
View larger version (37K):
[in a new window]
|
Fig. 4. Phosphorylation of the Dup N terminus in vivo depends on cyclin E/CDK2. (A)
Immunoprecipitation (IP) of embryo extracts with polyclonal guinea pig Dup
antibody followed by western blot with affinity-purified polyclonal rabbit Dup
antibody detects 106 kDa, 105 kDa and 82 kDa isoforms (lane 2). (Lane 4) IP of
these three isoforms with Geminin. (Lane 1) Input, (lane 3) pre-immune. (B)
The 106 kDa isoform differs from the 105 kDa by a cyclin E/CDK2-dependent
phosphorylation. Western blot labeled with Dup antibody of extracts from
wild-type third instar larval brains (lanes 1, 2), hsp70:GAL4; UAS:cyclin
E (lanes 3, 4, 5), or hsp70:GAL4; UAS: Dacapo (lanes 6, 7, 8).
Lanes 4, 5, 7, 8: induced expression of UAS trangenes; lane 2, 5, 8: lambda
phosphatase (PPase) treated. (C) Cyclin E/CDK2 alters migration of Myc:FL-Dup
but not Myc:Dup 10(A). Transgene expression from Myc:FL-Dup (lanes 1-4) and
Myc:Dup 10(A) (lanes 5-8) was detected with Myc antibodies. Over-expression of
cyclin E (lanes 3, 4, 7, 8). Lanes 1, 3, 6, 8: phosphatase treated. Blots were
reprobed for lamin C as a loading control. (D) Abundance of FL-Dup (lanes 1-6)
and Dup 10(A) (lanes 7-12) at different times after a 30-minute heat pulse of
expression without (lanes 1-3, 7-9) or with (lanes 4-6, 10-12) coexpression of
cyclin E. (E) Quantification of Myc-tagged protein from the experiment shown
in D. Each point shows the average value for three replicates normalized
against the lamin C loading control and a linear regression curve.
|