Biology Reference
In-Depth Information
Invariant signaling pathways in spliceosome assembly
Structural analyses resulted in a complete coverage of the network by 71 T-invariants (Supplementary
Table S2). All T-invariants describe at least one partial pathway within the spliceosomal assembly process
and are biologically meaningful. 12 T-invariants (17%) are trivial as they describe solely the in- and
efflux of the DExD/H box helicases Prp5, Prp28, the proteins hPrp6, SKIP and hDib1, the splicing factors
SF1, ASF/SF2, SC35 and the SF3b components SF3b10, SF3b14a, SF3b14b and SF3b49. However,
these proteins also constitute a set of spliceosomal components for which it is possible to narrow the
putative point of pullout from the assemply process, which so far is unknown for the majority of proteins
participating in spliceosome assembly. For example, the four smaller SF3b proteins are known to be
required for SF3b formation, but are not detected at the stage of C-complex formation (see supplements
of Makarov
et al.
, 2002).
Due to the non-decomposability of T-invariants, the removal of a single reaction from an invariant
signaling pathway disables the pathway. In case of spliceosome assembly, this does not mean that
the entire assembly process is stalled. The model clearly illustrates that the different results from
experimental studies shape up to a network with an inherent redundancy to sustain specific check points
which represent crucial intermediate assembly stages. For example, a critical step of early spliceosome
assembly is the donor or 5' splice site recognition, for which several parallely occurring signaling
pathways were modeled. These pathways contribute to the intermediate stage of E-complex assembly
and are reflected in the model by several sets of T-invariants, which are listed below. Note that enumerated
T-invariants are labeled by “i”. T-invariants in brackets describe spliceosomal subpathways that involve
5'ss recognition, but do not proceed via the productive branch of C-complex formation that results in
exon ligation:
1. T-invariants i13, i14 (i68, i69) describe the U1 snRNP independent 5'ss recognition (see Figs. 8
and 9, Supplementary Table S2, for i13 and Tab. 2). The central function of this branch is transition
t16.U1 indep 5ss act
, which models the activity of the SR protein SC35. The presence of SC35 is
sufficient to define a 5'ss in absence of a functional U1 snRNP [Crispino and Sharp, 1995], and initi-
ates contacts to the BP occupying proteins SF1 and U2AF. This can result in selection of competing
5'ss, which render this pathway a potential candidate for the production of alternatively spliced
mRNAs [Tarn and Steitz, 1994]. The remaining transitions of this branch are
t31.U2 BPS bdg2
and
t32.U6
5ss bdg2
, which feed U2 snRNP and U4U5U6 tri-snRNP respectively and proceed the
assembly pathway to the B-complex stage. Further differences in the T-invariants, sharing this
otherwise unique branch of 5' selection, exist in two different ways of A-complex assembly via
early
t13.17S U2 matur2
(i13, i68) and late
t22.17S U1 matur1
(i14, i69) action of the enzyme
SF3b125. However, for i14 it is not clear at which time SF3b125 leaves the assembly process.
Hence, this T-invariant additionally includes the input reaction
t109.SF3b125 in
. Another variation
between these similar T-invariants is the different proceeding during C-complex assembly after the
remodeling step
t93.U2 3ss U6 5ss U5 remod
. Either a productive spliceosome is formed via one
type of pathway (i13, i14, Figs. 8 and 9) or the premature disassembly occurs via another type of
discard pathway (i68, i69).
2. T-invariants i15-i20 (i62-i67) describe the ASF/SF2 dependent 5'ss recognition. They model the
5'ss recognition by contacts of the U1 snRNP factor U170K with the exon-bound splicing factor
ASF/SF2 (
t12.ASFp U170K bdg
) via RS domains [Cao and Garcia-Blanco, 1998]. Additionally,
contacts of U1C to another splicing factor, TIA1, which binds intronic elements close to the 5' splice
site (
t9.U1C TIA1 bdg1
), can direct the splice site selection [Puig
et al.
, 1999; Del Gatto-Konczak
