Control of organ size is a fundamental aspect in biology and plays important roles in development. The Hippo pathway is a conserved signaling cascade that controls tissue and organ size through the regulation of cell proliferation and apoptosis. Here, we report on the roles of Hcf (host cell factor), the Drosophila homolog of Host cell factor 1, in regulating the Hippo signaling pathway. Loss-of-Hcf function causes tissue undergrowth and the down-regulation of Hippo target gene expression. Genetic analysis reveals that Hcf is required for Hippo pathway-mediated overgrowth. Mechanistically, we show that Hcf associates with the histone H3 lysine-4 methyltransferase Trithorax-related (Trr) to maintain H3K4 mono- and trimethylation. Thus, we conclude that Hcf positively regulates Hippo pathway activity through forming a complex with Trr and controlling H3K4 methylation.
Initially discovered in Drosophila by genetic studies, the Hippo pathway is a highly conserved regulator of tissue growth and organ size in mammals [1–5]. The core of the Hippo pathway in Drosophila is a network of kinases which controls the subcellular localization and activities of the downstream transcriptional co-activator Yki (Yorkie) [6,7]. In this kinase cascade, the Ste20-like kinase Hippo (Hpo) phosphorylates the downstream Dbf2-related kinase Warts (Wts), which then phosphorylates Yki to prevent its nuclear localization [6,7]. Nuclear Yki binds to other transcription factors, including Scalloped (Sd), Homothorax (Hth), Teashirt (Tsh) and Mad, to activate Hippo target gene expression [8–11]. Recent studies have found that Yki recruits many chromatin modifiers, such as GAGA factor, the Brahma complex, the mediator complex and the histone H3 lysine 4 (H3K4) methyltransferase complex, to promote transcription [12–16]. The association between Yki and the histone methyltransferase complex is crucial for the chromatin modification and transcriptional activation of Yki target genes [14,15].
Hcf (host cell factor) is the Drosophila homolog of human HCF-1, one cofactor mediated the interaction between chromatin modifiers and DNA-bound transcriptional regulators [17,18]. Through recruiting specific chromatin-modifying complexes, HCF-1 can function in both activation and repression of gene expression [17–19]. Both Hcf and HCF-1 share similar conserved sequence structure and undergo a process of proteolytic cleavage to produce a heterodimeric complex of Hcf-N and Hcf-C subunits [20–22]. Biochemical studies have shown that Drosophila Hcf is present in both dSet1 and Trr (trithorax-related) H3K4 methlyltransferase complexes, indicating its potential roles in histone H3K4 methylation [23,24]. In Drosophila, mutation of Hcf causes a pleiotropic phenotype, including female sterility, size reduction and abnormal morphology . In the present study, we identify Drosophila Hcf as a positive regulator of the Hippo pathway. We further show that Hcf forms a complex with Trr in S2 cells. Last, we report the evidence that Hcf1 functions to activate Hippo target gene expression by promoting histone methylation, especially H3K4me1 (H3K4 monomethylation) and H3K4me3 (H3K4 trimethylation). In summary, Hcf promotes tissue growth through positively regulating the Hippo pathway.
Materials and methods
The fly stocks were maintained at 25°C on standard food. We used the following fly stocks: UAS-hcf RNAi 1 (VDRC, V46999), UAS-hcf RNAi 2 (Tsinghua Fly Center THU0900), UAS-hcf RNAi 3 (V105433), UAS-wts RNAi 1 (V106174), UAS-wts RNAi 2 (THU0564), en-GAL4 UAS-GFP, Nub-GAL4, en-GAL4 UAS-GFP/Cyo; Diap1-lacZ/TM6B, ex-lacZ/Cyo; hh-GAL4 UAS-GFP/TM6B, wts RNAi/Cyo; UAS-hcf RNAi 1/TM6B.
Immunostaining and microscopy
Wing imaginal discs of third instar larvae were dissected in cold PBS and then fixed in 1× PBS with 4% formaldehyde for 15–20 min with gentle rotation. After washing for three times in PBT (0.1% Triton X-100 in PBS), discs were blocked in 3% BSA in PBT for 1 h before incubated with primary antibodies overnight at 4°C. Discs were then washed for four times and incubated with secondary antibodies for 2 h and DAPI for 20 min. After four times further wash, discs were mounted in Vectorshield medium. The following primary antibodies were used: rabbit anti-GFP (1 : 2000, Abcam, ab290), mouse anti-GFP (1 : 2000, Abcam, ab1218), rabbit anti-H3K4 monomethylation (1 : 2000, Abcam, ab8895), rabbit anti-H3K4 dimethylation (1 : 2000, Abcam, ab7766), mouse anti-β-galactosidase (1 : 2000, Promega, Z378A), rabbit anti-H3K4 trimethylation (1 : 2000, Activemotif, 39159). Fluorescent secondary antibodies were Alexa Fluor 488, 555 or 633-conjugated and used at 1 : 500 dilution. Images were acquired on an Olympus FV1000 confocal microscope and further processed by Adobe Photoshop CS4.
The full-length cDNA fragment of Hcf was amplified by PCR using the forward primer 5′-CAGAGACCCCGGATCGGGGTACCATGGAAGGCTCAGACTTT-3′ and the reverse primer 5′-TCAGCTTCTGTTCCATGCGGCCGCgATCATGCAATCCGTTGCG-3′, and then cloned into a pAC5.1-myc vector to give rise to the pAC5.1-Hcf-myc construct with a C-terminal myc tag. The N-terminal of Hcf was amplified by PCR using the forward primer 5′-CTACTAGTCCAGTGTGGTGATGGAAGGCTCAGACTTTGT-3′ and the reverse primer 5′-TGATCAGCTTCTGTTCCATCTCTATAATATCATCCATTG-3′, and then cloned into a pAC5.1-myc vector to give rise to the pAC5.1-Hcf-N-myc construct. The C-terminal of Hcf was amplified by PCR using the forward primer 5′-CTACTAGTCCAGTGTGGTGATGCAGTTGGATGGAGCCGG-3′ and the reverse primer 5′-TGATCAGCTTCTGTTCCATATCATGCAATCCGTTGCGTC-3′ and then cloned into a pAC5.1-myc vector to give rise to the pAC5.1-Hcf-C-myc construct. The full-length cDNA fragment of Trr was amplified by PCR using the forward primer 5′-GACCCCGGATCGGGGTACCTATGAATATACCGAAGGTGACAAC-3′ and the reverse primer 5′-GGATATCTGCAGAATTCCAGTTCATCCACTTGCGACAGTTG-3′ and then cloned into a pAC5.1-flag vector to give rise to the pAC5.1-Trr-flag construct. Plasmids for expression of glutathione S-transferase (GST) fusion proteins in Escherichia coli were generated by inserting PCR fragments into pGEX-4T-1. The SET domain-containing Trr-C421 (residues 2011–2431) fragment was amplified by PCR using the forward primer 5′-CCGGAATTCATGGGTGCAGGAGCAG-3′ and the reverse primer 5′-TAAAGCGGCCGCGTTCATCCACTTGCGACA-3′ and then cloned into a pGEX-4T-1 vector to give rise to the pGEX-Trr-C421 construct. The full-length cDNA fragment of Hcf was amplified by PCR using the forward primer 5′-CGCGTGGATCCCCGGAGTTGGATGGAGCCGGAGAT-3′ and the reverse primer 5′-AGTCACGATGCGGCCATCATGCAATCCGTTGCGTC-3′, and then cloned into a pGEX-4T-1 vector to give rise to the pGEX-Hcf construct.
S2 cell culture and transfection
Drosophila S2 cells were cultured at 25°C in Schneider's medium (Gibco, 21720-001) supplemented with 10% fetal bovine serum (FBS, Gibco, 10099-141). Plasmids were transfected using the X-tremeGene HP (Roche, Cat. No. 06366236001) reagent.
Co-immunoprecipitation and western blotting
S2 cells were lysed in modified RIPA lysis buffer [50 mM Tris–HCl, pH 8.0, 150 mM NaCl, 1% (v/v) IGRPA CA-630, 0.5% (w/v) sodium deoxycholate] supplemented with protease inhibitor cocktail (Roche, Cat. No. 04693132001) and PhosStop phosphatase inhibitor cocktail (Roche, Cat. No. 4906845001). Myc-tagged and Flag-tagged proteins were pre-cleared using Protein-A agarose beads (Roche, Cat. No. 11134515001) for 1.5 h. Next, Myc-tagged proteins were purified using ∼5 µg mouse anti-myc antibody (ab9132, Abcam) and Flag-tagged proteins were purified using ∼5 µg mouse anti-Flag antibody (CW0287A, Beijing ComWin Biotech) using Protein-G agarose beads (Roche, Cat. No. 11243233001). After 4 h of incubation, the myc or Flag immunoprecipitates were washed for five times. When cross-linking was required, the harvest cells were incubated in the solution (1 mM DTSSP, 0.1 M HEPES, 1× PBS) for 30 min. About 1 M Tris (pH 7.5) was used to stop the cross-linking reaction. Samples were then subjected to SDS–PAGE and transferred to the polyvinylidene fluoride membrane. Membranes were immunoblotted with mouse anti-myc antibody (ab9132, Abcam) and mouse anti-Flag (1 : 1000, CW0287A, Beijing ComWin Biotech) antibodies. Detection of purified and associated proteins was performed using the ChemiLucent™ ECL detection reagents (Millipore, WBKLS0500). Images were taken using the Chemiluminescence Imaging System (Clinx Science Instruments, Shanghai).
Total RNA was purified from 50 third instar larval imaginal discs by the use of Trizol reagent (Ambion, 15596-026), and 1 µg of total RNA were reversed transcribed with PrimeScript RTase (TaKaRa, PrimeScript™ II 1st Strand cDNA Synthesis Kit, Code No. 621A). Quantitative PCR was performed using the ABI 7900HT Fast Real-Time PCR system. Primers used were as follows: rp49: 5′-GCTAAGCTGTCGCACAAA-3′ and 5′-TCCGGTGGGCAGCATGTG-3′; Hcf: 5′-ATACATACGTTCGGACGGGC-3′ and 5′-TGTGGAAGGAAGGCACACAA-3′.
In vitro histone methyltransferase assay
Histone methyltransferase assays were performed as previously described . Briefly, GST-Trr-C421 (6 µl) was mixed with 3 µg of recombinant histone H3 protein (Active motif, cat#31295) and 0.5 mM S-adenosyl methione (SAM) in a total volume of 30 µl of methylase activity buffer (0.05 M Tris, pH 8.2, 0.2 M NaCl, 3 mM DTT, 5 mM MgCl2, 5% glycerol). Reactions were incubated for 1 h at 30°C, and stopped by adding 2× SDS loading buffer. Proteins were separated by 12% SDS–PAGE and analyzed by western blot. Rabbit anti-H3K4 mono-methylation (1 : 500, Abcam, ab8895) and mouse anti-H3 (1 : 1000, Active motif, 61476) antibodies were used. GST-Trr-C421 and GST-Hcf proteins were expressed in BL-21 cells and induced by IPTG. GST-Trr-C421 proteins were purified on Glutathione Sepharose 4B (GE Healthcare, 17-0756-01). GST-Hcf proteins were insoluble and extracted with urea (3 M).
Adult wing analysis
For adult wing analysis, wings from female flies were removed and mounted 80% glycerol. Wings were imaged on a Nikon Eclipse 80i microscope. Quantification of the wing size was done in ImageJ.
Results and discussion
Knockdown of Hcf leads to tissue undergrowth and the down-regulation of Hippo target gene expression
Hcf is located on chromosome 4 in the Drosophila genome (www.flybase.org). A previous study has shown that loss of Hcf causes 50% lethality during pupal stage and some of the survivors show reduced body size . To further investigate the role of Hcf in growth control, we first used the posterior expressed en-Gal4 (engrailed-Gal4) to knock down Hcf activity in the developing wing. Consistent with the loss-of-function phenotype, RNAi of Hcf by en-Gal4 caused a reduction in the posterior size of the adult wing (Figure 1A,B, quantified in E). In addition, we also observed an ectopic vein nearby the posterior crossvein in the Hcf knockdown wing (Figure 1B). To avoid the off-target effect of RNAi, we next knocked down Hcf activity using two additional Hcf RNAi lines, in which the hairpin sequences target different regions of the Hcf mRNA. Tissue size reduction was also observed when these two RNAi lines were used (Figure 1A,C,D, quantified in E). Moreover, we used another wing specific Gal4, nub-Gal4 (nubbin-Gal4), to knock down Hcf activity and confirmed that reduction in Hcf caused tissue size reduction in the Drosophila wing (Supplementary Figure S1A–D, quantified in E). qRT-PCR analysis revealed that these three RNAi lines showed 60–70% knockdown efficiency, and they lowered endogenous Hcf in wing imaginal discs to 30%–40% when driven by nub-Gal4 (Supplementary Figure S1F).
Knockdown of Hcf causes tissue undergrowth.
To determine the potential relationship between Hcf and the Hippo pathway, we sought to examine the expression of Hippo target genes upon Hcf knockdown in the wing imaginal disc. For this purpose, two well-characterized Hippo target gene reporters, DIAP1 (Death-associated inhibitor of apoptosis 1)-lacZ and ex-lacZ, were used. For all three RNAi lines, knockdown of Hcf by en-Gal4 led to a decrease in Diap1-lacZ expression in the posterior compartment compared with that in the anterior compartment (Figure 2A–D′). Consistently, ex-lacZ was also down-regulated in the posterior compartment of wing discs when Hcf was knocked down with hedgehog-Gal4 (hh-Gal4) (Figure 2E–H′).
Knockdown of Hcf decreases transcription of Hippo pathway target genes.
Taken together, these results imply that Hcf promotes tissue growth and is required for Hippo downstream target gene expression.
Hcf is required for Hippo pathway inactivation-mediated overgrowth
To explore the role of Hcf in the Hippo pathway, we performed genetic epistasis experiments. For this purpose, we examined whether Hcf RNAi was able to suppress the tissue overgrowth caused by RNAi-mediated knockdown of the core Hippo kinase cascade component Wts. RNAi knockdown of wts with en-Gal4 led to an increase in wing size in the posterior compartment (Figure 3A,C, quantified in E). However, knockdown of Hcf by RNAi suppressed the wts-mediated wing overgrown phenotype (Figure 3A–D, quantified in E). This result provides genetic evidence that Hcf is involved in the regulation of Hippo pathway activity.
Hcf is required for Hippo pathway inactivation-induced tissue overgrowth.
Hcf associates with Trr
Given the fact that Hippo target gene expression is down-regulated in the absence of Hcf, we hypothesized that Drosophila Hcf functions to active Hippo target gene expression directly. Previous studies have shown that HCF-1 can interact with chromatin modifiers as well as transcription factors to regulate gene expression [18,19]. Two recent reports reveal that the histone methytransferase complex Trr controls Hippo target gene expression through histone methylation [14,15]. To gain the molecular mechanisms by which Hcf actives Hippo target gene expression, we investigated whether Hcf associates with Trr to control histone methylation. To this end, we performed co-immunoprecipitation (Co-IP) experiments for assaying the interaction between these two proteins. Myc-tagged Hcf and Flag-tagged Trr constructs were cotransfected in S2 cells. Following the immunoprecipitation of Myc-Hcf, Trr-Flag was detected in the immune complexes (Figure 4A). In a converse Co-IP experiment, Myc-Hcf was recovered in the immune complexes prepared by the immunoprecipitation of Trr-Flag (Figure 4B). As Hcf undergoes proteolytic maturation in which two subunits Hcf-N and Hcf-C are produced by processing of the full-length precursor, two bands represented the full-length Hcf and C-terminal Hcf subunit were observed (Figure 4A,B). We further determined which subunit of Hcf interacts with Trr. Hcf-N-Myc or Hcf-C-Myc was co-expressed with Trr-Flag and immunoprecipitated. Trr-Flag was not recovered in Hcf-N-Myc immunoprecipitation, and Hcf-N-Myc failed to co-precipitate with Trr-Flag in the reciprocal experiment (Figure 4C,D). However, we found that Trr-Flag was co-precipitated with Hcf-C-Myc and Hcf-C-Myc was recovered in Trr-Flag immunoprecipitation (Figure 4E,F). Therefore, we concluded that Hcf interacts with Trr and this interaction is mediated by the C-terminal region of Hcf.
Hcf associates with Trr in S2 cells.
Knockdown of Hcf impairs H3K4 mono- and trimethylation in wing discs
Trr is required for H3K4 methylation at Hippo target genes, especially in induction of both H3K4me1 and H3K4me3 [14,15]. The association between Hcf and Trr drove us to assess the effect of Hcf knockdown on histone H3K4 methylation in wing discs. We stained the wing imaginal discs with antibodies specific to H3K4 mono-, di- and trimethylation, and observed that Hcf knockdown led to significant reduction in H3K4me1 and H3K4me3 signals in the posterior compartment as compared with that in the anterior compartment (Figure 5A–H′). However, Hcf knockdown had no effects on the levels of H3K4me2 (H3K4 dimethylation) (Supplementary Figure S2A–D′). Together, these results suggest that the interaction between Hcf and Trr contributes to the induction of H3K4me1 and H3K4me3.
Impairment of H3K4 mono- and trimethylation in Hcf knockdown tissues.
Trr histone methyltransferase activity on histone H3 is enhanced by Hcf
To test whether the association between Hcf and Trr influences Trr histone methyltransferase (HMT) activity, we performed in vitro HMT assay. A previous study has shown that a purified C-terminal Trr (Trr-C421, C-terminal 421 residues) fragment possesses a robust H3K4 monomethyltransferase activity . We produced a similar Drosophila Trr fragment (GST-Trr-C421) and confirmed that it has a H3K4 monomethyltransferase activity on a free histone H3 substrate in vitro (lanes 1 and 2 in Figure 6). The addition of Hcf to a GST-Trr-C421 HMT assay increased monomethylation of H3K4 by Trr, suggesting that Hcf directly modulates H3K4 monomethylation by Trr (lanes 3 and 4 in Figure 6).
Enhancement of Trr HMT activity by Hcf.
Here, we focus on the role of Hcf in growth control during Drosophila wing development. Our results reveal that Hcf functions as a positive regulator of the Hippo signaling pathway in Drosophila. Furthermore, Hcf associates with Trr and promotes histone methylation, which might be crucial for Hippo target gene transcription. Thus, it is likely that Hcf promotes tissue growth partly through the regulation of the Hippo pathway in Drosophila.
The transcriptional co-activator Yki is the key component of the Hippo pathway, and it activates Hippo target gene expression through recruiting the histone methyltransferase complex [2,14,15]. This activation is dependent on the binding between Yki and NcoA6 (Nuclear receptor co-activator 6), one subunit of the Trr methyltransferase complex [14,15]. It has been suggested that the interaction between Yki and the Trr complex functions to promote histone methylation at Hippo target genes [14,15]. Hcf associates with Trr and regulates histone methylation, indicating that Hcf could act as an additional mediator between the Yki and Trr methyltransferase complex. Further studies are required to examine the possible interaction between Yki and Hcf.
Together, these findings demonstrate that chromatin modification plays crucial roles in tissue growth control.
Death-associated inhibitor of apoptosis 1
histone H3 Lysine 4
host cell factor
Z.N. performed most of the experiments with the help from J.L., F.W. and Q.D. W.Y. performed qRT-PCR and HMT activity experiments. Y.X. and X.Y. provided supervision and project direction. W.G. conceived the project and supervised students. W.G. and Z.N. wrote the manuscript.
The present study was supported by the National Key Research and Development Program of China [2018YFC1003200] and a grant  from the National Natural Science Foundation of China.
We thank the Vienna Drosophila Resource Center, the Bloomington Drosophila stock center and the Tsinghua Fly Center for fly stocks.
The Authors declare that there are no competing interests associated with the manuscript.