence microscopy. For the latter, cells were fixed using chilled (-20°C) methanol and processed as previously described (6). The following immunoreagents were used: anti-actin (clone AC40, 1 : 200), anti--tubulin (clone B-5-1-2, 1 : 2000), anti-vinculin (clone vin-11-5, 1 : 400) (all from Sigma-Aldrich Corp., St Louis, MO), anti-vimentin (clone V9, 1 : 100; Dako, Denmark), and anti-ATPase (-subunit, oxidative complex V, clone 7H10, 1 : 200; Molecular Probes, Eugene, OR). TexasRed-conjugated AffiniPure goat anti-mouse IgG (H+L) was used as secondary antibody (1 : 200; Jackson ImmunoResearch Laboratories, West Grove, PA). Mitochondria were alternatively visualized by incubating cells with TexasRed-conjugated MitoTracker (Molecular Probes) for 45 min prior to fixation. Specimens were viewed in a Zeiss Axiophot fluorescence microscope and confocal images were obtained using the LSM510 module (Carl Zeiss, Oberkochen, Germany).

Expression of fusion proteins in bacteria and in vitro protein interaction assay
cDNA constructs used for the in vitro binding assay and the experimental details have been described previously (6,12). In brief, microtiter plates were coated with one protein, and the other, Eu³⁺-labeled, soluble protein was overlaid using different concentrations. After washing, the amount of bound protein was determined by measuring Eu³⁺-fluorescence and comparison with a standard.

Immunoblotting analysis
Cultured cells were directly lysed in SDS-sample buffer, lysates subjected to SDS–PAGE, and proteins transferred to nitrocellulose sheets (Schleicher and Schuell, Dassel, Germany). Blots were developed using antibodies to GFP (diluted 1 : 1000; clones 7.1 and 13.1, Roche Diagnostics, Basel, Switzerland), and rabbit antiserum no. 9 raised against a recombinant N-terminal protein fragment corresponding to exons 9–12 (E419–V451) of rat plectin (diluted 1 : 3000) (13), followed by incubation with AP-conjugated goat anti-rabbit IgG (Jackson Immuno-Research Laboratories, West Grove, USA; dilution 1 : 5000) and visualization of proteins with BCIP/NBT. For quantification, stained membranes were scanned and bands evaluated using the ImageQuant 5.1 software package (Molecular Dynamics/Amersham Biosciences, Piscataway, NJ). Normalization factors based on EGFP-reporter signals were applied to plectin isoform values.

RNase protection assays
cDNA sequences used as probes were subcloned into pSP64 (Promega, Madison, WI) by PCR-cloning using primers flanked with BamHI/EcoRI restriction sites. Protected sequences were 120 nt (myc-specific probe) and 90 nt (EGFP) in length, allowing simultaneous detection. The experimental procedure was as described previously (39), with these modifications: mRNA was isolated from cultured cells using TRIzol reagent (Invitrogen). One to two micrograms of RNA and an excess of radiolabeled probe (100 000 cpm) were used per assay. Signals of protected fragments were quantified using an InstantImager (PerkinElmer, Boston, MA). Normalization factors were calculated from the signals obtained with the EGFP-specific probe and applied to the values obtained for plectin isoform transcripts detected by the myc-probe.

Calculation of ‘Kozak scores’
This score was generated by dividing the sum of the frequencies obtained for nucleotides surrounding the ATGs in plectin sequences with that obtained for the Kozak consensus sequence GCCRCCaugG (21). The frequency table used was based on a database of 1534 reviewed cytoplasmic protein transcripts (19). Bases in all positions were weighted equally.

	ACKNOWLEDGEMENTS

 
G.A.R. was in part supported by a pre-doctoral fellowship awarded by the Austrian Academy of Sciences. This work was supported by grants P14520 and F006-11 from the Austrian Science Research Fund.

	FOOTNOTES

 
^* To whom correspondence should be addressed. Tel: +43 1427752851; Fax: +43 1427752854; Email: wiche{at}abc.univie.ac.at

Present address: Division of Gynecology, Molecular Oncology Group, Department of Obstetrics and Gynecology, University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Leung, C.L., Green, K.J. and Liem, R.K. (2002) Plakins: a family of versatile cytolinker proteins. Trends Cell Biol., 12, 37–45.[CrossRef][Web of Science][Medline]

2. Wiche, G. (1998) Role of plectin in cytoskeleton organization and dynamics. J. Cell Sci., 111, 2477–2486.[Abstract]

3. Fuchs, E. and Karakesisoglou, I. (2001) Bridging cytoskeletal intersections. Genes Dev., 15, 1–14.[Free Full Text]

4. Andrä, K., Nikolic, B., Stöcher, M., Drenckhahn, D. and Wiche, G. (1998) Not just scaffolding: plectin regulates actin dynamics in cultured cells. Genes Dev., 12, 3442–3451.[Abstract/Free Full Text]

5. Wiche, G., Krepler, R., Artlieb, U., Pytela, R. and Aberer, W. (1984) Identification of plectin in different human cell types and immunolocalization at epithelial basal cell surface membranes. Exp. Cell Res., 155, 43–49.[CrossRef][Web of Science][Medline]

6. Rezniczek, G.A., de Pereda, J.M., Reipert, S. and Wiche, G. (1998) Linking integrin alpha6beta4-based cell adhesion to the intermediate filament cytoskeleton: direct interaction between the beta4 subunit and plectin at multiple molecular sites. J. Cell Biol., 141, 209–225.[Abstract/Free Full Text]

7. Nievers, M.G., Kuikman, I., Geerts, D., Leigh, I.M. and Sonnenberg, A. (2000) Formation of hemidesmosome-like structures in the absence of ligand binding by the (alpha)6(beta)4 integrin requires binding of HD1/plectin to the cytoplasmic domain of the (beta)4 integrin subunit. J. Cell Sci., 113, 963–973.[Abstract]

8. Eger, A., Stockinger, A., Wiche, G. and Foisner, R. (1997) Polarisation-dependent association of plectin with desmoplakin and the lateral submembrane skeleton in MDCK cells. J. Cell Sci., 110, 1307–1316.[Abstract]

9. Andrä, K., Lassmann, H., Bittner, R., Shorny, S., Fässler, R., Propst, F. and Wiche, G. (1997) Targeted inactivation of plectin reveals essential function in maintaining the integrity of skin, muscle, and heart cytoarchitecture. Genes Dev., 11, 3143–3156.[Abstract/Free Full Text]

10. Rouan, F., Pulkkinen, L., Meneguzzi, G., Laforgia, S., Hyde, P., Kim, D.U., Richard, G. and Uitto, J. (2000) Epidermolysis bullosa: novel and de novo premature termination codon and deletion mutations in the plectin gene predict late-onset muscular dystrophy. J. Invest. Dermatol., 114, 381–387.[CrossRef][Web of Science][Medline]

11. Koss-Harnes, D., Hoyheim, B., Anton-Lamprecht, I., Gjesti, A., Jorgensen, R.S., Jahnsen, F.L., Olaisen, B., Wiche, G. and Gedde-Dahl, T., Jr (2002) A site-specific plectin mutation causes dominant epidermolysis bullosa simplex Ogna: two identical de novo mutations. J. Invest. Dermatol., 118, 87–93.[CrossRef][Web of Science][Medline]

12. Fuchs, P., Zörer, M., Rezniczek, G.A., Spazierer, D., Oehler, S., Castanon, M.J., Hauptmann, R. and Wiche, G. (1999) Unusual 5' transcript complexity of plectin isoforms: novel tissue-specific exons modulate actin binding activity. Hum. Mol. Genet., 8, 2461–2472.[Abstract/Free Full Text]

13. Andrä, K., Kornacker, I., Jörgl, A., Zörer, M., Spazierer, D., Fuchs, P., Fischer, I. and Wiche, G. (2003) Plectin-isoform-specific rescue of hemidesmosomal defects in plectin (-/-) keratinocytes. J. Invest. Dermatol., 120, 189–197.[CrossRef][Web of Science][Medline]

14. Elliott, C.E., Becker, B., Oehler, S., Castanon, M.J., Hauptmann, R. and Wiche, G. (1997) Plectin transcript diversity: identification and tissue distribution of variants with distinct first coding exons and rodless isoforms. Genomics, 42, 115–125.[CrossRef][Web of Science][Medline]

15. Schröder, R., Fürst, D.O., Klasen, C., Reimann, J., Herrmann, H. and van der Ven, P.F. (2000) Association of plectin with Z-discs is a prerequisite for the formation of the intermyofibrillar desmin cytoskeleton. Lab. Invest., 80, 455–464.[Web of Science][Medline]

16. Pedersen, A.G. and Nielsen, H. (1997) Neural network prediction of translation initiation sites in eukaryotes: perspectives for EST and genome analysis. Proc. Int. Conf. Intell. Syst. Mol. Biol., 5, 226–233.[Medline]

17. Salamov, A.A., Nishikawa, T. and Swindells, M.B. (1998) Assessing protein coding region integrity in cDNA sequencing projects. Bioinformatics, 14, 384–390.[Abstract/Free Full Text]

18. Nikolic, B., Mac Nulty, E., Mir, B. and Wiche, G. (1996) Basic amino acid residue cluster within nuclear targeting sequence motif is essential for cytoplasmic plectin-vimentin network junctions. J. Cell Biol., 134, 1455–1467.[Abstract/Free Full Text]

19. Peri, S. and Pandey, A. (2001) A reassessment of the translation initiation codon in vertebrates. Trends Genet., 17, 685–687.[CrossRef][Web of Science][Medline]

20. Kozak, M. (1978) How do eucaryotic ribosomes select initiation regions in messenger RNA? Cell, 15, 1109–1123.[CrossRef][Web of Science][Medline]

21. Kozak, M. (1981) Possible role of flanking nucleotides in recognition of the AUG initiator codon by eukaryotic ribosomes. Nucl. Acids Res., 9, 5233–5262.[Abstract/Free Full Text]

22. Kozak, M. (1984) Compilation and analysis of sequences upstream from the translational start site in eukaryotic mRNAs. Nucl. Acids Res., 12, 857–872.[Abstract/Free Full Text]

23. Meijer, H.A. and Thomas, A.A. (2002) Control of eukaryotic protein synthesis by upstream open reading frames in the 5'-untranslated region of an mRNA. Biochem. J., 367, 1–11.[CrossRef][Web of Science][Medline]

24. Niesler, B., Flohr, T., Nothen, M.M., Fischer, C., Rietschel, M., Franzek, E., Albus, M., Propping, P. and Rappold, G.A. (2001) Association between the 5' UTR variant C178T of the serotonin receptor gene HTR3A and bipolar affective disorder. Pharmacogenetics, 11, 471–475.[CrossRef][Web of Science][Medline]

25. Winder, S.J., Hemmings, L., Maciver, S.K., Bolton, S.J., Tinsley, J.M., Davies, K.E., Critchley, D.R. and Kendrick-Jones, J. (1995) Utrophin actin binding domain: analysis of actin binding and cellular targeting. J. Cell Sci., 108, 63–71.[Abstract]

26. Geerts, D., Fontao, L., Nievers, M.G., Schaapveld, R.Q., Purkis, P.E., Wheeler, G.N., Lane, E.B., Leigh, I.M. and Sonnenberg, A. (1999) Binding of integrin alpha6beta4 to plectin prevents plectin association with F-actin but does not interfere with intermediate filament binding. J. Cell Biol., 147, 417–434.[Abstract/Free Full Text]

27. Garcia-Alvarez, B., Bobkov, A., Sonnenberg, A. and de Pereda, J.M. (2003) Structural and functional analysis of the actin binding domain of plectin suggests alternative mechanisms for binding to F-actin and integrin beta4. Structure, 11, 615–625.[Medline]

28. Bernier, G., Mathieu, M., De Repentigny, Y., Vidal, S.M. and Kothary, R. (1996) Cloning and characterization of mouse ACF7, a novel member of the dystonin subfamily of actin binding proteins. Genomics, 38, 19–29.[CrossRef][Web of Science][Medline]

29. Yang, Y., Bauer, C., Strasser, G., Wollman, R., Julien, J.P. and Fuchs, E. (1999) Integrators of the cytoskeleton that stabilize microtubules. Cell, 98, 229–238.[CrossRef][Web of Science][Medline]

30. Foisner, R. and Wiche, G. (1987) Structure and hydrodynamic properties of plectin molecules. J. Mol. Biol., 198, 515–531.[CrossRef][Web of Science][Medline]

31. Yaffe, M.P. (1999) Dynamic mitochondria. Nat. Cell Biol., 1, E149–150.[CrossRef][Web of Science][Medline]

32. Reipert, S., Steinböck, F., Fischer, I., Bittner, R.E., Zeöld, A. and Wiche, G. (1999) Association of mitochondria with plectin and desmin intermediate filaments in striated muscle. Exp. Cell Res., 252, 479–491.[CrossRef][Web of Science][Medline]

33. Appaix, F., Kuznetsov, A.V., Usson, Y., Kay, L., Andrienko, T., Olivares, J., Kaambre, T., Sikk, P., Margreiter, R. and Saks, V. (2003) Possible role of cytoskeleton in intracellular arrangement and regulation of mitochondria. Exp. Physiol., 88, 175–190.[Abstract]

34. Rappaport, L., Oliviero, P. and Samuel, J.L. (1998) Cytoskeleton and mitochondrial morphology and function. Mol. Cell Biochem., 184, 101–105.[CrossRef][Web of Science][Medline]

35. Milner, D.J., Mavroidis, M., Weisleder, N. and Capetanaki, Y. (2000) Desmin cytoskeleton linked to muscle mitochondrial distribution and respiratory function. J. Cell Biol., 150, 1283–1298.[Abstract/Free Full Text]

36. Correia, I., Chu, D., Chou, Y.H., Goldman, R.D. and Matsudaira, P. (1999) Integrating the actin and vimentin cytoskeletons: adhesion-dependent formation of fimbrin-vimentin complexes in macrophages. J. Cell Biol., 146, 831–842.[Abstract/Free Full Text]

37. Modrek, B. and Lee, C.J. (2003) Alternative splicing in the human, mouse and rat genomes is associated with an increased frequency of exon creation and/or loss. Nat. Genet., 34, 177–180.[CrossRef][Web of Science][Medline]

38. Kriventseva, E.V., Koch, I., Apweiler, R., Vingron, M., Bork, P., Gelfand, M.S. and Sunyaev, S. (2003) Increase of functional diversity by alternative splicing. Trends Genet., 19, 124–128.[CrossRef][Web of Science][Medline]

39. Spazierer, D., Fuchs, P., Pröll, V., Janda, L., Oehler, S., Fischer, I., Hauptmann, R. and Wiche, G. (2003) Epiplakin gene analysis in mouse reveals a single exon encoding a 725 kDa protein with expression restricted to epithelial tissues. J. Biol. Chem., 278, 31657–31666.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				P. Fuchs, M. Zorer, S. Reipert, G. A. Rezniczek, F. Propst, G. Walko, I. Fischer, J. Bauer, M. W. Leschnik, B. Luscher, et al. Targeted Inactivation of a Developmentally Regulated Neural Plectin Isoform (Plectin 1c) in Mice Leads to Reduced Motor Nerve Conduction Velocity J. Biol. Chem., September 25, 2009; 284(39): 26502 - 26509. [Abstract] [Full Text] [PDF]

				J. Kostan, M. Gregor, G. Walko, and G. Wiche Plectin Isoform-dependent Regulation of Keratin-Integrin {alpha}6{beta}4 Anchorage via Ca2+/Calmodulin J. Biol. Chem., July 3, 2009; 284(27): 18525 - 18536. [Abstract] [Full Text] [PDF]

				L. Winter, C. Abrahamsberg, and G. Wiche Plectin isoform 1b mediates mitochondrion-intermediate filament network linkage and controls organelle shape J. Cell Biol., October 21, 2008; 181(6): 903 - 911. [Abstract] [Full Text] [PDF]

				P. Konieczny, P. Fuchs, S. Reipert, W. S. Kunz, A. Zeold, I. Fischer, D. Paulin, R. Schroder, and G. Wiche Myofiber integrity depends on desmin network targeting to Z-disks and costameres via distinct plectin isoforms J. Cell Biol., October 17, 2008; 181(4): 667 - 681. [Abstract] [Full Text] [PDF]

				T. Hijikata, A. Nakamura, K. Isokawa, M. Imamura, K. Yuasa, R. Ishikawa, K. Kohama, S. Takeda, and H. Yorifuji Plectin 1 links intermediate filaments to costameric sarcolemma through {beta}-synemin, {alpha}-dystrobrevin and actin J. Cell Sci., June 15, 2008; 121(12): 2062 - 2074. [Abstract] [Full Text] [PDF]

				G. A. Rezniczek, P. Konieczny, B. Nikolic, S. Reipert, D. Schneller, C. Abrahamsberg, K. E. Davies, S. J. Winder, and G. Wiche Plectin 1f scaffolding at the sarcolemma of dystrophic (mdx) muscle fibers through multiple interactions with {beta}-dystroglycan J. Cell Biol., March 26, 2007; 176(7): 965 - 977. [Abstract] [Full Text] [PDF]

				C. Abrahamsberg, P. Fuchs, S. Osmanagic-Myers, I. Fischer, F. Propst, A. Elbe-Burger, and G. Wiche Targeted ablation of plectin isoform 1 uncovers role of cytolinker proteins in leukocyte recruitment PNAS, December 20, 2005; 102(51): 18449 - 18454. [Abstract] [Full Text] [PDF]

				K. Wilhelmsen, S. H.M. Litjens, I. Kuikman, N. Tshimbalanga, H. Janssen, I. van den Bout, K. Raymond, and A. Sonnenberg Nesprin-3, a novel outer nuclear membrane protein, associates with the cytoskeletal linker protein plectin J. Cell Biol., December 5, 2005; 171(5): 799 - 810. [Abstract] [Full Text] [PDF]

				C.-H. Yang, J. Szeliga, J. Jordan, S. Faske, Z. Sever-Chroneos, B. Dorsett, R. E. Christian, R. E. Settlage, J. Shabanowitz, D. F. Hunt, et al. Identification of the Surfactant Protein A Receptor 210 as the Unconventional Myosin 18A J. Biol. Chem., October 14, 2005; 280(41): 34447 - 34457. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 12/23/3181    most recent ddg345v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (29) Request Permissions Google Scholar Articles by Rezniczek, G. A. Articles by Wiche, G. Search for Related Content PubMed PubMed Citation Articles by Rezniczek, G. A. Articles by Wiche, G. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics