Prominent Isoform Research Articles

Mamestra configurata serpin-1 homologues: cloning, localization and developmental regulation

A screen of a Mamestra configurata (bertha armyworm) midgut cDNA library identified three types of cDNA clones that resemble the Manduca sexta serpin-1 gene family. Two serpins, 1b and 1c, possess a common conserved serpin amino terminal scaffold domain but bear no similarity to any members of the M. sexta gene family within the reactive centre loop. These serpins differ from one another by only two amino acids in the reactive centre loop (S 363→P) and serpin signature (M 369→T) regions. The other member, denoted serpin-1a, is closely related to the M. sexta serpin-1Z. M. configurata serpins as a group were expressed in all insect developmental stages including eggs, larvae and adult moths. Within larvae, serpin gene expression was restricted to the early to middle instar developmental phase and mainly in the fat body and hemocytes. Stress imposed by starvation strongly induced expression in fat body and to a lesser degree in alimentary organs, nervous system and Malphigian tubules. Conversely, starvation decreased expression in hemocytes. Wounding or inoculation with bacteria did not induce serpin gene transcription but did lead to the formation of higher and lower molecular weight forms, presumably serpin-protease complexes and resultant truncated serpin, respectively. Two dimensional PAGE and western blotting analysis revealed at least 12 distinct serpins consisting primarily of neutral, but also highly acidic and basic isoforms, as well as additional high and low molecular weight immuno-reactive species. Serpins-1b/1c are the more prominent serpin isoforms and are expressed predominantly in the fat body and subsequently exported to the hemolymph as revealed by western blotting and immunolocalization. The serpin-1b/1c isoform was found only as the fully glycosylated species within the hemolymph. Hemolymph protease activity was comprised mostly of serine proteases whose overall activity increased dramatically at the onset of the molt concomitant with a sharp decline in serpin gene expression.

β-1,3-Glucanase and chitinase activities in soybean root nodules

Summary β-1,3-Glucanase and chitinase specific activities were measured in soybean nodules and the adjoining root tissue during nodule ontogeny. Activities of both enzymes exhibited a transient pattern in root nodules with a maximum rate occurring following the decline in nitrogen fixation rate at 28 days post inoculation. β-1,3-Glucanase and chitinase levels were higher in the adjacent root tissue than in the nodules. β-1,3-Glucanase activity was optimal at a pH of 5.3 and an incubation temperature of 37 °C. The product formation increased linearly with time. TLC analysis of the hydrolytic products from the nodule reaction mixture revealed the presence of at least seven oligomers. ‘Western’ immunoblot analysis showed the nodule glucanases cross-reacting with antisera raised against virus-inducible tobacco ( Nicotiana tabacum L.) protein PR-1 (17 kDa), tobacco basic protein PR-2 (21 kDa), lima bean ( Phaseolus vulgaris L.) acidic protein PL1 (18 kDa), pinto bean basic protein PR-4d (21 kDa) and tobacco (36 kDa) and cowpea ( Vigna unguiculata L.) (36 kDa) PR-proteins. Chitinase was heat-labile and sensitive to mercury but was protease-resistant. A non-denaturing polyacrylamide gel activity stain revealed the presence of at least four prominent chitinolytic isoforms in nodule extract as well as in root tissue. ‘Western’ immunoblot analysis showed cross-reactivitiy of nodule chitinases with an anti-vacuolar bean leaf chitinase.

Isoform specific expression of the neuronal F-actin binding protein, drebrin, in specialized cells of stomach and kidney epithelia.

To further understand the functional role that the F-actin binding protein, drebrin (developmentally regulated brain protein), plays in the regulation of F-actin, we characterized its expression in non-neuronal cells. Using nanoelectrospray mass spectrometry methods, we initially identified drebrin in non-neuronal cultured cells. Using a drebrin-specific monoclonal antibody, we were able to detect drebrin protein in several different cell lines derived from fibroblasts, astrocytomas, and simple epithelia, but not in cell lines derived from stratified epithelia. Double-label immunofluorescence experiments of cultured cell monolayers revealed the localization of drebrin at the apical plasma membrane together with a pool of submembranous F-actin. Immunoblot analysis of mouse organs revealed that, in addition to its high levels of expression in brain, drebrin was present in stomach and to a lesser degree in kidney, colon, and urinary bladder. Drebrin protein detected in the non-brain organs migrated faster through SDS-PAGE gels, indicating that the lower molecular weight embryonic brain isoform (E2) may be the prominent isoform in these organs. RT-PCR experiments confirmed the specific expression of the E2 isoform in adult stomach, kidney, and cultured cells. In situ immunofluorescence experiments revealed a cell-type specific pattern in both stomach and kidney. In stomach, drebrin was specifically expressed in the acid-secreting parietal cells of the fundic glands, where it accumulated at the extended apical membrane of the canaliculi. In kidney, drebrin was expressed in acid-secreting type A intercalated cells, where it localized specifically to the apical plasma membrane. Drebrin was expressed as well in the distal tubule epithelial cells where the protein was concentrated at the luminal surface and present at the interdigitations of the basolateral membranes.

The rate and anisotropy of impulse propagation in the postnatal terminal crest are correlated with remodeling of Cx43 gap junction pattern.

Disruptions to intermyocyte coupling have been implicated in arrhythmogenesis and development of conduction disturbances. At present, understanding of the relationship between the microscopic organization of intercellular coupling and the macroscopic spread of impulse in the normal and diseased heart is largely confined to theoretical analyses. The abundance and arrangement of gap junctions, as well as conduction properties, were assessed in terminal crest preparations isolated from the atria of neonate, weanling, and adult rabbits. We report that the connexin composition of terminal crest was uncomplicated, with Cx43 being the most prominent isoform detectable by Western blotting and immunostaining. Terminal crest myocytes showed little change in total Cx43-gap junction per cell during postnatal growth as assessed by stereology. However, marked non-uniformities emerged in the sarcolemmal distribution of Cx43-gap junctions. Cx43-gap junction area at myocyte termini increased 3.5-fold from birth to adulthood. Correlated with this change in Cx43, impulse propagation velocity parallel to the myofiber axis, as assessed by multi-site optical mapping using voltage-sensitive dye (di-4-ANEPPS), increased 2.4-fold. Conversely, the amount of Cx43-gap junctions on myocyte sides, and the conduction velocity transverse to the myofiber axis, remained relatively invariant during maturation. Hence, the increasing electrical anisotropy of maturing terminal crest was wholly accounted for by increases in conductance velocity along the bundle. This increase in longitudinal conduction velocity was correlated with changes in the sarcolemmal pattern, but not the overall density, of Cx43-gap junctions. This study provides the first correlative structure/function analysis of the relationship between the macroscopic conduction of impulse and the microscopic cellular organization of gap junctions in a differentiating cardiac bundle. Confirmation is provided for theoretical predictions which emphasize the importance of the cell-to-cell geometry of coupling in determining the spread and pattern of myocardial activation.

Striking multiplicity of eIF4E-BP1 phosphorylated isoforms identified by 2D gel electrophoresis regulation by heat shock.

Eukaryotic initiation factor eIF4E-binding protein 1 (eIF4E-BP1), or PHAS-I, is multiply phosphorylated by insulin-stimulated protein kinase(s). Estimates for the number of phosphorylation sites range from two to greater than eight. IEF/SDS/PAGE can precisely differentiate protein isoforms based on their differences in charge (phosphorylation) and molecular mass. In this study, the diversity of eIF4E-BP1 isoforms was determined using IEF/SDS/PAGE/immunoblotting of unfractionated cell lysates. To investigate the molecular regulation of phosphorylation, alterations in eIF4E-BP1 in response to heat shock in HeLa cells were determined. In exponentially growing cells, 8-10 prominent eIF4E-BP1 isoforms were detected. Following heat shock, a rapid, temperature-dependent dephosphorylation of eIF4E-BP1 occurs roughly concurrent with protein synthesis inhibition; during recovery from heat shock rephosphorylation of eIF4E-BP1 parallels restoration of protein synthesis. However, eIF4E-BP1 and eIF4E kinases remain highly active during heat shock, as okadaic acid treatment restores phosphorylation of both factors in heat shocked cells. eIF4E-BP1 dephosphorylation is associated with eIF4E dissociation from large molecular mass complexes and increased binding to eIF4E-BP1. The amount of eIF4E-BP1 converted to the dephosphorylated state is sufficient to titrate all the eIF4E present. eIF4E-BP1 phosphorylation changes regulated by heat shock also occur in Drosophila. Of the 10 isoforms of eIF4E-BP1 resolved by IEF/SDS/PAGE, at least seven are labelled with [32P] and all 10 are recognized by (eIF4E-BP1)-specific antibodies. These results identify a complex set of eIF4E-BP1 phosphorylation isoforms; changes in the expression of these isoforms in response to stresses such as heat shock may contribute to translation repression.

Specific inhibition of skeletal α-actin gene transcription by applied mechanical forces through integrins and actin

Skeletal α-actin (skA), a prominent fetal actin isoform that is re-expressed by adult cardiac myocytes after chronic overload in vivo, provides a model for studying cytoskeletal gene regulation by mechanical forces in vitro. We have determined the mechanisms by which perpendicular applied forces acting through integrins and the actin cytoskeleton regulate the expression of skA. Rat-2 fibroblasts were transiently transfected with plasmids containing 5′-regulatory regions of the skA gene fused to luciferase coding sequences. A constant, perpendicular force (0.2 pN/μm2) was applied by using a collagen-magnetic bead model; a 25% deformation was obtained on the dorsal cell surface. In this system, force is applied through focal adhesion integrins and strongly induces actin assembly [Glogauer, Arora, Yao, Sokholov, Ferrier and McCulloch (1997) J. Cell Sci. 110, 11-21]. skA promoter activity was inhibited by 68% in cells subjected to 4 h of applied force, whereas Rous sarcoma virus promoter activity was unaffected. In cells transiently transfected with a skA expression vector there was also a parallel 40% decrease in skA protein levels by force, as shown by Western blotting. In L8 cells, constitutive skA expression was decreased by more than 50%. Analyses of specific motifs in the skA promoter revealed that transcriptional enhancer factor 1 and Yin and Yang 1 sites, but not serum response factor and Sp1 sites, mediated inhibitory responses to force. In cells treated with cycloheximide the force-induced inhibition was abrogated, indicating a dependence on new protein synthesis. Inhibition of actin filament assembly with either cytochalasin D or Ca2+-depleted medium blocked the inhibitory effect induced by the applied force, suggesting that actin filaments are required for the regulation of skA promoter activity. Western blot analysis showed that p38 kinase, but not Jun N-terminal kinase or extracellular signal-regulated protein kinase 1/2, was activated by force; indeed, the p38 kinase inhibitor SB203580 relieved the force-induced inhibition of skA. We conclude that the force-induced inhibition of skA promoter activity requires an intact actin cytoskeleton and can be mapped to two different response elements. This inhibition might be mediated through the p38 kinase.

Specific inhibition of skeletal α-actin gene transcription by applied mechanical forces through integrins and actin

Skeletal alpha-actin (skA), a prominent fetal actin isoform that is re-expressed by adult cardiac myocytes after chronic overload in vivo, provides a model for studying cytoskeletal gene regulation by mechanical forces in vitro. We have determined the mechanisms by which perpendicular applied forces acting through integrins and the actin cytoskeleton regulate the expression of skA. Rat-2 fibroblasts were transiently transfected with plasmids containing 5'-regulatory regions of the skA gene fused to luciferase coding sequences. A constant, perpendicular force (0.2 pN/micrometer(2)) was applied by using a collagen-magnetic bead model; a 25% deformation was obtained on the dorsal cell surface. In this system, force is applied through focal adhesion integrins and strongly induces actin assembly [Glogauer, Arora, Yao, Sokholov, Ferrier and McCulloch (1997) J. Cell Sci. 110, 11-21]. skA promoter activity was inhibited by 68% in cells subjected to 4 h of applied force, whereas Rous sarcoma virus promoter activity was unaffected. In cells transiently transfected with a skA expression vector there was also a parallel 40% decrease in skA protein levels by force, as shown by Western blotting. In L8 cells, constitutive skA expression was decreased by more than 50%. Analyses of specific motifs in the skA promoter revealed that transcriptional enhancer factor 1 and Yin and Yang 1 sites, but not serum response factor and Sp1 sites, mediated inhibitory responses to force. In cells treated with cycloheximide the force-induced inhibition was abrogated, indicating a dependence on new protein synthesis. Inhibition of actin filament assembly with either cytochalasin D or Ca(2+)-depleted medium blocked the inhibitory effect induced by the applied force, suggesting that actin filaments are required for the regulation of skA promoter activity. Western blot analysis showed that p38 kinase, but not Jun N-terminal kinase or extracellular signal-regulated protein kinase 1/2, was activated by force; indeed, the p38 kinase inhibitor SB203580 relieved the force-induced inhibition of skA. We conclude that the force-induced inhibition of skA promoter activity requires an intact actin cytoskeleton and can be mapped to two different response elements. This inhibition might be mediated through the p38 kinase.

Multiple NF-ATc Isoforms with Individual Transcriptional Properties Are Synthesized in T Lymphocytes

The transcription factor NF-ATc that controls gene expression in T lymphocytes and embryonic cardiac cells is expressed in three prominent isoforms. This is due to alternative splice/polyadenylation events that lead to the predominant synthesis of two long isoforms in naive T cells and a shorter NF-ATc isoform in effector T cells. Whereas the previously described isoform NF-ATc/A contains a relatively short C terminus, the longer isoforms, B and C, span extra C-terminal peptides of 128 and 246 aa, respectively. We show here that in addition to the strong N-terminal trans-activation domain, TAD-A, which is common to all three NF-ATc isoforms, NF-ATc/C contains a second trans-activation domain, TAD-B, in its C-terminal peptide. Various stimuli of T cells that induce the activity of TAD-A also enhance the activity of TAD-B, but, unlike TAD-A, TAD-B remains unphosphorylated by protein from 12-O-tetradecanoyl 12-phorbol 13-acetate-stimulated T cells. The shorter C-terminal peptide of isoform NF-ATc/B exerts a suppressive transcriptional effect. These properties of NF-ATc/B and -C might be of importance for gene regulation in naive T lymphocytes in which NF-ATc/B and -C are predominantly synthesized.

Alternative Polyadenylation Events Contribute to the Induction of NF-ATc in Effector T Cells

The transcription factor NF-ATc is synthesized in three prominent isoforms. These differ in the length of their C terminal peptides and mode of synthesis. Due to a switch from the use of a 3′ polyA site to a more proximal polyA site, NF-ATc expression switches from the synthesis of the two longer isoforms in naive T cells to that of short isoform A in T effector cells. The relative low binding affinity of cleavage stimulation factor CstF-64 to the proximal polyA site seems to contribute to its neglect in naive T cells. These alternative polyadenylation events ensure the rapid accumulation of high concentrations of NF-ATc necessary to exceed critical threshold levels of NF-ATc for gene induction in effector T cells.

Failure of short‐term mannose therapy of patients with carbohydrate‐deficient glycoprotein syndrome type 1A

Carbohydrate‐deficient glycoprotein syndrome type 1A (CDGS1A) is an inherited disorder with multi‐systemic abnormalities resulting from failure to generate sufficient lipid‐linked oligosaccharide precursor or to transfer the sugar chain to many glycoproteins. Cultured fibroblasts from these patients have reduced incorporation of mannose into glycoproteins which can be corrected by adding D‐mannose to the culture medium. Providing dietary mannose to elevate mannose concentrations in vivo therefore might remedy some of the underglycosylation in the patients. Five children with CDGS1A aged 15 months to 14 y completed a protocol of enteral supplementation with D‐mannose 100 mg/kg every 3 h for 9 d. The mean S‐mannose level increased from 32 μM (range 22‐42 μM) to a trough value of 72 μM (range 39–103 μM). No serious side effects were observed. Surprisingly, the mean serum concentration of four glycoproteins (transferrin, α1‐antitrypsin, antithrombin, and thyroxine‐binding globulin) tended to decrease, and the mean serum concentration of carbohydrate‐deficient transferrin (CDT) increased. Furthermore, the initially present abnormal isoforms of these glycoproteins and of protein C became more prominent and/or additional abnormal isoforms appeared. This short‐term trial does not support a benefit of mannose to the deficient glycosylation of CDGS1A patients.

Two kinds of synchronized rhythmic activity of immature rat spinal motoneurons

To investigate the distribution and effects of hyperpolarization-activated cyclic nucleotide-gated (HCN) channel and its isoforms in bladder, especially in bladder interstitial cells of Cajal (ICC).Four HCN isoforms were detected in bladder tissue from rats using reverse transcription–polymerase chain reaction and Western blotting. The HCN1 subtype was observed in bladder ICCs by double-labeled fluorescence. The effect of the HCN blocker, ZD7288, was investigated using the bladder smooth muscle strip test.HCN1-4 isoforms were all identified in bladder ICCs using reverse transcription–polymerase chain reaction and Western blotting. Based on our semiquantitative analysis, HCN1 was found to be the most prominent isoform. The expression of HCN1 was confirmed in bladder ICCs by double-labeled fluorescence through colabeling of HCN1 and kit (CD117). ZD7288 significantly decreased the bladder excitation.All 4 HCN channel isoforms exist in the bladder, and they affect the bladder excitation, presumably via bladder ICCs.

Localization of myosin II A and B isoforms in cultured neurons.

Tension generated by growth cones regulates both the rate and the direction of neurite growth. The most likely effectors of tension generation are actin and myosins. We are investigating the role of conventional myosin in growth cone advance. In this paper we report the localization of the two most prominent isoforms of brain myosin II in growth cones, neurites and cell bodies of rat superior cervical ganglion neurons. Affinity purified polyclonal antibodies were prepared against unique peptide sequences from human and rat A and B isoforms of myosin heavy chain. Although each of these antibodies brightly stained nonneuronal cells, antibodies to myosin heavy chain B stained neurons with greater intensity than antibodies to myosin heavy chain A. In growth cones, myosin heavy chain B was most concentrated in the margin bordering the thickened, organelle-rich central region and the thin, actin-rich peripheral region. The staining colocalized with actin bundles proximal and distal to the marginal zone, though the staining was more prominent proximally. The trailing edge of growth cones and the distal portion of the neurite often had a rimmed appearance, but more proximal regions of neurites had cytoplasmic labelling. Localizing MHC-B in growth cones previously monitored during advance (using differential interference contrast microscopy) revealed a positive correlation with edges at which retraction had just occurred and a negative correlation with lamellipodia that had recently undergone protrusion. Cell bodies were brightly labelled for myosin heavy chain B. Myosin heavy chain A staining was dimmer and its colocalization with filamentous actin bundles in growth cones was less striking than that of myosin heavy chain B. Growth cones stained for both myosin heavy chain A and B revealed that the two antigens overlapped frequently, but not exclusively, and that myosin heavy chain A lacked the elevation in the marginal zone that was characteristic of myosin heavy chain B. The pattern of staining we observed is consistent with a prominent role for myosin heavy chain B in either generating tension between widely separated areas of the growth cone, or bundling of actin filaments, which would enable other motors to effect this tension. These data support the notion that conventional myosin is important in growth cone advance and turning.

Localisation of the high affinity factilitative glucose transporter protein GLUT 1 in the placenta of human, marmoset monkey (Callithrix jacchus) and rat at different developmental stages

In the present study, the facilitative D-glucose transporter protein GLUT 1 was localised by immunohistochemistry in the placenta of human, marmoset (Callithrix jacchus) and rat at different developmental stages. A polyclonal antiserum against a 13-amino-acid peptide of the GLUT 1 carboxy terminus was used. It identified a protein of around 50 kDa molecular weight in immunoblotting of the placental tissues. GLUT 1 was located in the syncytiotrophoblast, in cytotrophoblast cells and in fetal endothelium. Similar staining patterns, except in human extravillous cytotrophoblast cells, were observed at all differentiation stages, despite differences in the internal placental architecture of the species. In the marmoset placenta, GLUT 1 was undetectable in endothelial cells of maternal vessels. In rat placentae, trophoblastic giant cells, epithelial cells of both visceral and parietal yolk sac, yolk sac vessels and the stratum spongiosum were stained. Reichert's membrane did not immunoreact. Preadsorption of the antiserum with a 13-amino-acid peptide resulted in the loss of immunoreactivity. The results suggest that GLUT 1 is a prominent isoform of glucose transporters in mammalian placentae. It is generally abundant in placental cell populations bordering on the maternal and fetal circulations and may therefore facilitate an effective glucose supply to the fetus and placenta.

Na,K-ATPase β subunit isoform expression in the peripheral nervous system of the rat

Using the RNase protection assay (RPA) to study the distribution of isoforms of the non-catalytic (β) subunit of Na,K-ATPase in the peripheral nervous system, we found both β1 and β2 isoform mRNAs in dorsal root ganglion (DRG), but only β2 mRNA in sciatic nerve. Using Western blot to measure accumulation of the polypeptides at a ligature on the nerve we found that β1 but not β2 polypeptide is carried by rapid axonal transport in the sciatic nerve. These results imply that β1 is the prominent isoform of Na,K-ATPase in neurons and β2 the prominent isoform in Schwann cells.

In Vivo Neuronal Synthesis and Axonal Transport of Kunitz Protease Inhibitor (KPI)‐Containing Forms of the Amyloid Precursor Protein

We have shown previously that the amyloid precursor protein (APP) is synthesized in retinal ganglion cells and is rapidly transported down the axons, and that different molecular weight forms of the precursor have different developmental time courses. Some APP isoforms contain a Kunitz protease inhibitor (KPI) domain, and APP that lacks the KPI domain is considered the predominant isoform in neurons. We now show that, among the various rapidly transported APPs, a 140-kDa isoform contains the KPI domain. This APP isoform is highly expressed in rapidly growing retinal axons, and it is also prominent in adult axon endings. This 140-kDa KPI-containing APP is highly sulfated compared with other axonally transported isoforms. These results show that APP with the KPI domain is a prominent isoform synthesized in neurons in vivo, and they suggest that the regulation of protease activity may be an important factor during the establishment of neuronal connections.

Differential expression of the TGF-beta isoforms in embryogenesis suggests specific roles in developing and adult tissues.

The TGF-beta's are multifunctional, pleiotropic molecules with major effects in control of cellular migration, cellular proliferation, and elaboration of extracellular matrix. Thus far, five distinct isoforms of TGF-beta have been described, each approximately 65-85% homologous and containing the characteristic 9 positionally conserved cysteine residues. Although the actions of the activated mature forms of the different isoforms on cells are qualitatively similar in most cases, there are a few examples of distinct activities. For example, TGF-beta's 1 and 3, but not TGF-beta 2, inhibit the growth of large vessel endothelial cells, and TGF-beta's 2 and 3, but not TGF-beta 1, inhibit the survival of cultured embryonic chick ciliary ganglionic neurons. In addition, selective targeting of the latent forms of the TGF-beta's is suggested by the observation that latent TGF-beta 2 is the prominent isoform found in body fluids such as amniotic fluid, breast milk, and the aqueous and vitreous humor of the eye; it is noteworthy in this regard that TGF-beta 2 is unique among various isoforms in that it lacks a RGD integrin-binding sequence in its precursor. The most dramatic differences in the TGF-beta isoforms are seen at the level of expression, where there is now a wealth of data demonstrating both spatially and temporally distinct expression of both the mRNAs and proteins in developing tissues, regenerating tissues, and in pathologic responses.(ABSTRACT TRUNCATED AT 250 WORDS)

Glutathione S-transferases in normal and cancerous human colon tissue

In order to evaluate the role of the placental form of glutathione S-transferase (GST-Pi) as a tumour marker, activity and composition of GSTs from human colon were investigated. GSTs were purified from normal colon mucosa and from colonic tumours by affinity chromatography on glutathione-agarose. After SDS-PAGE or isoelectric focusing these purified preparations revealed only one band that comigrated with GST-Pi from human placenta. A monoclonal antibody (mAb) very specific for GST-Pi was developed and characterized. On immunoblot this mAb stains purified GST from normal and diseased colon tissue. GST activity was significantly higher in most cancerous (247 +/- 38 nmol/min/mg protein; n = 7), compared with the corresponding normal tissues (171 +/- 18 nmol/min/mg protein; n = 7). In colon from patients without large bowel malignancies GST-Pi is also by far the most prominent isoform detectable. In conclusion, both normal and tumorous colon tissue predominantly express GST-Pi and therefore GST-Pi is not suitable as a tumour marker for colonic carcinomas. However, the increased GST-Pi levels in colonic tumours could possibly contribute to the relatively high resistance to anti-cancer drugs.