The role of p19 and p21 H-Ras proteins and mutants in miRNA expression in cancer and a Costello syndrome cell model

P19 overexpression regulates specific miRNA expression

We have shown previously that cell growth and metastatic genes are regulated by p19
7]. In this latter work we compared p19 expression with the specific p19mut expression
(mutation not seen in p21), where p19mut was a p19 mutant (W164A) that prevented p19-RACK1
and p19-p73? binding 7]. These previous analysis are further complemented here by checking whether p19 can
regulate other RNA genes by studying miRNA expression in cells overexpressing p19
or p19mut. First, we incubated miRNA micro-arrays with RNAs from cells overexpressing
p19 wild-type protein versus RNAs from cells overexpressing empty vector. A second incubation was performed with
RNAs from cells overexpressing p19mut in a similar way. Afterwards, both results were
mathematically compared, and miRNAs that are upregulated by p19 overexpression but
not by 19mut overexpression were selected (see Fig. 1a). Candidate miRNAs were found to be miR-342, miR-206, miR-330, miR-138, and mirR-99b
(Fig. 1a and Additional file 1), which vary with p19 but not with the specific p19mut overexpression. We consider
that the values obtained with p19mut overexpression could be considered the miRNAs
basal levels respect p19 wild-type protein (see asterisks values in Fig. 1a and Additional file 1). The most relevant observation of this assay is that the expression of very few
miRNAs varies upon overexpression of p19 and not by p19mut overexpression and so,
these miRNAs might be also related to the pathway process activated by RACK-p19 and
p73?-p19 interactions. miRNA upregulation by p19 H-Ras was re-validated by RT-PCR
with a specific Taqman assay for mature miR-206, miR-342, miR-138 and miR-330 and
was found to increase 2-, 1.6-, 16- and 2.5-fold, respectively, upon overexpression
of p19 in three independent experiments and quadruplicate analysis. miR-206 has been
reported to be the miRNA whose expression is most downregulated in metastatic cells
and has also been shown to regulate cell migration and morphology 23]. Our previous work showed that p19 H-Ras induces a G1/S phase delay, thereby maintaining
cells in a reversible quiescence state 7]. Taken together, these findings prompted us to study the contribution of miR-206
upregulation to the G1/S delay, both of which are induced by p19. Figure 1b shows that anti-miR-206 partially antagonizes the effect of p19 on G1/S phases and
restores the G2 phase, thus indicating that miR-206 upregulation partially contributes
to the G1/S delay observed upon p19 overexpression.

Fig. 1. Anti-miR-206 partially antagonizes the effect of p19 on G1/S and G2 phases. a Fold change is the value used to measure overexpression of pRK5-p19 as compared first
to pRK5 empty vector and then to overexpression of pRK5-p19mut for miRNA microarrays.
(*) indicates the value obtained with pRK5-p19mut. These miRNAs varies upon overexpression
of 19 and not by p19mut overexpression. Additional gene information is available in
Additional file 1 (P??0.01). RT: miR-206 activation was confirmed by Taqman RT-PCR with a specific Taqman
assay for mature miR-206 (Applied Biosystems). b HeLa cells were first transfected with the pSuper-GFP vector containing the sequence
of anti-miR-206 and incubated for 2 days under cell-culture conditions. Cells were
then re-transfected with GFP-p19 vector and analyzed by FACS (fluorescent-activated
cell sorting) the following day. A representative flow cytometry histogram with GFP-p19,
pSuper-GFP anti-miR-206, GFP-p19?+?pSuper-GFP anti-miR-206 or GFP only (negative control)
is shown in the upper portion, whereas G1, S, and G2 percentages are given below (*P??0.001 as compared G1 and G2 to GFP-19 versus GFP-p19?+?anti-miR-206). HeLa cells
line were used in all the experiments stated above

H-Ras mutants alter the H-Ras splicing rate

We have previously demonstrated that SR proteins (SFRs) activate inclusion of the
alternative exon IDX and subsequently p19 expression 6], 24]. p68 RNA helicase (DDX5) was also shown to inhibit p19 expression and reduce IDX
inclusion, whereas SC-35 (SFRS2) and SRp40 (SFRS5) strongly increase IDX inclusion
in vivo6], 24]. We analyzed the obtained miRNAs by the TARGETSCAN tool and observed that miR-330
may target SC-35 and many other SR proteins as well as an UsnRNP core protein, mir-342
may target SF4, and miR-206 may target p68 RNA helicase and many other SR proteins.
As splicing factors are also targets of the miRNAs upregulated by p19, therefore we
decided to study whether the overexpression of p19 can also affect its own alternative
splicing. Thus, two Taqman RT-PCR assays were designed to map total/endogenous H-Ras
mRNA (directed to E3-E4A exons) and the endogenous p19 level and the endogenous/total
p21 H-Ras ratio determined in HeLa cells transiently overexpressing p19 (see Fig. 2a). Figure 2b shows that overexpression of p19 increases total endogenous H-Ras and endogenous
p21 expression by 5.3- and 3.9-fold, respectively, and that this activation is reverted
to endogenous basal levels by p19mut. The splicing rate for endogenous p21 formation
was slightly reduced in cells overexpressing p19 (Fig. 2b) but increased in the presence of p19mut (10 % higher than the negative control),
thus indicating that p19mut can alter the p19:p21 ratio by acting on the H-Ras alternative
splicing processes. In contrast, p19 was observed to directly regulate total mRNA
H-Ras expression but not to significantly alter the splicing rate of H-Ras pre-mRNA,
thus indicating that the p19:p21 ratio is maintained. Furthermore, we showed that
the overexpression of the strong p21 mutant (p21Q61L) also regulated this alternative
splicing, increasing the amount of p19 mRNA (Fig. 2c).

Fig. 2. Gene expression and alternative splicing regulation by H-Ras proteins. a Scheme of the Taqman assays used to determine endogenous p19, p21, and endogenous
total H-Ras mRNA expression. E3-IDX, E3-E4A, and E4A-E4B recognize endogenous p19
H-Ras, p21 H-Ras, and both p19 and p21 H-Ras, respectively. Taqman assays were performed
with E3-E4A and E4A-E4B for overexpressed p19 and with E3-IDX and E4A-E4B for overexpressed
p21mut. b Endogenous total H-Ras (Taqman E4A-E4B) and p21 (Taqman E3-E4A) mRNA levels change
upon overexpression of p19 and p19mut. The results are presented as fold changes with
respect to basal levels obtained for empty vector transfections, which were set to
1 (no change). Graphic: the percentage of alternative splicing p21/endogenous total
H-Ras (Taqman E3-E4A/Taqman E4A-E4B) was obtained for each individual experiment and
standard deviations of three separate experiments calculated. (?) is the % p21 in
transfections with empty vector set to 33 %; (?) p19 and (?) p19mut, % is the p21
in HeLa cells overexpressing p19 and p19mut, respectively. (?) p19, P?=?0.07; (?) p19mut, P?=?0.08. c Endogenous total H-Ras (Taqman E4A-E4B) and p19 (Taqman E3-IDX) mRNA level changes
upon overexpression of p21Q61L. The results are presented as fold changes with respect
to the basal levels obtained for empty vector transfections, which were set to 1 (no
change). Graphic: the percentage of alternative splicing p19/total H-Ras (Taqman E3-IDX/Taqman
E4A-E4B) was obtained for each individual experiment and standard deviations of three
separate experiments calculated. (?) is the % of p19 in transfections with empty vector
was set to 15 %; (?)p21mut is the % of p19 after expression of the p21mut (p21Q61L).
P?=?0.09. Hela cells line were used in all the experiments stated above

P19 alters miRNAs expression but not the cell growth in H-Ras
^(?/?)
KO cell lines

In order to further understand how the alternative splicing of H-Ras alters miRNAs,
we differentially expressed pEGFP-p19 or pEGFP-p21 in H-Ras
^(?/?)
KO MEFs and analyzed their effect on miR-206 and miR-138. Figure 3a shows that p19 has a large effect on miR-206 expression whereas pGFP-p21 increases
it only slightly. pEGFP-p19 also has a large effect on miR-138 in both KO MEFs and
HeLa cells (Fig. 3b, c, respectively), thus indicating that the alternative splicing of H-Ras towards p19
or p21 affects the expression of several miRNAs. Figure 4 shows that pEGFP-p21 increases cell growth in H-Ras
^(?/?)
/N-Ras
^(?/?)
DKO MEFs whereas pEGFP-p19 does not. This finding is in accordance with our previous
results, where we showed that p19 causes a G1/S delay in HeLa cells.

Fig. 3. P19 and p21 H-Ras differentially regulate miR-206 and miR-138. a and b show KO H-Ras
^(?/?)
fibroblasts stably expressing pEGFP (negative control), pEGFP-p19 or pEGFP-p21. The
regulation of miR-206 (a) and miR-138 (b) was analyzed in these cells with specific miRNA Taqman assays. c Regulation of miR-138 in transient transfections in HeLa cells

Fig. 4. P19 H-Ras does not activate cell growth. DKO H-Ras
^(?/?)
/N-Ras
^(?/?)
fibroblasts that stably express pEGFP (negative control), pEGFP-p19 and pEGFP-p21
were studied by direct cell-proliferation assay. Cells were harvested and plated in
96 wells (10,000 cells/well) on six microplates in sextuplicate and incubated at 37 °C,
5 % CO
₂
with DMEM/10 % FCS. After the desired time, the microplates were washed and frozen.
Cells were quantified with the green fluorescent dye CyQuant (Invitrogen) according
to the manufacturer’s instructions. Fluorescence measurements were performed using
a microplate reader with excitation at 485 nm and detection at 530 nnm

The p19 CS mutant G12S alters miRNA expression

Aoki et al. 25] reported that CS is a result of mutations in the H-Ras gene. The most common mutation is found on codon 12 with the amino acid change G12S
13]. All human cells contain both p19 and p21 H-Ras proteins at different levels as a
consequence of the alternative splicing, therefore when H-Ras gene is mutated in CS patients, both these proteins p19 and p21, are present in their
mutated forms. We obtained pRK5 plasmids containing p19(G12S) and determined how this
mutant alters a selected group of miRNAs in HeLa cells. These miRNAs included miR-342
and miR-330, which are already known to be upregulated by p19 overexpression (Fig. 1), miR-126 and miR-335, which significantly reduce the ability of CN34-LM1 and CN34-BoM1
cells to metastasize to the lung 23], miR-374, which is known to be associated with aggressive small cell lung cancer
26], and let-7, which targets Ras genes 27]. Figure 5 shows that p19(G12S) clearly upregulates all these miRNAs (column 3), thus indicating
that the p19 CS mutant contributes significantly to the syndrome by altering specific
miRNA levels. We also analyzed how p21 mutants affect the same set of miRNAs by overexpressing
PRK5-p21, pRK5-p21Q61L and pRK5-p21G12S in HeLa cells. Figure 5 shows that p21 and p21 mutants also alter expression of these miRNAs, although to
a lesser extent than p19G12S. Thus, p21 was found to upregulate miR-374, miR-126,
miR-342, miR-335 and let-7 but not miR-330, and p21G12S was found to have the same
effect as p21 on miR-374, miR-342 and miR-126. In contrast, p21G12S (column 5) was
found to have no effect on miR-335 and let-7 but upregulates miR-330, whereas p21
does not. This indicates that p21 and p21 mutants do not have such a linear effect
as p19 and p19G12S but that both p19 and p21 CS mutants alter the expression of several
miRNAs, which may contribute to the development of CS.

Fig. 5. P19 H-RasG12S CS mutant clearly upregulates several miRNAs. HeLa cells transiently
expressing: 1) pRK5 (negative control); 2) pRK5-p19; 3) pRK5-p19G12S; 4) pRK5-p21;
5) pRK5-p21Q61L; and 6) pRK5-p21G12S. MiRNAs levels were analyzed with specific miRNAs
Taqman assays. Three independent experiments were done, each of them per quadruplicate
(n?=?12, and statistic significative difference calculated by Anova, one way analysis
of variance, p 0.05, p value?=?0.036), GraphPad Prism version 4.0