IV- DISCUSSION
These research tasks relate to the study of the role of TLS in
the ways of indication of the rétinoïque acid. TLS is a
bifunctional protein in the mechanism of action of AR on the level of the
transcriptionnelle regulation and the post-transcriptionnelle regulation by AR.
Indeed, of the experiments of transactivation and delay on freezing highlight
that TLS takes part in the complex transcriptionnel of
hétérodimère RXR-RAR, called RANC allowing to increase the
transactivation dependant on AR on the target gene promoters. Initially, of the
techniques of transitory transfection in Cos-6 cells and myeloblastic cells
HL-60 show that TLS is able to increase the transcriptionnelle activity of the
natural promoter of RAR sensitive to the action of the rétinoïque
acid in the presence of rétinoïque acid. It had been previously
highlighted that TLS had a transcriptionnelle activity per. Indeed,
the N-final part of TLS implied in protein of fusion TLS/ERG is regarded as
pre-necessary for the factor of transcription ERG deteriorated in its potential
leucemogene by increasing its transcriptionnelle activity and/or by changing
its specificity on the level of genes target (Zinszner and coll, 1994 ;
Ichikawa and coll, 1999). Moreover, it was highlighted that the N-final part of
TLS amalgamated with the field of connection of the ADN of Gal4 has a strong
transcriptionnelle activity (Uranishi and coll, 2001).
In order to determine if TLS belonged to the RANC, of the
experiments of delay on freezing were carried out while using like probes
oligonucleotidic DR5. For this purpose, extracts of Cos-6 cells
transfectées by RAR, RXR or TLS were used. The results show that TLS
belonged to RANC. Another study carried out in vitro shows that this
interaction is direct and independent of the presence of the
rétinoïque acid and seems specific to the receivers with
rétinoïque acid RXR, the glucocorticoïdes, with the estrogens,
and with the thyroid hormone (Powers and coll, 1998). This direct interaction
is independent of the presence of the ligand. This observation is due to the
fact that the complex is formed between the field of connection of the ADN
(field C) and the area D of the nuclear receiver with the N-final part of TLS
(Powers and coll, 1998). The area C, made up of 66 amino acids, is preserved
among all the members of the superfamille of the nuclear receivers. It contains
two structures of « finger with
zinc » and corresponds to the field of connection of the
nuclear receiver with the ADN. The area D, also called zone hinge, is between
the field of connection to the ADN and the area E. It is subdivided in three
under-areas (D1, D2 and D3) of which the N-final D1 area, the most preserved,
contains many basic amino acids which would correspond to a signal of nuclear
localization. The central area D2 is more variable. It is interesting to note
that the interaction of TLS with the field of connection to the ADN of RXR does
not deteriorate the connection of the nuclear receiver on its specific brief
reply.
The function Co-activatrice of TLS on the level of the complex
transcriptionnel of the receivers to the rétinoïque acid could seem
paradoxical in the fact that the interaction of TLS with RXR implies its field
of connection of the ADN. However, this interaction is not destabilizer of the
complex protein/ADN. Quite to the contrary, since it is established that this
interaction protein/protein is stabilizing complex RXR-ADN (Powers and coll,
1998). Moreover, this interaction does not prevent the change of conformation
of the nuclear receivers. Thus, TLS by its action would stabilize the complex
Co-activator of the receivers to the rétinoïque acid. The principal
complexes Co-activators of the nuclear receivers include/understand the complex
Brg (SWI/SNF), Co-integrators CBP and p300, the family of the proteins p160,
the protein p/CIF and the complexes TRAP/DRIP/ARC.
TLS was already implied in the complex transcriptionnel of
another factor of transcription, NFB. Thus, TLS increases the transactivation
dependant on NFB induced by physiological stimuli such as TNF and IL-1
(Uranishi and coll, 2001). TLS thus acts as an Co-activator of various factors
of transcriptions.
Lastly, TLS shares structural characteristics with hTAFII68.
These proteins were found associated complexes TFIID (Bertolotti and coll, 1997
and 1998) and are implied in activation transcriptionnelle (Prasad and coll,
1994 ; Zinszner and coll, 1994 ; Bertolotti and coll, 1999 ;
Ichikawa and coll, 1999). TLS interacts in vivo with TFIID (Uranishi
and coll, 2001). Moreover, TLS is associated the ARN polII via its
field N-terminal (Yang and coll, 2000). The whole of these interactions could
make it possible TLS to establish the molecular link between factors of
transcription and the complex of initiation of the transcription.
The whole of these studies implies in an undeniable way TLS in
the transcriptionnelle regulation on the level of the ways of indication of the
rétinoïdes.
The second part of the research project relates to the
possible bond between the ways of indication of the rétinoïdes and
the post-transcriptionnelle regulation through the modulation of the alternate
épissage. Indeed, the fact that TLS, factor of épissage, is
implied in the ways of indication of AR through its direct role in the
transcriptionnelle coactivation dependant on the RANC encourages to study the
role of the rétinoïque acid and its receivers in the
épissage.
The experimental results show that TLS acts in vivo
on the selection of sites 5 ' of épissage alternate of ARN pre-m E1A in
the hematopoietic cells K562. The selection of distal site 5 ' of E1A
corresponding to the formation of the isoforme 9S is favoured with the
detriment of that of the proximaux sites 12S and 13S.
Certain data in vivo switch us on the role of TLS in the
regulation of the alternate épissage. TLS is associated RNPs by
in vivo forming a complex with the hnRNP A1, a factor of
épissage able to support the selection of distal site 5 ' of
épissage during an alternate épissage of any pre-m ARN (Uranishi
and coll, 2001). It was previously shown that TLS acted on the selection of the
distal site of E1A in the erythroblastic cells of Mouse IW1-32 (Hallier and
coll, 1998). The experiments carried out on the K562 cells transfectées
by E1A make it possible to highlight an increase in the rate of isoforme 9S in
the presence of AR and TLS. The functional interference between TLS and the
rétinoïque acid can be due to molecular interferences, i.e.
physical interactions protein/protein. This assumption can be based on the
direct interaction shown by preceding work. Consequently, these results
identify the rétinoïque acid as an actor implied in the regulation
of the épissage. It is the first time that it is highlighted the action
of a hormone on the épissage. This work makes it possible to define a
new level of regulation of the ways of indication of the
rétinoïdes.
The fact that TLS interacts with RXR, TR and GR. make it
possible reasonably to think that this phenomenon is found on a more general
level.
The épissage of the ARN, critical stage of the form of
genes, is regarded more and more as an event Co-transcriptionnel. Experimental
obviousnesses state from now on that stages of transcription, it
« capping » and the polyadenylation, are closely related to
the ARN polII through its association with the factors implied in these
mechanisms molecular (Cho and coll, 1997 ; Hirose and coll, 1998). The
proteins of regulation of the épissage, the proteins SR, are associated
the ARN polII. However, the means by which this association is carried out were
not identified. The Co-activator transcriptionnel p52 is able to interact with
the protein SR, ASF/SF2. p52 thus acts as an adapter in order to coordinate the
transcription and the épissage (Ge and coll, 1998). TLS also interacts
with the ARN polII and the proteins SR (Yang and coll, 2000). TLS could thus
act as a recruteuse molecule of the factors of regulation of the
épissage SR towards the ARN polII, thus coupling the transcription with
the épissage. TLS could also act directly on the épissage by its
interaction with ARN (Lerga and coll, 2001).
TLS contains three fields potentially implied in the
connection with the ARN, RGG1, RRM and RGG2-3. The mechanism implied in the
recognition of the ARN by TLS is still unknown. The secondary structure of the
ARN represents a significant part of the interactions protein/ARN. The selected
sequences are fixed at TLS with affinities of about 250 Nm with 600 Nm, Kd of
250 Nm corresponding to the complex ggugARN/TLS (Lerga and coll, 2001).
Consequently, TLS seems to be a protein having a weak affinity for its sequence
ARN. Thus, it is not excluded that Kd of in vitro given complex
TLS/ggugARN is far away from Kd from a complex ARN/protein on the level of
the spliceosome. However, a weak affinity of a factor of épissage for
its target sequence ARN could represent a condition compatible with the
assembly of « spliceosome » which is held through the
exchange and replacement of a great number of proteins on the level of the
substrate, the ARN pre-Mr.
The functions of TLS in the regulation of the transcription
and the épissage make it possible to consider the consequences of a
deterioration of TLS in certain pathologies. In the leukaemic cells
myéloïdes having the translocation T (16; 21) and in the cells of
liposarcome in the translocation T (12; 16), only one allele are stopped
whereas the other allele is intact (Crozat and coll, 1993 ; Rabbitts and
coll, 1993 ; Yamamoto and coll, 1997). These observations suggest a role
of protein of generated fusion of dominant type negative. The protein of fusion
TLS/ERG, observed in human leukemia myéloïde T (12;16), in vivo
pertuberait the épissage of E1A observed in cells HeLa (Yang and
coll, 2000). It should be noted that TLS was not considered able to act on E1A
in these cells, which seems erroneous starting from other observations carried
out (F. Moreau-Gachelin, personal communication). Nevertheless, TLS is able to
inhibit the function of the TASR on the épissage of ARN pre-m E1A (Yang
and coll, 2000). Moreover, it is interesting to note that the épissage
CD44 is deteriorated by the presence of protein of fusion TLS/ERG in stable
clones generated starting from cells of the line K562 (Yang and coll, 2000).
Gene CD44 codes for a molecule of adhesion made up of ten exons
constitutive and ten let us exons variable. Various combinations formed by let
us exons variable is at the origin of a large variety of isofomes of
épissage of CD44 which differ on the level from their extracellular
field. The abnormal épissage of ARN pre-m CD44 was found in various
solid tumors and of leukemias (Cooper and coll, 1995). Moreover, It was
suggested that the variations in proteinic rate of SR according to various
stages' of the development of the breast cancer can be directly related to the
variations of various the isoformes of CD44 (Stickeler and coll, 1999). Two
mechanisms are planned to explain the negative effect dominating of TLS/ERG
over the épissage of CD44. The first considers that TLS clinging to the
specific isoformes CD44, TLS/ERG would block this way of regulation, at the
origin of a premature degradation of the pre-m ARN of CD44 not spliced
completely. In the second mechanism, if one of the ways of regulation is
blocked and that another way of regulation of the alternate épissage
exists, the inhibition of the first way by TLS/ERG could encourage the
épissage CD44 through another way of regulation, thus increasing the
risk to generate an aberrant épissage.
If TLS is expressed in way ubiquitaire, it is interesting to
note differences in expression in the normal and pathological hematopoietic
cells. The cells hematopoietic stocks purified of blood of cord have lower
rates of TLS compared with the cells myéloïdes. In addition, an
important expression of TLS was highlighted in cells of LAM (Millets and coll,
2000). It is interesting to note that TLS was also identified by its rate of
expression decreased in cells HL-60 treated by the rétinoïque acid
during one hour. Moreover, the expression of TLS is strongly decreased in cells
HL-60 during the granulopoïèse induced by the DMSO or the ATRA.
There is a specificity with the granulopoïèse since differentiation
monocytaire induced by the TPA or the D3 vitamin does not act on the expression
of TLS (Millets and coll, 2000). This reduction in expression is associated the
fall of the cellular proliferation in the various LAM. The whole of these
observations suggests that the expression of TLS is a key regulator of the
myélopoïèse where a strong rate of expression of TLS
supports the cellular proliferation with the detriment of differentiation.
The prospects for this work are articulated on two axes. The
first relates to the fundamental molecular mechanisms which imply TLS in the
ways of indication of the rétinoïque acid. We plan to study TLS in
the regulation of the épissage constitutive and the alternate
épissage into 3 ' of épissage. The effect of AR on the
épissage of E1A will be also studied through the roles of RAR and RXR.
It is also a question of determining with precision the association of TLS in
the complex transcriptionnel of AR. Thus, of the experiments of identification
of the direct interactions between TLS and RAR will be carried out knowing that
they were highlighted with RXR (GST sweater down and
Co-immunoprécipitation). We want to study the coordination of the
épissage and the transcription to know if these mechanisms are dependant
or independent one of the other (a construction joining together genes E1A and
the luciférase under the dependence of a transferred promoter sensitive
to AR, not allowing the RANC to fix itself to increase the transcription). The
biological role of TLS will be determined by its transitory invalidation
(RNAi). The second axis consists in confirming the bond between TLS and
pathology. It is a question of including/understanding the consequences of the
deterioration of the expression of TLS and its targeting within the framework
of a new therapeutic and diagnostic futurology.
| 
