Phenotypic Mutation 'feckless' (pdf version)
Mutation Type missense
Coordinate82,547,895 bp (GRCm39)
Base Change T ⇒ C (forward strand)
Gene Hcfc2
Gene Name host cell factor C2
Synonym(s) 1700129L13Rik, fkls
Chromosomal Location 82,531,994-82,578,262 bp (+) (GRCm39)
MGI Phenotype FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] This gene encodes one of two proteins which interact with VP16, a herpes simplex virus protein that initiates virus infection. Both the encoded protein and the original Herpes host cell factor interact with VP16 through a beta-propeller domain. The original Herpes host cell factor, however, is effective at initiating viral infection while the encoded protein is not. Transcripts of varying length due to alternative polyadenylation signals have been described. [provided by RefSeq, Jul 2008]
PHENOTYPE: Mice homozygous for null or severely hypomorphic allele exhibit reduced poly(I:C)-mediated TLR3 signaling and increased mortality following viral infection. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_001081218; MGI: 1915183

Limits of the Critical Region 77790618 - 79893800 bp
Amino Acid Change Tryptophan changed to Arginine
Institutional SourceBeutler Lab
Gene Model not available
AlphaFold Q9D968
SMART Domains Protein: ENSMUSP00000020478
Gene: ENSMUSG00000020246
AA Change: W296R

Pfam:Kelch_1 22 60 2.1e-6 PFAM
Pfam:Kelch_5 68 106 1.1e-6 PFAM
Pfam:Kelch_3 81 135 8.8e-7 PFAM
Pfam:Kelch_5 186 230 8.4e-7 PFAM
Pfam:Kelch_3 206 253 1.6e-11 PFAM
Pfam:Kelch_1 244 302 7.5e-9 PFAM
Pfam:Kelch_3 254 323 3.4e-7 PFAM
Pfam:Kelch_5 312 356 1.4e-6 PFAM
FN3 357 591 8.43e-9 SMART
FN3 607 703 6.06e-1 SMART
Predicted Effect probably damaging

PolyPhen 2 Score 1.000 (Sensitivity: 0.00; Specificity: 1.00)
(Using ENSMUST00000020478)
SMART Domains Protein: ENSMUSP00000124472
Gene: ENSMUSG00000020246

Pfam:Kelch_1 1 38 8.5e-7 PFAM
Pfam:Kelch_5 46 84 3.7e-8 PFAM
Pfam:Kelch_3 59 113 2.6e-8 PFAM
Predicted Effect probably benign
Meta Mutation Damage Score 0.9660 question?
Is this an essential gene? Probably nonessential (E-score: 0.194) question?
Phenotypic Category Autosomal Semidominant
Candidate Explorer Status loading ...
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI


Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00847:Hcfc2 APN 10 82577112 splice site probably null
IGL01799:Hcfc2 APN 10 82536825 missense probably damaging 1.00
IGL01916:Hcfc2 APN 10 82570217 missense possibly damaging 0.94
IGL02150:Hcfc2 APN 10 82545852 missense probably damaging 1.00
IGL02378:Hcfc2 APN 10 82544905 missense possibly damaging 0.64
IGL02580:Hcfc2 APN 10 82564256 missense probably benign 0.00
IGL02641:Hcfc2 APN 10 82538383 missense probably damaging 1.00
Backstabbing UTSW 10 82547659 splice site probably null
Minions UTSW 10 82575079 missense probably damaging 1.00
scaffold UTSW 10 82574242 missense probably damaging 1.00
R0380:Hcfc2 UTSW 10 82564272 splice site probably benign
R0528:Hcfc2 UTSW 10 82575079 missense probably damaging 1.00
R0534:Hcfc2 UTSW 10 82574242 missense probably damaging 1.00
R1646:Hcfc2 UTSW 10 82536861 missense probably damaging 1.00
R1903:Hcfc2 UTSW 10 82538392 missense probably damaging 0.98
R1939:Hcfc2 UTSW 10 82538284 missense probably damaging 0.99
R2014:Hcfc2 UTSW 10 82574814 missense probably benign 0.23
R2015:Hcfc2 UTSW 10 82574814 missense probably benign 0.23
R2571:Hcfc2 UTSW 10 82544857 missense probably damaging 1.00
R4540:Hcfc2 UTSW 10 82568481 missense probably benign 0.10
R4694:Hcfc2 UTSW 10 82559534 missense probably damaging 1.00
R4735:Hcfc2 UTSW 10 82547914 missense probably damaging 1.00
R4833:Hcfc2 UTSW 10 82544980 missense probably null 0.01
R6837:Hcfc2 UTSW 10 82575030 missense probably damaging 0.96
R7268:Hcfc2 UTSW 10 82544846 nonsense probably null
R7683:Hcfc2 UTSW 10 82535063 missense probably benign 0.00
R7733:Hcfc2 UTSW 10 82575013 missense probably benign 0.00
R7742:Hcfc2 UTSW 10 82547659 splice site probably null
R8319:Hcfc2 UTSW 10 82574201 missense probably damaging 0.98
R8829:Hcfc2 UTSW 10 82574179 missense probably damaging 1.00
R8989:Hcfc2 UTSW 10 82536822 missense probably damaging 1.00
R9189:Hcfc2 UTSW 10 82535041 missense probably benign 0.06
R9241:Hcfc2 UTSW 10 82568485 missense probably benign
R9362:Hcfc2 UTSW 10 82574258 missense probably damaging 1.00
R9363:Hcfc2 UTSW 10 82574258 missense probably damaging 1.00
R9386:Hcfc2 UTSW 10 82574937 missense probably damaging 1.00
R9701:Hcfc2 UTSW 10 82574269 nonsense probably null
R9802:Hcfc2 UTSW 10 82574269 nonsense probably null
V3553:Hcfc2 UTSW 10 82547895 missense probably damaging 1.00
X0022:Hcfc2 UTSW 10 82545801 missense probably damaging 0.99
Z1176:Hcfc2 UTSW 10 82535006 missense probably damaging 0.97
Mode of Inheritance Autosomal Semidominant
Local Stock Live Mice, Sperm, gDNA


Last Updated 2018-03-06 10:16 PM by Diantha La Vine
Record Created unknown
Record Posted 2017-10-04
Other Mutations in This Stock Stock #: F2404 Run Code: SLD00106
Coding Region Coverage: 1x: 87.2% 3x: 68.2%
Validation Efficiency: 81/91

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Abhd18 T to G 3: 40,888,313 (GRCm39) C386G probably damaging Het
AC091683.1 G to A 13: 50,054,528 (GRCm39) noncoding transcript Het
Adgrv1 G to A 13: 81,568,125 (GRCm39) T5016M probably benign Het
Epha3 A to G 16: 63,366,531 (GRCm39) L972P probably benign Het
Gpm6a G to A 8: 55,511,917 (GRCm39) D264N probably damaging Het
H1f7 T to A 15: 98,155,090 (GRCm39) I20F possibly damaging Het
Pvr A to T 7: 19,639,102 (GRCm39) S393T probably benign Het
Rsbn1 T to A 3: 103,821,892 (GRCm39) C3* probably null Het
Thoc2l T to A 5: 104,668,096 (GRCm39) W873R possibly damaging Homo

Stock #: V3553 Run Code: HSQ01008
Coding Region Coverage: 10x: 96.7% 20x: 94.0%
Validation Efficiency: 0/0

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Bcas3 G to A 11: 85,712,926 (GRCm39) probably benign Het
Bicral G to T 17: 47,141,021 (GRCm39) N20K probably damaging Het
Dnah9 C to A 11: 65,860,902 (GRCm39) G2823C probably damaging Het
Hcfc2 T to C 10: 82,547,895 (GRCm39) W296R probably damaging Het
Lrp1 C to T 10: 127,407,311 (GRCm39) G1877S probably damaging Het
Rhbdf1 G to A 11: 32,161,583 (GRCm39) P575L probably damaging Het
Sh3pxd2a G to A 19: 47,255,658 (GRCm39) P1048L probably benign Het
Tgm5 C to T 2: 120,901,983 (GRCm39) V229I probably damaging Het
Tmem167 A to C 13: 90,246,585 (GRCm39) K36N probably damaging Het
Tnfrsf21 G to A 17: 43,348,822 (GRCm39) A145T probably benign Het
Zfp84 G to T 7: 29,476,672 (GRCm39) A455S probably benign Het

Stock #: Z2264 Run Code: HSQ01073
Coding Region Coverage: 10x: 0.0% 20x: 0.0%
Validation Efficiency: 0/0

GeneSubstitutionChr/LocMutationPredicted EffectZygosity

Stock #: Z2265 Run Code: HSQ01073
Coding Region Coverage: 10x: 0.0% 20x: 0.0%
Validation Efficiency: 0/0

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Phenotypic Description
Figure 1. TNF production by wild-type, Tlr3-/-, and Feckless mutant macrophages in response to polyI:C. Figure adapted from (1).
Figure 2. Immunoblot analysis of phosphoyrlated (p) ERK, p-JNK, p-p38, p-Akt, p-TBK1, p-STAT1, and the degradation of IκB at the indicated times after poly(I:C) treatment of peritoneal macrophages from Feckless or wild-type mice. Figure adapted from (1).
Figure 3. IFN-β in the culture medium of peritoneal macrophages from Feckless heterozygous, Feckless homozygous, or WT mice 4 hours after poly(I:C) treatment at the indicated concentrations. Tlr3-/- macrophages served as negative controls. Figure adapted from (1).
Figure 4. Immunoblot analysis of full-lenth and cleaved TLR3 in macrophages from WT, homozygous Feckless, and heterozygous Feckless mice 24 hours ater poly(I:C) treatmebn with or without DOTAP. Figure adapted from (1).
Figure 5. qRT-PCR measurement of Tlr3 mRNA normalized to Gapdh in macrophages from heterozygous Feckless, homozygous Feckless, and WT littermates. Expression level was plotted relative to that in WT cells. Figure adapted from (1).

The Feckless phenotype was identified among ENU-induced mutants in a screen for impaired ability to respond to TLR ligands. Peritoneal macrophages from Feckless mice failed to produce TNF and IL-12 in response to the dsRNA analog, poly I:C [Figure 1; (1)]. Type I interferon production was normal in response to poly I:C in Feckless macrophages, suggesting that the activation of NF-κB is impaired at a level downstream from TLR3. Signaling through other TLRs was intact, indicating that the Feckless mutation specifically affects the TLR3 signaling pathway.

Peritoneal macrophages from homozygous Feckless mice exhibited reduced phosphorylation of p38, JNK, and ERK as well as impaired IκB degradation in response to poly(I:C) stimulation (Figure 2), indicating that both MAPK and NF-κB signaling is aberrant. (1). Poly(I:C)-induced IFN-β production (Figure 3) as well as TBK1 phosphorylation (Figure 2) were impaired. Both TLR3 protein (Figure 4) and transcript levels (Figure 5) were reduced in macrophages from homozygous Feckless compared to that in wild-type macrophages. Taken together, the aberrant TLR3-associated signaling phenotype observed in the Feckless mice was due to loss of TLR3 expression.

Nature of Mutation
The Feckless mutation was mapped on approximately 2,500 meioses to Chromosome 10 based on the phenotype of reduced TNF production by peritoneal macrophages in response to poly I:C. Sequencing coverage (capillary electrophoresis) of 99.1% of coding/splicing sequences within the 2.1 Mb critical region, delimited by markers d10mit175 and SNP1 (77790618 and 79893800 bp from the centromere) was achieved, but no coding change was identified. A single base change in an intergenic region (A to C at position 78206261) was detected incidentally in the course of these efforts but its existence within an expressed gene product could not be established. This mutation was nonetheless used as a marker and a homozygote was used in whole genome sequencing using the SOLiD technology.  No mutations were identified in coding/splicing sequences within the critical region (either among successful nucleotide calls covered 3 or more times, or nucleotides with the “N” pattern denoting single or double coverage), although 5 single nucleotide changes were detected within intronic and intergenic regions (including the A to C substitution at 78206261). 88.7%, 79.9%, and 68.9% of coding/splicing sequences in the critical region were covered at least 1X, 2X or 3X by SOLiD sequencing, respectively. Validation of “N” pattern nucleotides (those flanked with ≥15 nt good sequence) was then expanded 8.1 Mb upstream and 12 Mb downstream of the defined critical region and revealed a single T to C transition at position 82,712,061 of chromosome 10, 2.3 Mb outside the critical region. Whole exome sequencing of DNA from an affected mouse confirmed the location of the Hcfc2 mutation (1).
The mutation occurs at position 1023 of the Hcfc2 mRNA, in exon 7 of 15 total exons. Four Hcfc2 transcripts are annotated in the Vega database.
The mutated nucleotide is shown in red text within the Hcfc2 mRNA sequence, and results in a tryptophan to arginine substitution at amino acid 296 of the host cell factor C2 (Hcfc2) protein (hereafter HCF-2).
Figure 6. TNF production by wild-type, Feckless, and Hcfc2TALEN/TALEN mutant macrophages in response to polyI:C.

An Hcfc2 knockout strain was generated by TALEN. Hcfc2TALEN/TALEN mice showed reduced TNF production upon poly(I:C) stimulation (Figure 6), which confirmed Hcfc2 is causative gene of Feckless phenotype.

Illustration of Mutations in
Gene & Protein
Protein Prediction
Figure 7. Domain structures of HCF-1, HCF-2, and VP16.  The human HCF proteins are depicted.  The β-propeller domain contains six kelch-like repeats predicted to fold into a six-bladed β-propeller structure.  A proline residue exists at the fourth position of every kelch repeat in each of HCF-1 and HCF-2; mutation of the proline residue in the third kelch repeat of HCF-1 (P) abrogates binding to VP16.  An enlarged schematic of the fifth kelch repeat is shown for HCF-2, with the position of the feckless mutation indicated by a red asterisk.  The eight HCFPRO repeats of HCF-1 are indicated by open (nonfunctional) and filled (functional) arrowheads.  The SASN, basic, and SASC regions are labeled.  The first 902 amino acids of HCF-1 comprising the β-propeller, SASN domain, and basic region are required to complement the tsBN67 defect.  The β-propeller domain of HCF-1 is necessary and sufficient to bind the HCF-binding motifs of VP16 and Luman.  VP16 contains a 355 amino acid central conserved core and a C-terminal transactivation domain (TAD).  Residues binding to HCF-1 and Oct-1 are indicated.  Image is interactive; click to see additional lab alleles of HCF-2 and to connect to the respective pages.

HCF-2 is a protein with unknown function identified based on amino acid sequence similarity to host cell factor C1 (Hcfc1, hereafter HCF-1) (2). The regions of homology between HCF-1 and HCF-2 are restricted to the N- and C-termini of the proteins. In HCF-2, these regions are separated by 201 amino acids without similarity to HCF-1 or other proteins. The conserved regions correspond to a β-propeller domain at the N-terminus (364 aa), and two self-association sequence (SAS) elements (43 and 192 aa for the N- and C-terminal self-association domains, respectively) (Figure 7). Between HCF-1 and HCF-2, the β-propeller, N-terminal self-association domain (SASN), and C-terminal self-association domain (SASC) share 69%, 54%, and 59% identity, respectively. However, HCF-2 lacks the large basic region in the middle of the HCF-1 sequence, as well as eight 26-amino acid HCFPRO repeats that serve as autoproteolytic cleavage sites for HCF-1 (3). Cleavage of the sequence PPCE↓THET within one or more HCFPRO repeats generates multiple N- and C-terminal fragments that remain non-covalently associated (4-6). HCF-2 is not proteolytically processed.  Human and mouse HCF-2 consist of 792 and 722 amino acids, respectively, and are 82.2% identical.

The β-propeller domain of HCF-2 is highly similar to that of HCF-1, and contains six kelch-like repeats (HCFKEL1-HCFKEL6) predicted to fold into a six-bladed β-propeller structure (8).  Each kelch repeat consists of four β-strands (β1-β4) connected by loops of variable length to form a β-sheet or blade. The β-propeller domain of HCF-1 (also known as the HCFVIC domain) is necessary and sufficient to bind VP16, a herpes simplex virus transcription factor, and to stabilize a DNA-bound complex containing HCF-1, VP16, and the cellular POU domain protein Oct-1 (see Background) (8-10). VP16 contains a C-terminal transactivation domain and a central conserved core necessary for DNA binding whose structure bears no similarity to that of other known DNA-binding proteins (Figure 7) (11;12). It binds to HCF-1 using a short sequence (D/EHXY) designated the HCF-binding motif (HBM) that is also used by other proteins to bind HCF-1 (13). When expressed with Oct-1, the HCFVIC domain is sufficient to support VP16 transcriptional activation (7). An apparently critical residue of the HCF-1 and HCF-2 kelch repeat is a proline at the fourth position of the repeat, which is conserved in every kelch repeat of both proteins. A point mutation of the invariant proline residue in the third kelch repeat of HCF-1 abrogates binding to VP16 (14).

Full length HCF-2 (15), or the β-propeller domain alone (2), is also able to bind to VP16 and to stabilize a complex with DNA, VP16 and Oct-1 in vitro. The β-propeller domain does this with greatly reduced efficiency compared to HCF-1 or full length HCF-2. No interaction between the HCF-2 β-propeller domain and VP16 was detected in a yeast two-hybrid assay. More importantly, HCF-2 failed to support transcriptional activation by the VP16-induced complex (15). However, when the fifth and sixth kelch repeats of HCF-2 were substituted with those of HCF-1, the chimeric HCF-2 β-propeller domain was able to mediate complex formation in vitro with VP16, Oct-1, and DNA as well as HCF-1. Thus, VP16 discriminates between HCF-1 and HCF-2 based on determinants in the fifth and sixth kelch repeats. The bZIP protein LZIP (also known as Luman) that associates with HCF-1 in a manner analogous to VP16 (13;16) also selectively interacts with HCF-1 based on differences in the fifth and sixth kelch repeats (2)

The function of the two self-association domains of HCF-2 is unknown. In HCF-1, SASN and SASC mediate non-covalent association of the N- and C-terminal portions of the protein after cleavage at one of the HCFPRO repeats. SASN is a 43 amino acid region identified in a natural splice variant of HCF-1 that cannot self-associate (4). SASC is a 190 amino acid region containing two tandem sequences with homology to fibronectin type 3 (Fn3) repeats (17). The cleaved N-terminal segment of HCF-1 is brought to the nucleus through SAS-mediated association with the C-terminal segment of the protein, which contains a nuclear localization signal at the extreme C-terminus. Both SAS elements are conserved in HCF-2, but HCF-2 lacks HCFPRO cleavage sites and migrates as a single species in protein gels under denaturing conditions (2).  Although there is no evidence that the two proteins associate in vivo, the HCF-2 SASN element can bind to the HCF-1 SASC element in immunoprecipitation experiments (17).

A single ortholog of HCF-1 exists in both flies and worms. Similar to HCF-2 in length, Caenorhabditis elegans ceHCF (782 aa) contains regions with homology to the β-propeller, SASN, and SASC domains, and nuclear localization signal of HCF-1 (15;18;19). ceHCF lacks the central HCFPRO repeats and is not proteolytically processed. Drosophila melanogaster dHCF is 1500 amino acids in length and contains β-propeller, SASN, and SASC domains, and a nuclear localization signal homologous to those of HCF-1 (20;21). dHCF does not contain HCFPRO repeats, but undergoes proteolytic processing via an unknown mechanism to yield N- and C-terminal fragments (21;22). Both ceHCF (15;19) and dHCF (21) can support VP16-induced complex formation and transcriptional activation. In addition, dHCF has been shown to associate with the Drosophila Ada2A-containing (ATAC) histone acetyltransferase complex (22), and with the cell cycle regulatory transcription factors dE2F1, dE2F4 (ortholog of human E2F4), as well as with human E2F1 (23) (see Background).

The Feckless mutation occurs within β-strand 4 of the fifth kelch repeat of the predicted β-propeller domain of HCF-2. Along with the SAS elements, the β-propeller domain was originally noted to be rich in the bulky hydrophobic residues tryptophan, tyrosine, and phenylalanine (6). The Feckless mutation affects such a residue.

Northern blot analysis demonstrated Hcfc2 expression in all tissues examined: spleen, thymus, prostate, testis, ovary, small intestine, mucosal lining of the colon, and peripheral blood leukocytes (2). At least five transcripts of different sizes were detected in the Northern blot. A 3.2 kb transcript was most abundantly expressed in testis. Indirect immunofluorescent staining of HeLa cells transiently transfected with a tagged HCF-2 protein revealed heterogeneity in the subcellular localization of HCF-2, which was detected variously as predominantly nuclear, predominantly cytoplasmic, or both nuclear and cytoplasmic (2). In contrast, HCF-1 is exclusively nuclear in most cell types due to a C-terminal basic nuclear localization signal (5). HCF-1 is localized in the cytoplasm associated with the Golgi apparatus in sensory neurons, where herpes simplex virus establishes latency (see Background) (24;25). The protein HPIP may serve as a nuclear export factor for HCF-1 (26).

Figure 8. Assembly of the VP16-induced transcriptional activation complex. The 3D structure of the VP16 conserved core region resembles the back and seat of a chair; the structure of the transactivation domain (TAD) has not been determined.  Once released from the virion, VP16 uses the sequence EHAY (aa 361-364) in the seat of the chair to bind to the β-propeller domain of HCF-1, an interaction reported to depend on phosphorylation of S375 in VP16.  The POU domain-containing transcription factor Oct-1 subsequently binds to at least three critical residues (YGS, aa 373-375) in the same area of the seat.  Both VP16 and Oct-1 associate with specific DNA sequences in viral immediate early gene promoters: VP16 with the sequence GARATT (R=purine), and the two segments of the POU domain, POU-specific (POUs) and POU homeodomain (POUh), with ATGC and TAAT, respectively.  The VP16-induced complex initiates HSV gene expression.

HCF-2 was identified in a search of the NCBI EST database for proteins with similarity to the N-terminal self-association domain of HCF-1 (2). Regions of HCF-2 shared similarity with the HCF-1 β-propeller, N-terminal self-association domain (SASN), and C-terminal self-association domain (SASC), and in the organization of these domains, as described above (Protein Prediction). However, the nucleotide sequence of HCF-2 was noted to differ significantly from that of HCF-1, although the importance of this fact is not known. Since its identification in 1999, no other published reports have focused specifically on the study of HCF-2, and therefore little is known about its function. The known facts about HCF-2 are discussed following a review of its closest relative, HCF-1.

Control of α-herpesvirus immediate early gene transcription by HCF-1

HCF-1 was discovered as a cellular activity required for transcriptional activation of herpes simplex virus (HSV) immediate early (IE) promoters by the viral transcriptional activator VP16 (6;7;27;28). VP16 (also known as Vmw65 or αTIF) is a HSV virion protein packaged in the tegument structure of the virus and released into the cell upon infection. It is subsequently assembled into a specific transcription enhancer complex recognizing the sequence TAATGARAT (where R represents a purine), a cis-regulatory element of viral IE promoters. The core of this enhancer complex is formed by the association of VP16 with HCF-1, which occurs in a DNA-independent manner (27-29) and may require phosphorylation of VP16 by casein kinase II (11;30), followed by association with the POU domain-containing transcription factor Oct-1 (Figure 8) (8-10). The formation of this VP16-induced complex initiates the cascade of HSV gene expression during viral lytic replication.  HCF-1 is also an essential factor for varicella zoster virus (VZV, also an α-herpesvirus) IE gene transactivation by ORF10, the VZV ortholog of VP16 (31)

It is now believed that HCF-1 serves as a critical factor regulating viral gene expression via interactions with multiple cellular transcription factors and chromatin modification components [reviewed in (32)]. The CREB/ATF family member Luman (13;16) and the bZIP protein Zhangfei (33;34) bind to the β-propeller domain of HCF-1 to either stimulate or inhibit, respectively, HCF-1-dependent activation of HSV IE genes. Sp1 and GA-binding protein (GABP) contribute to HSV IE gene expression by binding to recognition sequences adjacent to enhancer cores (35;36). HCF-1 interacts with both Sp1 (37) and GABP (38) to coordinate enhancer complex assembly and promote IE gene expression. 

Figure 9. HCF-1 couples LSD1 with Set1 or MLL to promote α-herpesvirus IE gene transcription. (A) Components of the MLL and Set1 histone methyltransferase complexes purified fromK562 erythroleukemiacells or HeLa cells by immunopurification using anti-MLL (p300) or anti-Flag antibodies (for Flag-tagged HCF-1N), respectively.  LSD1 was not identified in the Set1 complex, but coimmunoprecipitated in independent experiments with HCF-1 and RbBp5, a component of the Set1 and MLL complexes.  Although Sin3A histone deacetylase was isolated in the complex with Set1 and HCF-1, VP16 associated preferentially with complexes containing Set1 but lacking Sin3A, consistent with the role of the VP16-induced complex in transcriptional activation but not repression.  HCF-1 may associate with the Sin3A histone deacetylase complex in other contexts. (B) Model of HCF-1-dependent transcriptional activation.  Upon infection, the viral genome is marked with repressive chromatin modifcations such as histone H3K9 methylation.  A VP16-induced complex containing VP16, HCF-1, and Oct-1, as well as transcription factors Sp1 and GABP, associate with viral IE gene promoters and recruit components of the RNA Pol II machinery.  Transcriptional activation requires the HCF-1-dependent recruitment of a chromatin modification complex containing LSD1 and Set1 (or MLL, not depicted), which promote histone H3K9 demethylation and H3K4 methylation, respectively.  These modifications activate transcription by controlling the recruitment of heterochromatin-associated proteins, chromatin remodeling proteins, and histone acetyltransferases and deacetylases.

HCF-1 is a component of chromatin modification complexes that coordinate the introduction and removal of activating or repressive chromatin marks to modulate gene expression. The mixed-lineage leukemia (MLL) and Suppressor of variegation 3-9 (Su(var)3-9), Enhancer of zeste (E(z)), and Trithorax (Trx) (Set) histone methyltransferases mediate histone H3 lysine 4 methylation (H3K4), a modification associated with active transcription that promotes recruitment of nucleosome remodeling complexes, components of TFIID, and the spliceosome. HCF-1 is part of the MLL and Set1 complexes (39;40), and is required for the recruitment of Set1 and MLL1 to viral IE promoters during the initial stages of infection, leading to histone H3K4 trimethylation and transcriptional activation (Figure 9) (41). HCF-1 also associates with the histone demethylase lysine-specific demethylase1 (LSD1), likely as part of the same complex containing Set1 or MLL1. LSD1 removes either activating (H3K4) or repressive (H3K9) methyl groups depending on the identity of its binding partners.  During HSV infection, depletion of LSD1 or HCF-1 results in accumulation of repressive histone H3K9 methylation and inhibition of viral IE gene expression. The recruitment of LSD1 to the viral genome was shown to be HCF-1-dependent (42). Thus, HCF-1 couples LSD1 with Set1 or MLL1 to promote viral IE gene expression by both removing repressive chromatin and installing active chromatin marks.  HCF-1 may also associate with complexes that control histone acetylation, such as the Drosophila Ada2A-containing (ATAC) histone acetyltransferase complex (22) and the sin3a/HDAC histone deacetylase complex (39). It was recently demonstrated that HCF-1 regulates HSV DNA replication during late stages of infection by recruiting to the virus-specific DNA replication machinery the H3/H4 histone chaperone Asf1b, which modulates the flow of histones during nucleosome disassembly and reassembly to regulate the progression of the DNA replication fork (43).

Following a primary infection, HSV and VZV establish latency in neurons of the host sensory ganglia. Periodic interruption of the latent state can result in recurrent lytic infection, presumably initiated by induced transcription of viral IE genes. Because of the absolute requirement of HCF-1 for IE gene expression, HCF-1 has been postulated to control the initiation of viral reactivation from latency (44). HCF-1 is sequestered in the cytoplasm of unstimulated latently infected sensory neurons (24;25), but translocates to the nucleus and is recruited to HSV IE gene promoter-enhancer sequences upon stimulation of reactivation (45). HCF-1 occupancy of viral IE promoter domains is correlated withoccupancy by RNAPII and accumulation of viral IE mRNAs. Furthermore, latent HSV genomes are nucleosomal and inhibition of LSD1, which is recruited in an HCF-1-dependent manner during lytic infection, results in suppression of HSV reactivation from latency in explanted trigeminal ganglia (42).

Cellular transcription and cell cycle control by HCF-1

Analysis of a baby hamster kidney cell line (tsBN67) carrying a temperature sensitive proline-to-serine mutation in the third kelch repeat of the β-propeller domain of HCF-1 revealed a role for HCF-1 in cell cycle control (14). After 36-48 hours at the non-permissive temperature, tsBN67 cells stop proliferation and arrest in the G0/G1 stage of the cell cycle, from which they can reenter the proliferative cycle by transfer of the cells to the permissive temperature. The point mutation did not alter the stability or processing of HCF-1, but prevented interaction with VP16 and with chromatin at the non-permissive temperature (7;14;46).  Both the β-propeller domain and basic region of HCF-1 were shown to be important for G0-G1 progression (46-48), whereas the C-terminal fragment of HCF-1 was important for proper cytokinesis (49;50)

Figure 10. A model for regulation of G1 to S phase progression by HCF-1. During early G1, HCF-1 associates with the transcriptional repressor E2F4.  Recruitment of the Sin3 histone deacetylase into the complex by HCF-1 results in repression of S phage genes.  The transcriptional activator E2F1 replaces E2F4 as cells progress into late G1.  However, S phase genes continue to be repressed because E2F1 is bound to pRb, which recruits Sin3, Su(Var)39 histone methyltransferase, and the SWI/SNF chromatin remodeling complex to promoters.  At the G1/S border, HCF-1 binds to E2F1, an interaction that is mutually exclusive of E2F1-pRb binding.  In this complex with E2F1, HCF-1 recruits the MLL-containing histone H3K4 histone methyltranferase complex that trimethylates H3K4 and promotes transcriptional activation of genes required for S phase, including those encoding p107, cyclin A, E2F1, and Cdc25.

HCF-1 is thought to regulate cell cycle progression and cell proliferation in a manner similar to its regulation of VP16-induced transcriptional activation, through interactions with transcription factors and chromatin modifying enzymes required for cell cycling. HCF-1 interacts with E2F transcriptional regulators that orchestrate cell cycle progression (Figure 10). Individual E2F proteins function as either activators (E2F1, E2F2, E2F3a) or repressors (E2F3b, E2F4, E2F5) of transcription, and bind to members of the retinoblastoma (Rb) pocket protein family pRb [reviewed in (51)]. In serum-arrested cells, the majority of E2F-responsive promoters are occupied by repressor E2Fs such as E2F4, together with a pRb protein and a histone deacetylase (HDAC)-containing corepressor complex.  As cells reenter G1 phase, activator E2Fs replace E2F4 in this complex.  Subsequent release of pRb upon phosphorylation by a cyclin-dependent kinase leads to gene activation and initiation of S phase. This switch is accompanied by chromatin remodeling and the appearance of activating marks including acetylation of histone H3 and H4 and methylation of histone H3K4.

E2F1, E2F3a, and E2F4 contain HCF-binding motifs (HBM, sequence D/EHXY) and coimmunoprecipitate with HCF-1 (23;52;53). In addition, inactivation of the pRb proteins through expression of the simian virus 40 (SV40) large T antigen or the adenovirus E1A protein rescues the proliferation defect and G0/G1 arrest of tsBN67 cells at the non-permissive temperature (54). It was subsequently demonstrated that HCF-1 interacts with E2F1, E2F3a, and E2F4 via their HBMs in a cell cycle-dependent manner in HeLa cells and/or in yeast two-hybrid experiments (23). Although the overall abundance of HCF-1 is stable throughout the cell cycle, HCF-1 binds to E2F1 during late G1 and at the G1/S boundary, and to E2F4 during early G1 and S phases, i.e. when these E2Fs are active. These associations occur on the promoters of E2F-responsive genes. Furthermore, HCF-1-E2F1 complexes bind preferentially to activating MLL/Set histone methyltransferases, whereas HCF-1-E2F4 complexes bind preferentially to repressive Sin3 histone deacetylase. Thus, HCF-1 appears to recruit activating and repressive chromatin modifiers to E2F-responsive genes in a stage-specific manner to control cell cycle progression.

E2F proteins also link DNA damage recognition pathways to cell cycle checkpoints. For example, E2F1 can induce apoptosis as part of the DNA damage response by activating transcription of pro-apoptotic genes such as p14ARF, Apaf1, and p73 (51). A report suggests that HCF-1 is required for E2F1-induced apoptosis in the response to DNA damage. HCF-1 binds to E2F1 and MLL histone methyltransferase and transactivates pro-apoptotic genes (55).

Several other factors involved in cell cycle regulation are also reported to interact with HCF-1: Miz1 is a c-Myc interacting protein that induces G1 cell cycle arrest; binding of HCF-1 inhibits the recruitment of coactivator p300/CBP and thereby Miz1-mediated cell cycle arrest (56). HCF-1 binds to GABP to coactivate the pRb promoter in myoblasts, resulting in pRb expression and exit from the cell cycle during myogenesis (57). The C-terminal fragment of HCF-1 negatively regulates the level of expression of PR-Set7, a histone H4K20-specific methyltransferase (58). Knockdown of the HCF-1 C-terminal fragment results in excessive dimethylation of histone H4K20 and extensive mitotic defects in chromosome alignment and segregation, consistent with the reported effect of this portion of HCF-1 on cytokinesis. The mechanism by which HCF-1 downregulates PR-Set7, or through which H4K20 methylation by PR-Set7 controls chromosome behavior and progression through mitosis, remains unknown. As mentioned above, HCF-1 interacts with cellular transcription factors to control VP16-induced gene expression, and may similarly do so to regulate cell cycle progression. In addition to Luman, Zhangfei, Sp1, and GABP, the transcription factors Krox20 (53) and Thap1 (59), as well as transcriptional coactivators FHL2 (60), PGC-1β (61), and PRC (62) have been shown to interact with HCF-1. Luman does not appear to participate in cell cycle regulation by HCF-1 (63).  Ronin, a putative pluripotency factor of embryonic stem cells containing a zinc finger DNA binding motif, also binds directly to HCF-1 (64). Alterations in cellular gene expression were observed when HCF-1 was restricted from entering the nucleus (65).


HCF-2 shares significant similarity to HCF-1 in the β-propeller and the two self-association domains, and can stabilize the VP16-induced complex containing the IE gene enhancer DNA sequence, VP16, and Oct-1 (2;15). However, HCF-2 cannot support transcriptional activation by the VP16-induced complex. In addition, stable expression of HCF-2 fails to rescue the cell proliferation defect of tsBN67 cells at the non-permissive temperature. It may be that the difference in transcriptional activity of HCF-1 and HCF-2 is due to the absence in HCF-2 of a critical domain in HCF-1, or the presence in HCF-2 of an inhibitory domain lacking in HCF-1. Coexpression of HCF-2 actually inhibits rescue by HCF-1 (2). This finding suggests that the function of HCF-2 might be to inhibit VP16, perhaps by sequestering a common interacting partner, and thereby promote host resistance to HSV infection.

Like HCF-1, HCF-2 was identified as a member of the MLL histone methyltransferase complex (40). The complex was purified from K562 erythroleukemia cellsusing conventional chromatography followed by immunopurification with an antibody against the p300 subunit of MLL. In fact, HCF-2 was coimmunoprecipitated with MLL much more efficiently than HCF-1, suggesting that HCF-2 is preferred over HCF-1 in the MLL complex. The composition of the MLL complex shares considerable similarity with the Set1 complex; however, HCF-2 was not identified in the Set1 complex. Thus, HCF-2 may participate in the regulation of chromatin modification.

Figure 11. HCFC2 is necessary for Tlr3 transcription by promoting the binding of IRF1 and IRF2 to the Tlr3 promoter. HCFC2 forms a complex with IRF1 and IRF2, which are required for induced and basal Tlr3 transcription, respectively. HCFC2 appears to dissociate from IRF1 and IRF2 upon their binding to their DNA targets. We hypothesized that HCFC2 may generate a conformational change in the IRF that favors DNA binding.

Putative Mechanism
Figure 12. HCFC2 binds to IRF1 and IRF2 to facilitate their binding to the Tlr3 IRF-E. Immuno precipitation and immunoblot analysis of lysates of 293T cells transiently overexpressing FLAG-tagged IRF1 or IRF2 and HCFC2. Figure adapted from (1).
The β-propeller domain is the most highly conserved region between HCF-1 and HCF-2. In HCF-1, this domain has been implicated in numerous interactions: with proteins required for VP16-induced IE gene expression during α-herpesvirus infection, and with proteins required for cellular transcription, cell proliferation, and cell cycle progression. HCFC2 facilitates the binding of IRF1/2 to the Tlr3 promoter (namely to the Tlr3 IRF-E) by forming complexes with IRF1/2 (Figure 12(1). HCFC2 kelch repeats 5 and 6 were minimally required for interaction with IRF2. The mutation, which lies in a β-strand of the fifth kelch repeat, substitutes a conserved tryptophan residue with arginine, possibly disrupting the structure of the β-strand and the β-propeller as a whole.

In addition to regulating Tlr3 transcription, IRF1/2 and HCFC2 regulate the transcription of several interferon-regulated genes (1). ChIP sequencing determined that 381 DNA binding sites were differentially enriched in either wild-type or Hcfc2-deficient (Hcfc2-/-) mouse embryonic fibroblasts: 365 DNA sequences were immunoprecipitated at higher levels in wild-type samples, while 16 were immunoprecipitated at higher levels in the Hcfc2-/- samples. RNA sequencing of Hcfc2-/- bone marrow-derived macrophages (BMDM) found that 571 genes were differentially expressed in both Hcfc2-/- and Irf2-/- BMDMs compared to that in wild-type BMDMs. Analysis of the ChIP and RNA sequencing data showed that 31 genes exhibited reduced association with IRF2 and altered transcript levels in Hcfc2-/- samples compared to wild-type samples: 13 were significantly reduced and 18 were significantly increased in the Hcfc2-/- samples. Taken together, HCFC2 is proposed to facilitate IRF2 binding to several target genes, supporting the function of IRF2 as both a transcriptional activator and a repressor. Further analysis determined that HCFC2/IRF2 promotes the expression of known immune response genes that function in immune defense.

Primers Primers cannot be located by automatic search.
Feckless genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide insertion.
PCR program
1) 95°C             2:00
2) 95°C             0:30
3) 56°C             0:30
4) 72°C             1:00
5) repeat steps (2-4) 29X
6) 72°C             7:00
7) 4°C              ∞
Primers for sequencing
Feckless_seq(F): 5’- ACCTAAACCTCGGTGAGTCTG -3’
The following sequence of 693 nucleotides (from Genbank genomic region NC_­­­­­000076 for linear genomic sequence of Hcfc2, sense strand) is amplified:
14504                                                agatgca ccagctcgtt
14521 ttcttaccta aacctcggtg agtctggaaa ctgacccatt gacgtgatgt ctggtactaa
14581 agtaattttt gttattttgc tgcttaagag cagtacctga gtcatggaag ttcctttctt
14641 tctagataca gcagagtgga ctactctagt atcagattct caggaagata aaaaaaattc
14701 aaggccaaga ccaagagcag gacactgcgc tgttgcaatt ggcactcgac tatacttttg
14761 gagtggaaga gatggctaca aaaaagcact aaatagtcag gtctgctgca aggacctttg
14821 gtatcttgat actggtaggt aagagtactt aattaagttc taactgtagg aactgtgtct
14881 gtgtctaatg atgcaaagtg aatataattt cttctctttc ttagaacatg aaatgagtgt
14941 gaggaatgtg taactactcc ccactccaag gaagtctact gactcctctt tccttgaggc
15001 agtactcact cttaattact gagccatctc tccttctaat aacttttttg tacagatggt
15061 tgtaagcctt catgtggttg ttgggaattg aatttactac ctctgcttgc tcaggtcggc
15121 ccccctcgct cagtccctgc tggctctggc ccaaagatgt agttattact atactgtgac
15181 tgcacactgt atggct
Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated T is indicated in red.
Science Writers Eva Marie Y. Moresco, Anne Murray
Illustrators Diantha La Vine
AuthorsLei Sun, Zhengfan Jiang, Hua Huang, Michael Berger, Owen M. Siggs, Bruce Beutler