| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 36, 34691-34699, September 5, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶ ||
From the
Departments of Neurology and
¶Molecular and Human Genetics,
||Howard Hughes Medical Institute, Baylor College
of Medicine, Houston, Texas 77030, the **Institute of
Biochemistry and Molecular Cell Biology, Vienna Biocenter, University of
Vienna, A-1030 Vienna, Austria, the
Medical Molecular Biology Unit,
Institute of Child Health, University College, London WC1N 1EH, United Kingdom
and the Departments of Laboratory Medicine
and Pathology and Biochemistry, and the Institute of Human Genetics,
University of Minnesota, Minneapolis, Minnesota 55455
Received for publication, March 19, 2003 , and in revised form, May 23, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Since its first description in 1994, LANP has been implicated in myriad cellular functions from the cell surface to the nucleus. First described as a putative human leukocyte antigen class II-associated protein (and hence called PHAPI), it was suspected to be involved in signal transduction in lymphocytes (2). Matsuoka et al. (1994) independently described this protein in the developing cerebellum, and noting that it contained a leucine-rich repeat, called it by the acronym LANP. With a modular architecture reminiscent of a tadpole, LANP consists of a globular head formed by the N-terminal leucine-rich domain containing five leucine-rich repeats (LRR) and a C-terminal tail formed by the remaining length of acidic residues (3). As such, it belongs to a large and very interesting family of proteins that contain LRRs crucial for protein interactions, by forming a very characteristic secondary structure designed for protein-protein interactions (4–6). It was therefore proposed to be a modulator of signaling pathways in cerebellar morphogenesis.
LANP has since been implicated in a number of other functions: as a phosphorylated protein, LANP (known in this context as phosphoprotein 32 or pp32) was suggested to act as tumor suppressor (7–10); LANP has been shown to bind and shuttle the RNA-binding protein HuR, which is involved in RNA stability and transport. More recently it has been described as an inhibitor of histone acetylation and thus a transcriptional regulator (11) and in a very different role as a modulator of apoptosis (12–14). Ulitzur et al. (15, 16) were the first to suggest that LANP may also have cytoplasmic functions: in biochemical assays it binds to the microtubule-associated proteins (MAPs) MAP2, MAP4, and tau, stimulating the microtubule- and dynein-dependent localization of the Golgi apparatus in semi-intact cells. Since it is unclear which of the functions of LANP, if any, might be perturbed in SCA1 pathogenesis, we sought to understand the neuronal properties of LANP.
Here we report that LANP is typically nuclear in undifferentiated neuro2a cells. This compartmentalization is dependent on a nuclear localization signal in its C-terminal domain. LANP tends to be drawn to the cytoplasm during the process of neuronal differentiation. Intriquingly, in a search for LANP interacting proteins, we have identified the microtubule-associated protein, MAP1B, as a cytoplasmic protein that interacts with LANP. This interaction occurs via its light chain. Moreover the effects of MAP1B on neurite extension are altered by its interaction with LANP. This interaction has the potential to not only modulate neuritogenesis during neuronal development, but could also contribute to the loss of neurites and cytoarchitectural disarray seen in SCA1.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
To delimit the region of interaction of LANP we generated N- and C-terminally deleted LANP constructs. The C-terminally deleted LANP construct was engineered by digesting pGBK-T7 LANP with BsmI-SmaI to remove the C-terminal acidic domain, preserving the complete LRR region and then religating the backbone. To generate the N-terminally deleted construct we used a PCR-based strategy to clone a truncated LANP (residues 129–247) into the NdeI-SmaI sites of pGBK-T7.
Cell Culture, Transfection, Expression Constructs, and Immunofluorescence—Clones derived from the pACT2 yeast two-hybrid screen were subcloned into the mammalian expression vector pCMV-HA (Clontech). The construction of the mammalian tet-responsive expression plasmids (pMT5tet and pMT17tet) containing the heavy chain and full-length clones of rat MAP1B tagged at their C-terminal ends to the myc epitope has been described (17). The nuclear localization signal mutants of LANP were constructed by a PCR based mutagenesis strategy so as to alter residues 236 and 237 to alanine residues (from lysine and argine respectively) followed by subcloning into pCMV-myc (Clontech).
Tissue culture cells were obtained from American Type Tissue Collection (Manassas, VA). Neuro2a was used as a prototypical neuronal cell line. Differentiation of neuro2a cells was performed by growing cells in serum-free medium (Opti-MEM, Invitrogen) containing 0.3 mM dibutyryl cAMP (18). For experiments on non-neuronal cells, COS-7 cells or BHK-21 cells (a kind gift of Dr. O. Skalli, University of Illinois) were used. To quantify neuritogenesis, cells were counted as having extended neurites if they exhibited at least one process longer than two cell bodies in length.
Transfections were performed on coverslips using the LipofectAMINE Plus reagent (Invitrogen). Cells were fixed 48 h post-transfection before being processed for immunofluorescence (17). All images were captured by either light or confocal laser-scanning microscopy (Zeiss). Images were manipulated using Adobe Photoshop 5.0.
Antibodies—Anti-LANP antibody (antibody 3118) was generated by immunizing goat with bacterially expressed full-length LANP as a glutathione S-transferase fusion protein expressed and purified after subcloning into the bacterial expression vector pGEX5X3 (Amersham Biosciences). Anti-MAP1B C-20 antibody recognizes the light chain of MAP1B (Santa Cruz Biotechnology). The following antibodies to epitope tags were used: anti-myc epitope, clone 9E10 (Sigma); anti-FLAG, clone m2 or polyclonal F7425 (Sigma); and anti-HA, clone HA.11 (Covance); dilutions of 1:100 for immunofluorescence and 1:1000 for Western blotting.
Co-immunoprecipitation—Cells were transfected at 
80%
confluence on 150-mm dishes using 50 µg of DNA and LipofectAMINE Plus
reagent (Invitrogen). The light chain of MAP1B was expressed as an HA-tagged
fusion using the vector pCMV-HA; LANP was expressed as a FLAG-tagged fusion in
pFLAG CMV-2. Two days post-transfection cells were washed twice with
phosphate-buffered saline and then lysed in 3 ml of lysis buffer:
phosphate-buffered saline, 0.5% Nonidet P-40, 5 mM EDTA, and
protease inhibitors (Complete, Roche Applied Science) using the protocol
recommended in the antibodies protocol guide using the relevant antibody or
nonspecific immunoglobulins as controls (Clontech). Anti-HA
immunoprecipitation (IP) was performed with anti-HA beads (Sigma); anti-LANP
IP was performed by protein G beads coupled to FLAG M2 and 3118.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
Since LANP is a developmentally regulated protein in neuronal cells, with maximum abundance in the early postnatal life, we sought to determine whether its localization is altered depending on the process of differentiation, as has been observed for other proteins expressed at high levels during development (22, 23). In undifferentiated neuro2a cells, LANP typically is nuclear with little cytoplasmic staining. Upon differentiation by dbcAMP in the presence of low serum (18), LANP tends to be diffuse cytoplasmic, particularly in those cells with the most extensive neurites (Fig. 2, top panel). In addition to staining endogenous LANP with an LANP specific antibody, we also transfected epitope-tagged LANP into neuro2a cells so as to follow localization by an antibody specific to the epitope tag to rule out the possibility of nonspecific staining. Immunofluorescence microscopy once again revealed a dramatic alteration in LANP localization from the nucleus to the cytoplasm, often to the extent that the nucleus became completely devoid of LANP staining (Fig. 2, bottom panel). Approximately 80% of undifferentiated cells showed nuclear staining, while upon differentiation roughly the same percentage showed a cytoplasmic staining pattern for LANP. This translocation suggests that LANP shifts from the nucleus to the cytoplasm during neuritogenesis, where its cytoplasmic function(s) may be more critical.
|
LANP Interacts with MAP1B—To further delineate the neuronal properties of LANP, we decided to search for LANP interacting proteins expressed in neurons. To this end we performed a yeast two-hybrid screen using a mouse brain library and full-length LANP as bait. We identified MAP1B as a potential interacting partner. Three representative library clones of MAP1B that were fished out by the two-hybrid screening are shown in the schematic in relationship to the sequence of the rat MAP1B construct (Fig 3A). We also pulled out several clones corresponding to the C-terminal domain of MAP1A. These interacting clones were in general short. This result was not surprising in view of the fact that MAP1B and A share extensive similarity in their primary structure (80% identity in their last 100 amino acids). Intriguingly we also identified a tau isoform (GenBankTM accession number U12916 [GenBank] ) as an interacting bait. Since LANP had been demonstrated to interact with tau in biochemical assays we did not pursue the interaction of LANP with tau (15, 16, 24). Of our microtubule-associated interactors, we decided to pursue the MAP1B interaction further, as it would indicate that LANP has the intriguing property of binding to all classes of structural microtubule-associated proteins.
|
MAP1B, along with MAP1A, constitute the family of large molecular weight microtubule-associated proteins. Both these proteins have shared domains in addition to the C-terminal tail, some of which contain canonical repeats (Lys-Lys-Glu-X) involved in binding to microtubules (25, 26). The proteins are also processed similarly, undergoing proteolytic cleavage of a single precursor to produce a light chain and a heavy chain (27, 28). In the case of MAP1B the heavy chain is 243 kDa, while the light chain is 27 kDa; the cleavage site has been narrowed to within 40 residues of a proline-rich hydrophobic domain of the full-length MAP1B (29). One of our clones (clone 3-87; 242 residues) corresponded almost exactly to the complete light chain of MAP1B, beginning four residues downstream from the predicted cleavage event. This clone was the strongest interactor in our yeast two-hybrid screen, interacting more robustly with LANP than the other MAP1B clones picked up in the yeast two-hybrid that were either longer (clone 3-75, 559 residues) or shorter (clone 3-3, 168 residues) than the light chain. This suggests that LANP binds to the light chain per se (Fig. 3A).
Since we were keen to determine which domain of LANP mediates the interaction of LANP with MAP1B, we generated N- and C-terminal deletions of LANP. Using yeast two-hybrid assays we were able to demonstrate that the acidic C-terminal domain of LANP interacted more robustly with the light chain of MAP1B than did the N-terminal domain of LANP, the domain bearing the LRRs (Fig. 3B). Incidentally, it is the C-terminal domain that bears the nuclear localization signal. It should be mentioned in the context of discussing the domain structure of LANP that it is the N-terminal region, i.e. the leucine-rich region, that is responsible for the interaction of LANP with ataxin-1 (1).
LANP Interacts with the Light Chain of MAP1B in Vivo— Since the majority of the MAP1B fragments isolated by the yeast two-hybrid screen correspond to the C-terminal domain of the MAP1B precursor that eventually becomes the light chain, we sought to test the idea that the interaction of LANP is specific to the light chain of MAP1B.
We first performed immunfluorescence studies to look at the subcellular
localization of LANP in the presence of the MAP1B light chain
(Fig. 4). When transfected
alone, LANP localized predominantly to the nucleus in 80% of the cells.
In the small percentage of COS-7 cells in which LANP is cytoplasmic, the
staining is either diffuse or slightly vesicular, as was reported for CHO
cells (Fig. 4A)
(16). Incidentally, in our own
studies using CHO cells, we found that LANP (transfected or endogenous)
displays a predominant nuclear staining in addition to the lower intensity
cytoplasmic staining (data not shown). Discrepancy with earlier findings might
relate to the fact that the antibody used by Ulitzur et al.
(16) recognizes only a subset
of LANP in cells based on post-translational modifications (for instance
phosphorylation).
|
When co-tranfected with the MAP1B light chain, however, LANP relocated to
the cytoplasm in 80% of COS-7 cells. This relocation to the cytoplasm was
dramatic and in most of the cells LANP tends to be diffuse and cytoplasmic.
However, in a few cells the LANP distribution was clearly fibrillar, with both
MAP1B and LANP, co-localizing in a perinuclear filamentous pattern. This
pattern reflects the bundled and collapsed microtubule networks described
during earlier studies on its role in organizing tubulin networks (Ref.
17 and our own data not shown
using anti-tubulin antibodies) (Fig.
4). This suggests not only that the light chain of MAP1B interacts
with LANP, but that LANP can bind at least to some extent with the
microtubule-bound pool of the MAP1B light chain. In undifferentiated neuro2a
cells this relocation to the cytoplasm is even more dramatic with more than
90% showing a cytoplasmic distribution of LANP when MAP1B light chain is
overexpressed, reminiscent of the translocation of LANP seen during the
process of differentiation.
We next sought co-IP evidence. In Fig. 4H we demonstrated that we can co-immunoprecipitate the MAP1B light chain using antibodies targeted to LANP and also in the reverse direction, i.e. co-IP LANP when using antibodies directed against MAP1B light chain. The interaction between the two is thus fairly robust.
Finally, to determine whether the preference of LANP for the light chain
holds in cells, we used tagged constructs of MAP1B corresponding to the
full-length precursor and the heavy chain of rat MAP1B (95% identity to
mouse MAP1B) (Fig. 5 schematic)
and transfected cells with FLAG LANP along side either full-length MAP1B or
just with its heavy chain. We discovered that, much like the light chain of
MAP1B, full-length MAP1B also causes a dramatic shift of staining from the
nucleus to the cytoplasm, with a majority of cells (close to 70%) showing a
cytoplasmic staining (Fig. 5).
This was to be expected, since full-length MAP1B is cleaved into its heavy and
light chain in both neuronal and non-neuronal cells (17 and data not shown).
In contrast, expressing the heavy chain alone causes a less dramatic
transition from nuclear to cytoplasmic staining, although it did increase the
total number of cells showing a cytoplasmic staining from 20% to 40%. We
speculate that this increase might be due to alterations in the dynamics of
the endogenous light chain caused by the abundant quantities of newly
introduced heavy chain, although it is possible that there is an interaction
albeit less robust with the heavy chain as well. This is conceivable since the
heavy chain contains a microtubule binding domain consisting of several KKEX
and KKEE motifs, thought to contribute basic properties to the heavy chain and
that could also potentially interact with the acidic tail domain of LANP
(25).
|
LANP Modulates the Functions of MAP1B in Neuritogenesis—Recent evidence suggests that the ratio of light chain to heavy chain is under strict developmental control, mediated by differential proteolysis and clearance of these fragments to keep the light and heavy chains in a 6:1 to 8:1 molar ratio (30). We, therefore, tested whether different fragments of MAP1B might have differential effects on neurite outgrowth in neuro2a cells and whether the binding of LANP to the light chain could thereby affect neuritogenesis.
Full-length MAP1B, the light chain and the heavy chain each, had a different effect on the ability of neuro2a cells to extend neurites (Fig. 6). Although full-length MAP1B has been previously shown to have only modest effects on microtubule stability in non-neuronal cells (31, 32), it dramatically altered the morphology of transfected neuro2a cells. These cells became rounded (more than 80%) and were incapable of spreading out or extending neurites even after stimulation with dbcAMP, a potent inducer of neuritogenesis (up to 4 days after induction). Neither the light chain nor the heavy chain alone had such a significant effects on neuritic morphology (Fig. 6), although we noticed that the heavy chain had a greater inhibitory effect on neuritogenesis when compared with the light chain (80 and 60% of cells transfected with the light or heavy chain, respectively, were able to extend neurites). Thus, it appears that both the light chain and the heavy chain are presumably required to stabilize the microtubule cytoskeleton at the expense of neuritogenesis.
|
This inhibitory effect on neuritogenesis by full-length MAP1B now allowed us to test whether LANP modulates the ability of full-length MAP1B to suppress neurite formation induced by dbcAMP (Fig. 7). Co-transfecting LANP with full-length MAP1B significantly increased the ability of the cells to form neurites. On the other hand, LANP did not affect the ability of cells to express neurites when transfected with either the heavy chain or light chain alone, suggesting that the alleviation on the inhibition of neuritogenesis is because of its interaction with the MAP1B complex.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The MAP1 Proteins and Neurological Disease—MAP1B and MAP1A constitute the family of heavy molecular weight Type I microtubule-associated proteins. Both are expressed at high levels in the nervous system (33, 34), bind the acidic domain of tubulin with similar repeat sequences (Lys-Lys-Glu-X), and have several areas of homology in their primary structure (27, 28). The two proteins are reciprocally expressed over the lifespan of an organism, with MAP1B being expressed earlier during development, at a time when the cytoskeleton is quite dynamic. MAP1A typically is expressed later when MAP1B is down-regulated and is believed to confer stability to extended neurites (35–41).
Of the two microtubule-associated proteins MAP1A and MAP1B, the latter is better studied. Depleting MAP1B either by antisense approaches or by injecting antibodies prevents neurite outgrowth in tissue culture cells (42, 43). In Drosophila complete loss of MAP1B function causes embryonic lethality, whereas partial loss results in fewer and larger boutons at the neuromuscular junction (69, 70). Various lines of mice deficient in MAP1B have been generated, and their phenotypes variously manifest in abnormal brain architecture, delayed myelination, reduced axon caliber, tract malformation, and layer disorganization (44–47).
Intriguingly, MAP1B has been implicated in neuronal pathology in genetic diseases in mouse and man. In the fly, the fragile X-related gene (FMRP) appears to be a translational repressor of the MAP1B homolog, suggesting that MAP1B levels are dysregulated in human fragile X syndrome (48). These findings are consistent with microarray studies that identified MAP1B as an FMRP RNA target (49, 50). Perhaps the most compelling evidence that MAP1B, and particularly its light chain, is a hub for interactions with disease-causing proteins is the finding that the light chain of MAP1B interacts with gigaxonin, a protein that displays mutations in giant axonal neuropathy (51). This progressive, autosomal recessive disease afflicts the central and peripheral nervous systems and is characterized by aggregation of cytoskeletal components, particularly intermediate filaments, to such an extent as to cause an increase in axonal diameter. Surprisingly MAP1B has been implicated not just in these diseases that have onset early in life, but also in adult onset neurodegeneration including Alzheimer's disease and Lewy body Parkinsonism (52, 53).
As noted above, there are several levels of regulation of the MAP1 complex. Both MAP1A and B are cleaved to form a light and a heavy chain, both of which have microtubule-binding domains (17, 25, 26); the heavy chain is also phosphorylated (32, 43, 54–57). The light chains of MAP1B (also known as LC1) and MAP1A (also called LC2) share extensive similarity in their primary structure and are thought to be interchangeable with regard to their ability to interact with either of the heavy chains in the heavy molecular weight MAP complex (40). There is a third light chain, LC3, that is transcribed independently from a distinct genetic locus and can also interact with either heavy chain (58), although its contribution to the stability of tubulin networks is not yet clear.
LANP, Another Level of Regulation for MAP1B—Our findings suggest that there is another level of regulation of the MAP1B complex mediated by LANP that stems from the differential effects of the heavy chain, the light chain, and their precursor on the ability of neurons to extend neurites. Transfection of either the heavy chain or light chain alone allows neurite extension, but overexpression of full-length MAP1B inhibits neurite outgrowth. It is this inhibition that can be significantly relieved by co-transfecting LANP. The simplest interpretation of our findings might be that LANP binds to the light chain of MAP1B in the complex, thereby allowing greater flexibility of the neuronal cytoskeleton and thus neurite outgrowth. Consistent with this model is the finding that LANP does not modulate neurite outgrowth in the presence of either heavy or light chain in isolation, presumably because either chain alone would not have the stabilizing properties of a fully functional MAP1 complex. We have also tried to test for a more direct role for LANP in regulating the effect of MAP1B on neuritogenesis. Using the microtubule depolymerizing agent nocodazole we tested whether the normally stabilizing role of the light chain of MAP1B light chain described previously (17) can be altered by LANP. We could not, however, detect a major effect on microtubule networks (data not shown), although we cannot rule out the possibility that subtle effects on microtubule stability might only be evident in neurons over a prolonged period of time. We also sought to test the possibility that LANP binds to the full-length precursor regulating the endoproteolysis of the MAP1B precursor. We could not, however, quantify any significant alterations in the ratios of light chain to heavy chain in the presence and absence of LANP (data not shown).
Our results extend the results of Pfeffer and co-workers (15, 16, 24), who identified LANP as a modulator of microtubule-associated proteins, earning it the moniker mapmodulin. In their experiments they demonstrated that LANP binds to the free subunits of the structural MAPs, MAP2 and tau, and to the non-neuronal MAP, MAP4. Since LANP enhances vesicle transport, they hypothesized that LANP clears the microtubule tracks from associated proteins. Our findings indicate that LANP has the remarkable property of binding to all classes of structural MAPs, both the tau family of heat-stable MAPs (that also includes MAP2 and MAP4) and the high molecular weight heat-labile family represented by MAP1A and MAP1B, making LANP a potentially important regulator of microtubule function. The domain crucial for this interaction is the C-terminal domain of LANP that is rich in acidic residues and strikingly similar to the C terminus of tubulin that suggests a model where LANP and tubulin compete for MAPs (15). These results are in general consistent with such a model, although in a few cells LANP and MAP1B clearly display a microtubular pattern suggesting that LANP/MAP1B light chain complexes can in some instances still associate with the microtubule network. Additionally this acidic domain is shared by a variety of proteins, including the closely related leucine-rich family member APRIL or PAL31 that might also be implicated in similar functions (59, 60). It is an intriguing possibility that members of this LRR family might have distinct effects on cytoskeletal behavior and possess functions specific to different cell types and developmental time courses.
LANP, Neuronal Development and Neurological Disease— The role of LANP as a modulator of microtubule-associated proteins clearly has specific implications for neuronal development. LANP expression peaks during early postnatal life at a time when brain architecture is still being formed. But in addition to its role in development, LANP and MAP interaction could also be important for the neurodegenerative events in SCA1. Before the present studies, there had been no apparent link between the obvious neuritic pathology and the equally obvious nuclear events such as ataxin-1 accumulation. We speculate that mutant ataxin-1 binds to LANP in the nucleus, compromising the interactions of LANP with its cytoplasmic partners, i.e. MAPs, and the absence of these microtubule-stabilizing interactions leads to cytoskeletal derangements or abnormalities in subcellular trafficking. This is not only reminiscent of the havoc wrought in the hereditary tauopathies and Alzheimer's disease, where the MAP, tau, is implicated, but is also in keeping with the finding that MAP1B might be directly involved in the cytoskeletal derangements of giant axonal neuropathy (51). This raises the possibility that alteration of MAP behavior may be common to many neurodegenerative disorders, including SCA1. Indeed, these dystrophic changes of neurites are strikingly evident using antibodies to MAP1 light chain in SCA1 mouse models (data not shown). Moreover, with MAPs playing a pivotal role in modulating tubulin networks, LANP could be involved in a variety of microtubule-based functions, including maintaining cellular morphology, regulating the movement of mitochondria, lysosomes, peroxisomes, and vesicles, and in addition helping to maintain the integrity of the endoplasmic reticulum and the localization of messenger RNA (61–68). By engineering mice lacking LANP, it should be possible to test for a role of LANP in neuronal development and/or SCA1 pathogenesis.
|
FOOTNOTES |
|---|
To whom correspondence should be addressed: Davee Dept. of Neurology,
Northwestern University, Feinberg School of Medicine, Ward Ward 10-332, 303 E.
Chicago Ave., Chicago, IL 60611. Tel.: 312-503-4699; Fax: 312-503-0872;
E-mail:
p-opal{at}northwestern.edu.
1 The abbreviations used are: SCA1, spinocerebellar ataxia type 1; LANP,
leucine-rich acidic nuclear protein; LRR, leucine-rich repeat; MAP,
microtubule-associated protein; HA, hemagglutinin; IP, immunoprecipitation;
NLS, nuclear localization signal; DAPI,
4',6-diamidino-2-phenylindole.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Chen, B. Li, I. Grundke-Iqbal, and K. Iqbal I PP2A 1 Affects Tau Phosphorylation via Association with the Catalytic Subunit of Protein Phosphatase 2A J. Biol. Chem., April 18, 2008; 283(16): 10513 - 10521. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
-galactosidase staining
(Xgal, 6-h incubation) and relative growth by auxotrophic selection
on media deficient for Leu, Trp, His, and Ade (Growth, 4 days of
incubation). B, to delimit the domain of LANP responsible for
interacting with the light chain of MAP1B, we used a full-length LANP
construct (1) and deletion constructs that contained either the
complete leucine-rich region (2) or the acidic tail domain
(3). Deleting the acidic tail domain dramatically decreased LANP
interaction with the light chain of MAP1b. (Xgal, 3 h incubation;
Growth, 4 days of incubation) (+ = interaction; – = weak/no
interaction based on both calorimetric and growth assays).
