MicroRNA-378 promotes cell survival, tumor growth, and angiogenesis by targeting SuFu and Fus-1 expression
- Sunnybrook Health Sciences Centre and Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada M4N 3M5
-
Edited by Kevin V. Morris, The Scripps Research Institute, La Jolla, CA, and accepted by the Editorial Board October 30, 2007 (received for review July 25, 2007)
Abstract
MicroRNAs are single-stranded RNA of 18–24 nt expressed endogenously that play important roles in cancer development. Here, we show that expression of miR-378 enhances cell survival, reduces caspase-3 activity, and promotes tumor growth and angiogenesis. Proteomic analysis indicates reduced expression of suppressor of fused (Sufu), a potential target of miR-378, which was confirmed in vitro and in vivo. Expression of a luciferase construct containing the target site in Sufu was repressed when cotransfected with miR-378. Transfection of a Sufu construct reversed the effect of miR-378, suggesting an important role for miR-378 in tumor cell survival. We also discovered that miR-378 targets Fus-1. Expression of luciferase constructs harboring the target sites in Fus-1 was repressed by miR-378. Fus-1 constructs with or without its 3′ UTR were also generated. Cotransfection experiments showed that the presence of miR-378 repressed Fus-1 expression. Suppression of Fus-1 expression by siRNA against Fus-1 enhanced cell survival. Transfection of the Fus-1 construct reversed the function of miR-378 in cell survival. Our results suggest that miR-378 transfection enhanced cell survival, tumor growth, and angiogenesis through repression of the expression of two tumor suppressors, Sufu and Fus-1.
Over the past few years, microRNAs (miRNAs) have emerged as a prominent class of gene regulators (1). miRNAs are single-stranded RNA of 18–24 nt in length and are generated by an RNase III-type enzyme from an endogenous transcript that contains a local hairpin structure (2, 3). miRNA functions as a guide molecule in posttranscriptional gene silencing by partially complementing with the 3′ untranslated region (UTR) of the target mRNAs, leading to translational repression (4). In animals, miRNA genes are transcribed to generate long primary transcripts, which are processed by the RNase III-type enzyme Drosha to produce precursor miRNAs (premiRNAs) in the nucleus (5). PremiRNAs are then exported to the cytoplasm by exportin-5 (6). After arrival in the cytoplasm, premiRNAs are subjected to secondary processing by Dicer, a cytoplasmic RNase III-type enzyme (7, 8). miRNAs normally repress gene expression, opposing the activity of transcription factors, which initiate gene expression (9). By silencing various target mRNAs, miRNAs have key roles in diverse regulatory pathways, including control of development (10), cell differentiation (11), apoptosis (12–14), cell proliferation (15), division (16), protein secretion (17), and viral infection (18, 19). Most importantly, miRNAs have been known to play roles in cancer development (20–23). Previous studies have shown that miR-378 is expressed in a number of cancer cell lines (24) and is involved in the expression of vascular endothelial growth factor (25). To understand the biological functions of miR-378, we have generated a miR-378 expression construct for functional studies. Here, we show that stable expression of miR-378 plays a role in cell survival, tumor growth, and angiogenesis.
Results and Discussion
Expression of miR-378 Enhances Cell Survival.
We have generated a construct that can express the premiRNA of miR-378 and GFP with an antibiotic selection marker [Fig. 1 A; all primers used are listed in supporting information (SI) Table 1]. The advantage of this construct is the stability of miRNA expression after stable transfection. Furthermore, transfection of this construct allows selection with neomycin and rescue of positive clones by monitoring GFP or by cell sorting pooled cells. We have detected the expression of pre-miR-378 by RT-PCR, using RNA prepared from U87 cells stably transfected with miR-378 (Fig. 1 B). The pre-miR-378 was then successfully processed to mature miR-378, as detected by RT-PCR with primers designed by us (Fig. 1 C, SI Fig. 6A, and SI Table 1). The advantage of using RT-PCR for mature miRNA detection was the limited primer annealing, which could avoid nonspecific detection. In the reaction, we only designed 12-base pair annealing for reverse transcription and PCR. This should have greatly reduced nonspecific interaction that Northern hybridization normally produces. Increased expression of miR-378 was confirmed by real-time PCR, using kits available recently (see Methods, Fig. 1 D, and SI Fig. 7). To assure that processing of the expressed miR-378 did not interfere with the RNAi/miRNA pathways, we analyzed a few endogenous miRNAs and observed that transfection with miR-378 had no effect on these miRNAs (SI Fig. 6B).
Construction and expression of miR-378. (A) Diagram of a construct expressing GFP, a neomycin-resistance gene, and two hairpin structures of premiR-378. The boldface and capitalized letters represent two restriction sites BglII and HindIII. The boldface lowercase sequence represents an artifact sequence inserted between two premiRNAs. Six “t”s were added to terminate transcription. A few lowercase letters represent changes of nucleotides that are not part of the mature miR-378 but are useful for PCR purpose. (B) RT-PCR of premiR-378, using RNA prepared from U87 cells stably transfected with miR-378 and a control vector, using two primers miR-378N and miR-378C that amplify the premiR-378 sequence in A, confirming expression of premiR-378. As a control, Gapdh was amplified with two primers huGapdh421F and huGapdh720R. (C) RT-PCR of mature miR-378 using RNA prepared from U87 cells stably transfected with miR-378 and a control vector, confirming proper processing of miR-378. (D) Real-time PCR analysis confirmed expression of mature miR-378. **, P < 0.01. Error bars indicate SD (n = 3).
Although miR-378 has been detected in a number of human cancer cell lines (24), its function is not known. We analyzed the effect of miR-378 expression on mediating cell activity. A number of cell lines and a pooled cell line were selected in U87 cells transfected with miR-378 or a control vector. The cells were maintained in serum-free conditions or in serum-containing conditions and allowed to overgrow, resulting in extensive cell death. Transfection with miR-378 enhanced cell survival compared with control in all serum conditions by microscopic examination. Because most of the selected cell lines and the pooled cells exhibited similar phenotypes, only the results of the pooled cell line were shown (Fig. 2 A and SI Fig. 8A). The following experiments were performed by using the pooled cell line. The miR-378-transfected cells were cultured continuously, and the cells exhibited greater capacity for survival than the vector-transfected cells (SI Fig. 8B). We also carried out similar experiments, using the breast carcinoma cell line MT-1 and obtained similar results (SI Fig. 8C), suggesting a conservation of miR-378 functioning. Analysis of cell death in the detached and adherent cells revealed that transfection with miR-378 reduced cell death (SI Fig. 9A Upper) and increased cell survival (SI Fig. 9A Lower).
Cell survival affected by miR-378. (A) U87 cells transfected with miR-378 or a control vector were maintained in serum-free conditions. (Upper) Cell survival was monitored with a fluorescent microscope. Transfection with miR-378 enhanced cell survival. (Lower) Surviving cells were harvested from serum-free or serum-containing conditions and counted for statistical analysis. Asterisks indicate significant differences. **, P < 0.01. Error bars indicate SD (n = 4). (B) Caspase 3 activity was determined in cells transfected with miR-378 or a control vector in 1% or 10% serum-containing conditions. Asterisks indicate significant differences. **, P < 0.01. Error bars indicate SD (n = 6). (C) U87 cells were transiently transfected with miR-378, anti-miR-378, or a control vector, followed by culturing in serum-free conditions for 5 days. (Upper) Cell survival was assayed by trypan blue staining and cell counting. *, P < 0.05. Error bars indicate SD (n = 3). (Lower) Typical photographs.
To examine whether the cells died through apoptosis, we performed caspase 3 activity assays in the cells incubated in 1% or 10% serum-containing conditions. We detected decreased caspase 3 expression in the miR-378-transfected cells compared with the vector-transfected cells (Fig. 2 B). Because caspase 3 activity is essential for apoptosis, decreased caspase 3 activity suggests a decrease in apoptosis in the miR-378-transfected cells.
To confirm the function of miR-378, U87 cells were transfected with a construct expressing an antisence sequence against miR-378 (anti-miR-378) followed by culturing for 5 days. Cell survival experiments showed that, although miR-378 expression increased cell survival, expression of the anti-miR-378 construct decreased cell survival significantly (Fig. 2 C). Longer culturing times produced similar results (data not shown).
Expression of miR-378 Promotes Tumorigenesis and Angiogenesis.
When U87 cells transfected with miR-378 or a control vector were maintained in serum-free conditions in Petri dishes, we observed a dramatic change in cell morphology (SI Fig. 9B). The miR-378-transfected cells rounded up, whereas the vector-transfected cells were still spread on the dishes. The ability of the miR-378-transfected cells to survive without attachment, its increased survival, and its decreased apoptosis are all characteristic of fast-growing tumor cells. Both groups of cells were therefore injected s.c. into nude mice. Two weeks after the injection, we detected a clear difference in tumor size. By 4 weeks, mice injected with the miR-378-transfected cells had much larger tumors than mice injected with the GFP-transfected cells (Fig. 3 A Left). Analysis of the tumor growth curves indicates that the difference was significant 2 weeks after cell injection (Fig. 3 A Right). These data indicate that expression of miR-378 greatly promoted the process of tumor formation, because wild-type U87 cells normally take 4 weeks to form visible tumors. In an independent experiment, using a different batch of miR-378-transfected cells, similar results were obtained (data not shown). The tumors were removed and sectioned for immunohistochemistry, using anti-CD34 antibody. The tumors formed by miR-378-transfected cells contained larger blood vessels than those formed by the vector-transfected cells (Fig. 3 B), implying that miR-378 played a role in blood vessel formation. Although the number of blood vessels per unit field did not seem to increase, it is probable that the total number of blood vessels in the larger tumors had to be higher to allow expansion of the tumor. The formation of the large vessels may be important in facilitating such expansion. These results are in agreement with our previous studies that miR-378 binds to VEGF 3′ UTR competing with miR-125a for the same seed region and promotes VEGF expression (25).
Tumor formation and angiogenesis affected by miR-378 over expression. (A) U87 cells transfected with miR-378 or a control vector were injected s.c. into nude mice. (Left) Four weeks after the injection, mice were photographed and killed. (Right) Tumor sizes were measured and tumor growth curves were obtained. Asterisks indicate significant differences. *, P < 0.05. Error bars indicate SD (n = 3). (B) Tumors formed by cells transfected with the miR-378 construct or a control vector were subjected to immunohistochemistry probed with anti-CD34 antibody. Expression of the miR-378 construct elevated sizes of blood vessels. (C) Expression of miR-378 was analyzed by real-time PCR in RNAs isolated from papillary thyroid tumor and adjacent normal tissue. *, P < 0.05. Error bars indicate SD (n = 3).
We analyzed potential correlation of miR-378 overexpression in primary tumors. RNAs isolated from benign tumor papillary thyroid carcinoma and adjacent tissue (Ambion; catalog no. 7240) were analyzed for miR-378 expression, because the adjacent tissues to benign tumors are relatively “normal.” Real-time RT-PCR analysis demonstrated that the thyroid tumor expresses significantly higher levels of miR-378 than does the normal tissue (Fig. 3 C). These results are in agreement with previous studies indicating that numerous miRNAs are up-regulated in papillary thyroid carcinoma (26).
Repression of Suppressor of Fused (Sufu) Expression by miR-378 Transfection.
Cells stably transfected with miR-378 or vector were subjected to proteomic analysis performed by WEMB Biochem. A large number of proteins were down-regulated by miR-378 transfection. This analysis was repeated independently, using a different pool of cells. In both experiments, we detected that suppressor of fused (Sufu) was greatly down-regulated (Fig. 4 A). Sufu is reported to function as a tumor suppressor (27), because loss of its function causes excessive tumor cell proliferation (28). To test whether Sufu expression was correlated with miR-378 levels, a number of human cell lines were analyzed for Sufu expression by Western blots and miR-378 levels were analyzed by RT-PCR (SI Fig. 10A). Interestingly, in most of the cell lines tested, high levels of Sufu expression was correlated with low levels of miR-378 expression and vice versa. There were exceptions, because miR-378 can also regulate other protein expression, and Sufu may also be regulated by other miRNAs. Nevertheless, it seemed that miR-378 functioned as a suppressor of Sufu protein. To confirm this, a construct expressing an antisense miR-378 was expressed in U87 cells. Analysis of Sufu protein indicated that transient transfection of the antisense construct enhanced Sufu expression (SI Fig. 10B). Repression of Sufu expression may be an essential component of miR-378's ability to enhance cell survival. Other proteins that are the potential targets of miR-378 were also detected to be down-regulated in the miR-378-transfected cells (SI Table 2).
Targeting of Sufu by miR-378. (A) U87 cells transfected with miR-378 or a control vector were harvested and subjected to proteomic analysis. Repression of Sufu expression was found in two independent experiments. (B) Cell lysate from miR-378- or vector-transfected U87 cells was analyzed on Western blot for Sufu expression. Sufu expression was repressed by miR-378 transfection. Staining for actin confirmed equal loading. (C) Tumors formed by cells transfected with miR-378 or the control vector were sectioned and stained with H&E, secondary antibody [negative control (Neg)], and anti-Sufu antibody to examine repression of Sufu expression. (D) (Upper) Sufu 3′ UTR (nucleotides 4645–4677) was found to be the potential target of miR-378. (Lower) Mutations were generated on the potential target sequence (red). (E) COS-7 cells were cotransfected with miR-378 and a luciferase reporter construct, which was engineered with a fragment of the Sufu 3′ UTR harbouring the target sequence of miR-378 (Luc-Sufu) or a mutant (Luc-Sufu-mu). As a negative control, the luciferase reporter construct was engineered with a nonrelated fragment of cDNA (Ctrl). Asterisks indicate significant differences. **, P < 0.01. Error bars indicate SD (n = 3). (F) U87 cells stably transfected with miR-378 or GFP vector were transiently transfected with Sufu or SufuUTR. The cells were also grown on Petri dishes in serum-free conditions, followed by survival assays. Transfection of Sufu into the miR-378-expressing cells reversed the effect of miR-378 in enhancing cell survival. In the presence of the 3′ UTR, the effect of Sufu (SufuUTR construct) was repressed by miR-378 expression, resulting in high levels of cell survival. (Inset) SufuUTR expression conferred a higher level of cell survival than Sufu in GFP-transfected cells. Asterisks indicate significant differences. **, P < 0.01. Error bars indicate SD (n = 5).
Repression of endogenous Sufu expression was confirmed by Western blot probed with anti-Sufu monoclonal antibody (Novus Biologicals). We detected a clear reduction of Sufu expression in cells transfected with miR-378 compared with the control group (Fig. 4 B). RT-PCR analysis detected little difference (SI Fig. 10C), indicating that miR-378 repressed Sufu expression at the translational level. Immunocytochemical staining further confirmed the repression of Sufu expression (SI Fig. 10D).
We further examined Sufu repression in vivo. Tumors obtained from mice injected with miR-378- or GFP-transfected cells were sectioned and stained with H&E, secondary antibody (negative control), and anti-Sufu antibody. In the tumor sections derived from GFP-transfected cells, we observed cells with pink cytoplasm and condensed blue nuclei, characteristic of apoptotic cells after H&E staining (29). However, no such cells were found in the tumors from cells transfected with miR-378 (Fig. 4 C). Instead, Sufu expression was greatly repressed. This result confirmed that the physiological role of miR-378 in tumor growth and angiogenesis was mediated by the Sufu-associated pathway.
A potential target sequence of miR-378 was found in the 3′ UTR (nucleotides 4645–4676; GenBank accession no. NM_016169) of Sufu (Fig. 4 D Upper). This sequence is present in Sufu mRNAs in human, chimpanzee, mouse, rat, and canine genomes (SI Fig. 10E), suggesting conservational functions of this sequence. To confirm targeting by miR-378, we integrated a fragment of the Sufu 3′ UTR containing the target sequence, or the fragment whose target site was mutated, into a luciferase reporter vector (pMIR-Report; Ambion) (SI Fig. 11). Luciferase activity was significantly repressed in the construct harboring the miR-378 target sequence, compared with the control vector harboring a nonrelated fragment (Ctrl) or the mutated sequence (Fig. 4 E). To test the possibility of an additive effect by the miR-378 hairpin structure on the target sequence, we generated an miR-378 expression construct containing one copy of the miR-378 hairpin structure (SI Fig. 12A). Real-time PCR analysis indicated that the one-copy construct produced lower levels of mature mir-378 than the two-copy construct (SI Fig. 12B). Luciferase activity assays indicated that the two-copy miR-378 construct produced greater effect than the one-copy construct (SI Fig. 12C).
To finalize the function of miR-378-targeting Sufu in cell survival, rescue experiments were carried out. Sufu constructs with (SufuUTR) or without (Sufu) the 3′ UTR were generated (SI Fig. 13A). A Kozak sequence was engineered to the constructs to enhance protein expression (SI Fig. 13B). These constructs were transiently cotransfected with miR-378 or the control vector. Repression of Sufu expression by miR-378 was confirmed by Western blot (SI Fig. 14A). RT-PCR analysis of mRNA from these cells produced similar levels of PCR product (SI Fig. 14B), indicating that repression occurred posttranscriptionally.
U87 cells stably transfected with miR-378 or GFP vector were transiently transfected with Sufu or SufuUTR and grown on Petri dishes in serum-free conditions. The cultures were examined under a light microscope (SI Fig. 14C), and surviving cells were counted with trypan blue staining (Fig. 4 F). Cell survival assays indicated that the reintroduction of Sufu into miR-378-expressing cells reversed the effect of miR-378 on cell survival, reaching a level similar to transfection of Sufu into the GFP-expressing cells. In the presence of the 3′ UTR, the effect of Sufu (SufuUTR construct) was repressed by miR-378 expression, resulting in high levels of cell survival, confirming that the Sufu 3′ UTR is a target of miR-378. Nevertheless, without miR-378 expression (GFP controls), SufuUTR expression exhibited higher rates of cell survival compared with Sufu alone. This might be due to repression of SufuUTR by endogenous miRNAs.
Targeting of Fus-1 Expression by miR-378.
While searching for potential targets of miR-378, we also identified Tumor Suppressor Candidate 2 (TUSC2, or Fus-1; GenBank accession no. NM_007275) (30). Fus-1 appeared to harbor a standard target sequence for miR-378 at nucleotides 748–770 (Fig. 5 A). Sequence analysis indicated that this target sequence was conserved across different species (SI Fig. 15A). To test whether Fus-1 played a role in miR-378 action, we cloned Fus-1 cDNA by RT-PCR and linked it to a myc tag expressed in pcDNA3.1 (SI Fig. 15B), and then expressed the construct in GFP- or miR-378-transfected cells. Western blot analysis exhibited a band of expected size (SI Fig. 15C Upper). Cell survival experiments indicated that transfection of Fus-1 reduced miR-378-mediated enhancement in cell survival (SI Fig. 15 C Lower and D).
Targeting of Fus-1 by miR-378. (A) (Upper) Potential target of miR-378 was found in the 3′ UTR of Fus-1. (Lower) Mutations were generated on the potential target sequence (red). (B) U87 cells stably transfected with an siRNA construct against Fus-1 were grown on Petri dishes in serum-free conditions. Asterisks indicate significant differences in cell survival. **, P < 0.01. Error bars indicate SD (n = 4). (C) COS-7 cells were cotransfected with miR-378 and a luciferase reporter construct, which had been engineered with a fragment of the Fus-1 3′ UTR harbouring either the target sequence of miR-378 or mutants. Luciferase activities were determined. Asterisks indicate significant differences. **, P < 0.01. Error bars indicate SD (n = 3). (D) (Upper) Cell lysate prepared from U87 cells cotransfected with FusUTR and either miR-378 or a GFP vector was analyzed on Western blot probed with an anti-myc tag antibody or an anti-actin antibody. (Lower) Expression of Fus was also tested with RT-PCR, using primers FusKozak-BamHI and FusC-XhoI. (E) (Left) U87 cells stably expressing miR-378 were transfected with Fus or FusUTR. Survival assays indicated that transfection of Fus reversed the effect of miR-378-mediated enhancement in cell survival. (Right) GFP- and miR-378-transfected cells were transfected with FusUTR construct. Expression of FusUTR in the GFP-transfected cells greatly reduced survival rate compared with expression of FusUTR in the miR-378-transfected cells. **, P < 0.01. Error bars indicate SD (n = 5).
To corroborate this result, we generated a siRNA construct containing two hairpin structures complementary to Fus-1 sequences (SI Fig. 16A). Down-regulation of Fus-1 was confirmed by RT-PCR (SI Fig. 16B) and Western blot (SI Fig. 16C). Cell survival assays showed that transfection of Fus-1 siRNA greatly reduced cell death triggered by Fus-1 expression (Fig. 5 B), suggesting that a Fus-1-mediated pathway is essential for miR-378-enhanced cell survival.
To obtain direct evidence that Fus-1 3′ UTR is a target of miR-378, we generated a luciferase construct harbouring a fragment of the Fus-1 3′ UTR containing the target sequence of miR-378 (SI Fig. 16D). A mutant construct Luc-Fus-mu was also made (Fig. 5 A Lower). Luciferase assays showed that miR-378 reduced luciferase activity of Luc-Fus significantly, but had no effect on Luc-Fus-mu (Fig. 5 C). This implies that nucleotides 748–770 of Fus-1 are the target of miR-378. The construct expressing two copies of miR-378 exhibited a greater effect on the reduction of luciferase activities than that expressing one copy of miR-378 (SI Fig. 12D).
To confirm down-regulation of Fus-1 by miR-378, a fragment of the Fus 3′ UTR was cloned and inserted into the Fus-1 construct after the translation stop codon (SI Fig. 17A). Cotransfection of miR-378 with FusUTR produced lower levels of Fus-1 protein than cotransfection of GFP vector with FusUTR (Fig. 5 D Upper). RT-PCR analysis of mRNA from these cells produced similar levels of PCR products (Fig. 5 D Lower), indicating that miR-378 only affected Fus-1 at the protein level.
The miR-378-expressing cells were then transfected with Fus-1 or FusUTR and cultured on Petri dishes in serum-free conditions, followed by examination under a light microscope (SI Fig. 17B), and survival assays (Fig. 5 E Left). These experiments indicated that the reintroduction of Fus-1 reversed the effect of miR-378-mediated enhancement in cell survival. Re-expression of Fus-1 was sufficient to cause cell death, suggesting that the effect of miR-378 on enhanced survival was at least partly taking place through repression of Fus-1 expression. To further confirm the effect of Fus-1 3′ UTR, the GFP- and miR-378-transfected cells were transiently transfected with FusUTR, followed by microscopic examination (SI Fig. 17C) and survival assays (Fig. 5 E, Right). Although expression of FusUTR in the GFP-transfected cells greatly reduced survival rate, expression of FusUTR in the miR-378-transfected cells did not (Fig. 5 F Right), suggesting repression of FusUTR action by miR-378.
In summary, we have demonstrated that a construct harbouring a duplicate premiRNA sequence can be successfully expressed and processed to mature miRNA. With this approach, we have demonstrated that miR-378 functions as an oncogene by enhancing tumor cell survival, blood vessel expansion, and tumor growth. It acts on two tumor suppressors, Sufu and Fus-1. Other tumor suppressors may also be involved in miR-378 function, because miR-378 may target multiple mRNAs. This study sheds light on the possibility of targeting oncogenic miRNAs as a direction for gene therapy.
Methods
Construct Generation.
A miRNA construct expressing miR-378 was designed by our laboratory and the DNA synthesized by a biotech company (Top Gene Technologies). In brief, the premiRNA of miR-378 was ligated into a mammalian expression vector BluGFP that contains a Bluescript backbone, a CMV promoter driving green fluorescent protein (GFP) expression, and a promoter driving miR-378. This plasmid was developed in our lab and is expected to simultaneously express a small fragment of RNA and produce GFP (Fig. 1 A). Based on this plasmid, another plasmid named miR-378-1 was generated to express one copy of miR-378, made by using two primers, miR-378–1 copy and miR-378C. The sequence of the construct is given in SI Fig. 12. Using a similar approach, an antisense sequence to miR-378 was inserted in the expression vector producing an anti-miR-378 construct. In brief, the primer anti-miR-378 was designed to incorporate an anti-miR-378 sequence into the expression vector by PCR, followed by restriction digestion and ligation as above to produce the anti-miR-378 construct.
A luciferase reporter vector (pMir-Report; Ambion) was used to generate luciferase reporter constructs. Two primers (huSufuN3′-SacI and huSufuC3′-MluI) were used to clone a fragment of the Sufu 3′ UTR, using RT-PCR. The PCR product was digested with SacI and MluI, followed by insertion into a SacI- and MluI-open pMir-Report vector. To generate a mutant containing a mutation in the miR-378 target sequence, a primer (huSufuC3′mu-MluI) harboring this mutation was combined with the primer huSufuN3′-SacI in a PCR. The PCR product was digested with SacI and MluI and inserted into a SacI- and MluI-open pMir-Report vector (SI Fig. 13A). To serve as a negative control, a nonrelated sequence was amplified from the coding sequence of the chicken versican G3 domain, using two primers chver10051-SpeI and chver10350-SacI. The amplified PCR product was digested with SpeI and SacI, followed by insertion into an SpeI- and SacI-open pMir-Report vector.
To study the function of Sufu and Fus-1 in miR-378-regulated cell activities, we expressed the Sufu and Fus-1 constructs with or without their 3′ UTR. The Sufu construct was a kind gift of J. Rutka (University of Toronto). A Kozak sequence was engineered into the Sufu construct, using two primers, Kozak-MycHindIII and husufuCXbaI. Also, a fragment of the Sufu 3′ UTR was linked to the Sufu construct, using the primers Kozak-MycHindIII and husufuCSacI, producing the SufuUTR construct (SI Fig. 13A).
The Fus-1 coding sequence was cloned with two primers, FusKozak-BamHI and FusMyc-XhoI, using RT-PCR. The PCR product was digested with BamHI and XhoI and inserted into a BamHI- and XhoI-opened pcDNA3.1 vector (SI Fig. 15B). Also, the 3′ UTR of Fus-1 cloned above was amplified through PCR, using two primers, FusN3′-XhoI and FusC3′-ApaI. The PCR product was digested with XhoI and ApaI and inserted into the Fus-1 construct digested with XhoI and ApaI, producing the FusUTR construct (SI Fig. 17A).
The Fus-1 3′ UTR contains a potential target sequence for miR-378 located at nucleotides 748–770. A luciferase reporter construct was generated by using the primers huFusN3′-SacI and huFus780R-MluI by RT-PCR. The PCR product, after restriction digestion, was inserted into a SacI- and MluI-open pMir-Report vector, producing Luc-Fus. A mutant construct was generated by using two primers, huFusN3′-SacI and huFus780Rmu-MluI, which generated a mutation in Luc-Fus. This PCR product was used to produce the Luc-Fus-mu construct (SI Fig. 16D).
To generate a small interfering RNA against Fus-1, the primers huFus-si323P1 and huFus-si382P2 were designed for PCR. The PCR product was digested with BglII and HindIII and inserted into a BglII- and HindIII-digested BluGFP vector, resulting in a construct expressing siRNA aginst Fus-1 (SI Fig. 16A). The identities of all constructs were confirmed by restriction digestion and sequencing.
RT-PCR and RNA Analysis.
Cells (2.5×106) were harvested, and total RNA was extracted with the mirVana miRNA Isolation Kit (Ambion) according to the manufacturer's instructions. RT-PCR assays were performed as described in ref. 31. For RT-PCR of mature miRNAs, specific primers were designed, and the procedure for reactions is given in SI Fig. 6A. Real-time PCR was carried out according to the manufacturer's instructions (Qiagen; miScript reverse transcription kit, catalog no. 218060; miScript primer assay, catalog no. 218411; and miScript SYBR green PCR kit, catalog no. 218073).
Luciferase Activity Assay.
Luciferase activity assays were performed by using the Promega Luciferase Assay System as described in ref. 25.
Cell Survival Assay.
Cells (1.5 × 105 cells per well or 2 × 105 cells per well) were seeded on 35-mm Petri dishes in DMEM containing 0–10% FBS, and incubated for different time periods. The cell numbers were counted by using trypan blue staining as described in ref. 32.
Immunofluorescence.
Cells cultured on chambered slides were subjected to immunostaining as described in ref. 33.
Acknowledgments
We thank Jennifer Ma for assistance in real-time PCR experiments. This work was supported by Canadian Institutes of Health Research Grants MOP-62729 and MOP-74469 (to B.B.Y.) and Career Investigator Award CI 5958 from the Heart and Stroke Foundation of Ontario (to B.B.Y.).
Footnotes
- *To whom correspondence should be addressed. E-mail: byang{at}sri.utoronto.ca
-
Author contributions: D.Y.L. and Z.D. contributed equally to this work; D.Y.L., Z.D., and C.-H.W. performed research; B.B.Y. designed research; B.B.Y. analyzed data; and B.B.Y. wrote the paper.
-
The authors declare no conflict of interest.
-
This article is a PNAS Direct Submission. K.V.M. is a guest editor invited by the Editorial Board.
-
This article contains supporting information online at www.pnas.org/cgi/content/full/0706901104/DC1.
- © 2007 by The National Academy of Sciences of the USA