Abstract

Heart and neural crest derivatives expressed 1 (HAND1) is a basic helix-loop-helix (bHLH) transcription factor essential for mammalian heart development. Absence of Hand1 in mice results in embryonal lethality, as well as in a wide spectrum of cardiac abnormalities including failed cardiac looping, defective chamber septation and impaired ventricular development. Therefore, Hand1 is a strong candidate for the many cardiac malformations observed in human congenital heart disease (CHD). Recently, we identified a loss-of-function frameshift mutation (p.A126fs) in the bHLH domain of HAND1 frequent in hypoplastic hearts. This finding prompted us to continue our search for HAND1 gene mutations in a different cohort of malformed hearts affected primarily by septation defects. Indeed, in tissue samples of septal defects, we detected 32 sequence alterations leading to amino acid change, of which 12 are in the bHLH domain of HAND1. Interestingly, 10 sequence alterations, such as p.L28H and p.L138P, had been identified earlier in hypoplastic hearts, but the frequent p.A126fs mutation was absent except in one aborted case with ventricular septal defect and outflow tract abnormalities. Functional studies in yeast and mammalian cells enabled translation of sequence alterations to HAND1 transcriptional activity, which was reduced or abolished by certain mutations, notably p.L138P. Our results suggest that HAND1 may also be affected in septation defects of the human hearts, and thus has a broader role in human heart development and CHD.

INTRODUCTION

Heart and neural crest derivatives expressed 1 (HAND1) is a basic helix-loop-helix (bHLH) transcription factor implicated in several developmental programs and carcinogenesis ( 1–6 ). As a member of the class B (tissue-restricted) transcription factors, HAND1 is capable of forming heterodimers with class A (near-ubiquitous) E -factors, such as those encoded by TCF3 ( E2A namely, E12 or E47), to regulate transcription of downstream target genes by binding to an E-box or more specifically to a degenerate Thing 1/D-box (reviewed in 7 ). Furthermore, HAND1 may also form homodimers and can interact with other tissue-restricted bHLH proteins, such as HEY proteins and with its most closely related bHLH factor, notably HAND2 ( 8 ). Besides its broad ability to dimerize with other nuclear proteins, the mechanism by which HAND1 regulates transcription is complex, and may require tertiary HAND1–protein interactions with non-bHLH factors ( 9 ). Adding another layer of complexity, HAND1 may act as a transcriptional activator or a repressor depending on the target sequence (D-box or E-box, respectively) and dimerization partner ( 10 ). Such repression function occurs through sequestration of class A bHLH factors from E-boxes or by direct inhibition of their transcriptional activity.

Essentially, HAND1 plays a key role in heart development as has been shown in knock-out mice studies ( 1 , 2 ). Absence of Hand1 caused embryonal lethality, failed cardiac looping, defective chamber septation and impaired ventricular development. Its extraembryonic defects causing early lethality did, however, complicate dissection of the precise function of Hand1 in cardiac development especially in later stages. Such uncertainty was circumvented by cardiac-specific deletion of Hand1 in the developing mouse heart, which resulted in a spectrum of heart defects including ventricular septal defects (VSDs), atrioventricular valve (AV) and outflow tract abnormalities ( 11 ). Furthermore, the role of Hand1 in the dorsal–ventral patterning of the embryonic heart and the interventricular septum in transgenic mice had been determined ( 12 ). Specifically, Hand1 was targeted to the Mlc2v locus, which is expressed throughout the ventricular and septal myocardium. Hand1 knock-in mice died at midgestation and completely lacked interventricular septum, while cardiac deletion of Hand1 caused expansion of this region of the heart. Thus, diverse cardiac abnormalities such as hypoplastic ventricles, septation defects, valve defects and outflow tract abnormalities can result from a defective Hand1.

To better understand molecular causes of congenital heart disease (CHD) in humans, we have been studying different cardiac-specific transcription factor genes in a collection of malformed hearts. For instance, we analyzed HAND1 in a set of hypoplastic hearts and identified a frequent frameshift (p.A126fs) mutation in the bHLH domain ( 13 ). Indeed, p.A126fs is a loss-of-function mutation incapable of interacting with its class A partners, to stimulate the D-box- or to repress the E-box-containing promoters. We also found other infrequent HAND1 mutations, which although not predictive for the hypoplastic condition, may have contributed to the other cardiac malformations likewise present in the analyzed set of hearts.

We continued our search for HAND1 mutations in another set of 68 malformed hearts affected primarily by septal defects. In the affected heart tissues, we found 32 different mutations that would result in change of amino acids (non-synonymous), 12 of which are located in the sequences encoding the bHLH domain of HAND1. Functional studies with expressed HAND1 mutant proteins revealed that transcriptional activity was reduced or abolished by specific mutations. Our findings suggest that HAND1 may have a broader role in cardiac abnormalities, and sequence alterations leading to a defective protein may contribute to septation defects of the human hearts.

RESULTS

HAND1 sequence alterations are frequent in human hearts with septation defects

After direct sequencing of PCR fragments, we searched for sequence alterations leading to change in amino acid (non-synonymous mutations) within the coding region of HAND1 . We were particularly interested in mutations that would affect the bHLH domain at nt c.280–450 (amino acids 94–150). The HAND1 mutations detected in heart tissues in the vicinity of septal defects (‘affected’) as well as from matched (‘unaffected’) heart tissues are presented in Figure  1 A and B. Furthermore, Supplementary Material, Tables S1–S4 provide the mutation distribution in both affected and unaffected heart tissues, as well as description of the type of cardiac malformations in each patient. For clarification, unaffected tissues were taken away from the septal defects, of the same malformed hearts. For instance, in the case of VSDs, unaffected samples were obtained from normal atria. Noteworthy, the malformed hearts analyzed here had cardiac malformations other than septal defects ( Supplementary Material, Tables S1–S4 ).

Figure 1.

Schematic presentation of the coding sequence and conserved regions of HAND1, as well as spectrum and location of mutations in affected ( A ) and unaffected ( B ) tissues of patients with cardiac malformations, notably septal defects. The coding sequence of HAND1 is 645 bp (215 amino acids): c.25–51 poly-histidine, c.190–213 poly-alanine, c.280–318 basic and c.319–450 helix-loop-helix. More mutations were detected in the N-terminal and bHLH regions than in the C-terminal. Mutations detected also in hypoplastic hearts are indicated by asterisks.

Figure 1.

Schematic presentation of the coding sequence and conserved regions of HAND1, as well as spectrum and location of mutations in affected ( A ) and unaffected ( B ) tissues of patients with cardiac malformations, notably septal defects. The coding sequence of HAND1 is 645 bp (215 amino acids): c.25–51 poly-histidine, c.190–213 poly-alanine, c.280–318 basic and c.319–450 helix-loop-helix. More mutations were detected in the N-terminal and bHLH regions than in the C-terminal. Mutations detected also in hypoplastic hearts are indicated by asterisks.

In the 68 affected heart tissues, we found 32 different non-synonymous mutations. Of these 32, 12 are located in the sequences encoding the bHLH domain of HAND1 (Fig.  1 A). We also identified 24 mutations in matched unaffected heart tissues and 10 are in the bHLH region as well (Fig.  1 B). Altogether, we found 49 different sequence alterations, of which 25 (51%) were detected in affected tissues only, 17 (35%) in unaffected tissues only and 7 (14%) in both. These sequence alterations were not detected in 10 formalin-fixed, normal hearts of the same collection, 5 frozen normal hearts, 12 blood samples of CHD patients, as well as blood samples of 100 unrelated normal individuals.

Moreover, more mutations whether in affected or unaffected tissues were present in VSDs than in atrial septal defects (ASDs) and atrioventricular septal defects (AVSDs). Prominent among these sequence alterations was c.83T>A (p.L28H) detected in 24 cases, namely from affected tissues in 12 of 29 VSDs, 5 of 16 ASDs and 7 of 23 AVSDs. In the matched unaffected tissues, 17 were positive for this mutation. Of the remaining sequence alterations, most were detected in only one heart sample and especially in affected tissues suggesting their somatic origin. Importantly, in an earlier mutation analysis of HAND1 in a set of hypoplastic hearts ( 13 ), 10 of these sequence alterations were found (Fig.  1 ). But a frequent mutation identified in hypoplastic hearts c.376delG (p.A126fs) was absent in this set of hearts with septation defects, except in the affected heart tissue of one case (D01VSD, aborted fetus) who had septation as well as outflow tract defects, i.e. subaortic VSD with overriding aorta and double-outlet right ventricle.

Functional studies of HAND1 mutants in yeast system reveal impairment of transactivation activities

Using the recently developed transcriptional assay in the model organism Saccharomyces cerevisiae ( 13 ), we investigated the functional consequences of a panel of HAND1 sequence alterations selected for their frequency of occurrence or their location in conserved regions of the protein, notably the bHLH. The p.L28H mutation was frequently identified in DNA samples from affected heart sections with septal defects, as well as hypoplastic ventricles ( 13 ). The mutation is located in the proximity of the histidine-rich region of HAND1, a portion of the protein where there is paucity of information on structure/function relationships. Wild-type HAND1 or p.L28H was expressed in yeast at variable levels exploiting a galactose inducible promoter along with the class A bHLH transcription factor E47 (Fig.  2 A and B and Supplementary Material, Fig. S1 ). p.L28H appeared slightly less active than the wild-type when co-expressed at high levels with E47 in a yeast reporter strain containing 12 D-box response elements (REs) repeated in tandem. On the E-box reporter strain, the inhibition of E47 intrinsic activity was slightly enhanced in comparison with wild-type HAND1. Western blot analysis revealed that the p.L28H amino acid change had no appreciable effect on steady-state expression level of the protein (Fig.  2 C). To determine whether the subtle phenotypes of p.L28H could be dependent on the expression level of the protein or the strength of the cognate RE in the reporter strains, transactivation at the D-box was examined at variable expression levels in strains containing fewer repeats of the D-box ( Supplementary Material, Fig. S1 ). In those conditions, p.L28H exhibited a wild-type phenotype. Repression of E47 was also addressed at lower level of protein expression. As in those conditions, wild-type HAND1 was able to completely abolish E47 activity, the enhanced repressive capacity of p.L28H could not be assessed.

Figure 2.

Functional analysis of HAND1 mutants using a yeast-based luciferase assay.

Inducible expression vectors coding for wild-type or mutant HAND1 were transformed in the D-box ( A and D ) and E-box ( B and E ) yeast reporter strains, along with a plasmid expressing the class A bHLH E47 protein. (A and B) Impact of N-ter HAND mutants on sequence-specific transactivation. (D and E) Analysis of HAND1 missense mutations in the bHLH domain. In all cases, purified yeast transformants were grown for 24 h on liquid selective medium containing 2% galactose to achieve high level of expression of the transcription factors. Luciferase activity was measured from total protein extracts and normalized to a unit of soluble proteins. Presented are the average light units, normalized for amount of proteins and the standard deviations of three biological replicates obtained upon expression of the indicated proteins. ( C and F ) Western blot analysis to assay the expression of the HAND1 wt and mutant constructs from total protein extracts of HAND1 transformant cells. An antibody (sc-126) directed against the chimeric TAD domain was used. The amount of soluble protein extracts loaded on a 12% acrylamide gel is indicated.

Figure 2.

Functional analysis of HAND1 mutants using a yeast-based luciferase assay.

Inducible expression vectors coding for wild-type or mutant HAND1 were transformed in the D-box ( A and D ) and E-box ( B and E ) yeast reporter strains, along with a plasmid expressing the class A bHLH E47 protein. (A and B) Impact of N-ter HAND mutants on sequence-specific transactivation. (D and E) Analysis of HAND1 missense mutations in the bHLH domain. In all cases, purified yeast transformants were grown for 24 h on liquid selective medium containing 2% galactose to achieve high level of expression of the transcription factors. Luciferase activity was measured from total protein extracts and normalized to a unit of soluble proteins. Presented are the average light units, normalized for amount of proteins and the standard deviations of three biological replicates obtained upon expression of the indicated proteins. ( C and F ) Western blot analysis to assay the expression of the HAND1 wt and mutant constructs from total protein extracts of HAND1 transformant cells. An antibody (sc-126) directed against the chimeric TAD domain was used. The amount of soluble protein extracts loaded on a 12% acrylamide gel is indicated.

Previous studies of HAND1 deletion mutants provided evidence for an inhibitory role of the Nter region on transactivation ( 10 , 14 ). To interpret the results with p.L28H in the light of those observations, we examined whether a negative effect on transactivation in the HAND1 Nter region could also be inferred using the yeast-based system. We constructed deletion mutants devoid of the initial 48 or 78 amino acids and examined their relative activity when co-expressed with E47. The deletion mutants showed only a small reduction in the capacity to inhibit E47 at the E-box and this was paralleled by a slight increase in transactivation when co-expressed with E47 at the D-box reporter. The increase in transactivation was also detected with reporter strains containing three and five D-box REs ( Supplementary Material, Fig. S1 ). Western blot analysis showed that the expression level for the deletion mutants was comparable to that of wild-type HAND1 (Fig.  2 C). We concluded from these analyses that while the presence of a transcription repressive activity in the HAND1 Nter can be confirmed by the yeast-based assays, the p.L28H amino acid change found in association with heart malformations exhibited only a subtle phenotype with a slight reduction of transactivation potential at a D-box reporter that could be observed only at high expression levels and with a slight increase in repressive capacity at the E-box reporter.

The p.A126fs hotspot mutation recently identified in a collection of hypoplastic hearts ( 13 ) is basically absent in the collection of septal defects analyzed in the present study (see mutation spectra in Fig.  1 ). However, there were other infrequent amino acid changes present in both patient cohorts that affected residues throughout the bHLH domain. Although their clinical significance is uncertain, those mutations, which affect evolutionary conserved amino acids, provided an opportunity to explore structure/function relationships of the HAND1 bHLH domain using the D-box as well as the E-box reporter. Indeed, bioinformatic predictions based on sequence conservation show most mutations to affect protein function (Materials and Methods and Table  1 ).

Table 1.

Functional significance of HAND1 mutations as determined by bioinformatic prediction and functional studies in yeast and mammalian cells

Mutation  PMut a
 
PolyPhen b
 
SIFT c
 
Yeast assay
 
Mammalian assay 
 Prediction Score Prediction Score Prediction Score D-box + E47 E-box + E47 D-box + E47 
L28H Neutral 0.4902 Possibly damaging 1.723 Affect protein function 0.03 Slight reduction of transactivation Slight increase in repressive capacity Wt-like 
K102E Pathological 0.5528 Benign 1.477 Affect protein function 0.00 Loss of function Enhanced repressive potential towards E47 Reduced function 
S112G Pathological 0.9616 Possibly damaging 1.607 Affect protein function 0.02 Partially active Enhanced repressive potential towards E47 — 
E116V Pathological 0.6454 Probably damaging 2.020 Affect protein function 0.03 Wt-like Wt-like — 
E119G Pathological 0.8936 Probably damaging 2.010 Affect protein function 0.02 Wt- like, but more active Wt-like Wt-like 
A126fs d       Loss of function Loss of function Loss of function 
L138P Pathological 0.9708 Probably damaging 2.361 Affect protein function 0.00 Loss of function Loss of function Loss of function 
V149A Pathological 0.8814 Benign 0.669 Affect protein function 0.00 Partially active Enhanced repressive potential towards E47 — 
L150P Pathological 0.9806 Probably damaging 2.228 Affect protein function 0.00 Partially active Wt-like — 
Mutation  PMut a
 
PolyPhen b
 
SIFT c
 
Yeast assay
 
Mammalian assay 
 Prediction Score Prediction Score Prediction Score D-box + E47 E-box + E47 D-box + E47 
L28H Neutral 0.4902 Possibly damaging 1.723 Affect protein function 0.03 Slight reduction of transactivation Slight increase in repressive capacity Wt-like 
K102E Pathological 0.5528 Benign 1.477 Affect protein function 0.00 Loss of function Enhanced repressive potential towards E47 Reduced function 
S112G Pathological 0.9616 Possibly damaging 1.607 Affect protein function 0.02 Partially active Enhanced repressive potential towards E47 — 
E116V Pathological 0.6454 Probably damaging 2.020 Affect protein function 0.03 Wt-like Wt-like — 
E119G Pathological 0.8936 Probably damaging 2.010 Affect protein function 0.02 Wt- like, but more active Wt-like Wt-like 
A126fs d       Loss of function Loss of function Loss of function 
L138P Pathological 0.9708 Probably damaging 2.361 Affect protein function 0.00 Loss of function Loss of function Loss of function 
V149A Pathological 0.8814 Benign 0.669 Affect protein function 0.00 Partially active Enhanced repressive potential towards E47 — 
L150P Pathological 0.9806 Probably damaging 2.228 Affect protein function 0.00 Partially active Wt-like — 

a PMut predicts pathological mutations based on specific location in a protein sequence ( http://mmb2.pcb.ub.es:8080/PMut/ ).

b PolyPhen (= Poly morphism Phen otyping) predicts possible impact of an amino acid substitution on the structure and function of a human protein using straightforward physical and comparative considerations ( http://genetics.bwh.harvard.edu/pph/ ).

c SIFT ( S orts I ntolerant F rom T olerant amino acid substitutions) predicts whether an amino acid substitution affects protein function based on sequence homology and the physical properties of amino acids ( http://sift.jcvi.org/ ).

d Yeast assay reported previously [Reamon-Buettner et al . ( 13 )].

Overall, we tested seven single amino acid changes and two double mutants (Fig. 2 D and E). Using the D-box reporter assay, three groups of mutants could be distinguished. p.K102E, p.L138P and the two double mutants containing p.L138P appeared as loss-of-function. p.S112G, p.V149A and p.L150P were partially active, whereas p.E116V and p.E119G were wild-type like. p.E119G appeared even slightly more active than wild-type HAND1. This phenotype was confirmed with the weaker 3x D-box strain ( Supplementary Material, Fig. S2 ). In that strain, we also tested p.E116V, which confirmed the wild-type phenotype, and p.V149A appeared wild-type and not a partial-function mutant. With the E-box reporter assay, three different phenotypes were also observed, but the groupings of the mutants differed. p.K102E, p.S112G and p.V149A exhibited an enhanced repressive potential towards E47. p.E116V, p.E119G and p.L150P were wild-type-like. p.L138P and the two double mutants showed reduced repression, but were not completely defective as the previously described p.A126fs frameshift mutant ( 13 ). Western blot analysis confirmed similar expression for all mutated proteins and the wild-type (Fig.  2 F).

A recent report provided evidence that HAND1 can functionally interact with myocyte enhancer factor 2C (MEF2C) and stimulate transactivation in a MEF-box dependent manner ( 14 ). To address the impact of HAND1 mutations in the context of the functional interactions with MEF2C, we developed an inducible as well as a constitutive expression vector for MEF2C and a MEF-box reporter strain. However, expression of MEF2C alone did not result in transactivation of D-, E- nor MEF-RE strains (Fig.  3 ). Similarly, co-expression of MEF2C and HAND1 did not result in transactivation of any reporter strain. MEF2C showed a weak negative functional interaction with E47 in the E-box reporter assay. MEF2C appeared to reduce HAND1, E47-dependent transactivation at the D-box, but did not have an impact over that of HAND1 at the E-box. No activity was observed on the MEF-box strain even upon co-expression of the three proteins. Similar results were obtained with the constitutive expression vector for MEF2C ( Supplementary Material, Fig. S3 ). As we could not recapitulate in yeast the cofactor activity of HAND1 on MEF2C-dependent transactivation at its cognate RE, the HAND1 mutants could not be further evaluated.

Figure 3.

HAND1, E47 and MEF2C co-expression in yeast reporter cells. To demonstrate the possible interaction of three different partners involved in cardiomyogenesis, we performed luciferase assays as described in Figure  2 in yeast reporter strains containing either D-or E- or MEF-boxes upstream of the luciferase reporter. The strains were transformed with inducible expression vectors coding HAND1, E47 or MEF2C as indicated. Presented are the relative average light units and the standard deviations of three biological replicates.

Figure 3.

HAND1, E47 and MEF2C co-expression in yeast reporter cells. To demonstrate the possible interaction of three different partners involved in cardiomyogenesis, we performed luciferase assays as described in Figure  2 in yeast reporter strains containing either D-or E- or MEF-boxes upstream of the luciferase reporter. The strains were transformed with inducible expression vectors coding HAND1, E47 or MEF2C as indicated. Presented are the relative average light units and the standard deviations of three biological replicates.

We also addressed the impact of the HAND1 mutations on the functional interaction with E12, a splice variant of E47 that differ in the bHLH domain ( Supplementary Material, Fig. S4A and B ). E12 and E47 are both derived from a single gene, TCF3 ( E2A ) ( 15 ). The sequences encoding E47 is 1956 bp, while for E12 is 1965 bp. Sequence alignment showed 100% homology in the first 1587 and in the last 150 bases. In our system, E12 showed a reduced intrinsic activity toward the E-box RE compared to E47, which could not be linked to protein levels ( Supplementary Material, Fig. S4C ). While the overall results with HAND1 mutants were similar, some amino acid changes showed a partner-dependent impact ( Supplementary Material, Fig. S4A, B ). For example at the D-box, deletion of the initial 78 amino acids did not result in higher transactivation, p.K102E retained some activity, p.E119G was more active and the double mutant containing p.E119G+p.L138P that retained some activity although p.L138P alone confirmed to be a loss-of-function allele. In the E-box assay, p.L28H did not show enhanced repression and the double mutant with p.L138P differed from p.L138P alone and resulted in wild-type level of repression.

Functional studies of HAND1 mutants in mammalian cells further reveal impairment of transactivation activities

Additionally, we conducted studies with selected HAND1 mutations using a validated D-box sequence, that enabled specifically an examination of HAND1-dependent transactivation ( 10 ), by transient transfection of the mouse cell line P19Cl6, which differentiates into beating cardiomyocytes in the presence of 1% DMSO (reviewed in 16 ). On the D-box reporter, transfection of E47 or E12 alone showed no activity, but co-transfection with HAND1 led to transactivation (Fig.  4 A and B). Mutants p.A126fs and p.L138P abolished this effect, whereas p.K102E resulted in less transactivation (see also Supplementary Material, Fig. S5 ). Furthermore, p.L28H and p.E119G did not vary considerably from HAND1 wild-type. Results varied from the transfection of wild-type HAND1 alone (compare Fig.  4 A and B or Supplementary Material, Fig. S5 ). This difference might be related to a variation in transfection efficiencies among experiments, as revealed based on arbitrary scoring of DsRed2 fluorescence (Fig.  4 C). Endogenous bHLH proteins could interact with the transfected HAND1 leading to transactivation of the reporter and higher levels of transfected HAND1 protein might reveal this effect. Overall, this assay confirmed the results that p.A126fs and p.L138P are loss-of-function mutations, whereas p.K102E had a reduced transactivation activity. Moreover, the E-box reporter (Fig.  4 D and E) was active on its own. No further increase in activity was observed upon co-transfection with the E47 or E12 expression vectors nor repression was detected by cotransfection with HAND1 wt; thus preventing the assessment of the impact of the HAND1 mutants.

Figure 4.

Functional analysis of HAND1 mutants in a mammalian-based luciferase assay. ( A and B ) Transactivation activities of HAND1 wt and mutants, with either E12 or E47 on the D-box reporter consisting of three copies of CGTCTG cloned upstream of a minimal promoter driving firefly luciferase. ( C ) HAND1 wt and mutants as well as E12 and E47 were cloned in pIRES2-DsRed2 allowing control of transfection efficiency using fluorescence microscopy. Shown are transfected cells for the assay with E12 in (B). Relative light values were obtained from the ratio of firefly to renilla luciferase. Statistical analysis was performed using t -test on HAND1 mutants compared with HAND1_wt. Statistical significance is indicated by asterisk(s) and corresponding P -value indicated beneath it. ( D and E ) Transactivation activities of HAND1 wt and mutants, with either E12 or E47 on the E-box reporter consisting of three copies of CATCTG. ( F ) Transfection control for assay with E47 shown in (D).

Figure 4.

Functional analysis of HAND1 mutants in a mammalian-based luciferase assay. ( A and B ) Transactivation activities of HAND1 wt and mutants, with either E12 or E47 on the D-box reporter consisting of three copies of CGTCTG cloned upstream of a minimal promoter driving firefly luciferase. ( C ) HAND1 wt and mutants as well as E12 and E47 were cloned in pIRES2-DsRed2 allowing control of transfection efficiency using fluorescence microscopy. Shown are transfected cells for the assay with E12 in (B). Relative light values were obtained from the ratio of firefly to renilla luciferase. Statistical analysis was performed using t -test on HAND1 mutants compared with HAND1_wt. Statistical significance is indicated by asterisk(s) and corresponding P -value indicated beneath it. ( D and E ) Transactivation activities of HAND1 wt and mutants, with either E12 or E47 on the E-box reporter consisting of three copies of CATCTG. ( F ) Transfection control for assay with E47 shown in (D).

A summary of functional consequences of HAND1 mutations as determined by bioinformatic predictions and by assays in yeast and mammalian cells is presented in Table  1 . Predictions of the functional consequences on the mutations from three different software tools (see Materials and Methods) were in some cases conflicting, particularly for the p.L28H, p.K102E and p.V149A mutants. Nonetheless, there was a good correlation between the functional assays conducted in yeast and in the P19C16 mouse cells, except for the p.K102E mutation that appeared as a loss-of-function mutation in the yeast system, but was only partially defective in the mouse cells. The reason for this discrepancy is not clear, but might be related to the interactions with endogenous HAND1 and cofactor proteins in the P19Cl6 cells.

Potential pathogenic HAND1 mutations and corresponding malformations of human hearts

On the basis of functional studies in yeast and mammalian cells, as well as their frequency or presence in affected tissues of malformed hearts, a contribution to the disease by certain HAND1 mutations may be inferred. A strong candidate is p.L138P which has been consistently predicted to affect protein function and shown in both yeast and mammalian assays to result in loss of transactivation activities of HAND1 (Table  1 ). This mutation was detected in two malformed hearts affected by ASD or AVSD. The AVSD (F13AVSD), besides p.L138P was also positive for two other bHLH mutations p.E116V and p.E119G (Fig.  5 A), two mutations also predicted to be pathological but so far found to be wild-type like in functional assays used. The ASD (F24ASD) was likewise positive for p.E119G and p.L138P. Unlike p.E116V, we detected p.E119G and p.L138P only in affected heart tissues.

Figure 5.

Potentially pathogenic HAND1 mutations and corresponding cardiac malformations of patients. ( A ) Presence of three ‘heterozygous’ bHLH mutations [c.347A>T (p.E116V), c.356A>G (p.E119G), c.413T>C (p.L138P)] within an amplified fragment in F13AVSD. Cloning and sequencing of five clones revealed a clone with p.E116V only, another clone with p.E116V + p.L138P, and three clones with p.E119G + p.L138P. ( B ) Lifespan of patients in hypoplastic hearts group where p.A126fs, p.L138P or both were identified. ( C ) Cloning of a PCR fragment amplified from the hypoplastic ventricle of B35HLHS, and which was heterozygous for p.A126fs, p.E116V and p.L138P, shows a clone with p.A126fs only, another clone with p.E116V + p.L138P and a third clone with wild-type or reference sequence. ( D ) Cardiac malformations of patients in both hypoplastic hearts and septation defects groups where p.L138P was identified in different combinations with p.E116V, p.E119G and p.A126fs. ( E and F ) Sequence electropherograms and PCR-RFLP assays for c.304A>G (p.K102E) and c.83T>A (p.L28H) showing homozygous reference (wild-type), homozygous mutant and heterozygous genotypes in amplified PCR fragments. U ctrl., undigested PCR product.

Figure 5.

Potentially pathogenic HAND1 mutations and corresponding cardiac malformations of patients. ( A ) Presence of three ‘heterozygous’ bHLH mutations [c.347A>T (p.E116V), c.356A>G (p.E119G), c.413T>C (p.L138P)] within an amplified fragment in F13AVSD. Cloning and sequencing of five clones revealed a clone with p.E116V only, another clone with p.E116V + p.L138P, and three clones with p.E119G + p.L138P. ( B ) Lifespan of patients in hypoplastic hearts group where p.A126fs, p.L138P or both were identified. ( C ) Cloning of a PCR fragment amplified from the hypoplastic ventricle of B35HLHS, and which was heterozygous for p.A126fs, p.E116V and p.L138P, shows a clone with p.A126fs only, another clone with p.E116V + p.L138P and a third clone with wild-type or reference sequence. ( D ) Cardiac malformations of patients in both hypoplastic hearts and septation defects groups where p.L138P was identified in different combinations with p.E116V, p.E119G and p.A126fs. ( E and F ) Sequence electropherograms and PCR-RFLP assays for c.304A>G (p.K102E) and c.83T>A (p.L28H) showing homozygous reference (wild-type), homozygous mutant and heterozygous genotypes in amplified PCR fragments. U ctrl., undigested PCR product.

We cloned the ‘heterozygous’ PCR products obtained from F24ASD and F13AVSD to determine whether p.E116V, p.E119G or p.L138P are located on the same or different alleles. After sequencing of five clones from F13AVSD, we identified a clone with E116V only, another clone with p.E116V + p.L138P and three clones with p.E119G + p.L138P, indicating the presence of three haplotypes (see F13AVSD, Fig.  5 A). Moreover, analysis of four clones from F24ASD with p.E119G and p.L138P found two clones with p.E119G + p.L138P and another two clones with the reference (wild-type) sequence.

Interestingly, we detected p.L138P in five hypoplastic hearts previously ( 13 ). Owing to its infrequent detection when compared with the frameshift mutation p.A126fs which was present in 24 of 31 hypoplastic ventricles, no functional analysis had been undertaken at that time. Nonetheless, p.L138P was observed so far only in hypoplastic ventricles, but not in matched non-hypoplastic ventricles. We thus re-visited our data on the heart collection of hypoplastic hearts. In those five hearts with p.L138P, four had also p.A126fs and the presence of both mutations seems to lead to a much shorter lifespan of patients (Fig.  5 B). Again, p.L138P was found to be in combination with p.E116V or p.E119G.

Similarly, we cloned PCR fragments from three cases heterozygous for p.A126fs, p.E116V and p.L138P. We found three haplotypes (alleles) as follows: the frameshift mutation residing in one allele, the double mutant (p.E116V + p.L138P) in another allele and a third allele containing the wt sequence (see B35HLHS, Fig.  5 C). Furthermore, analysis of four clones amplified from the hypoplastic ventricle of one patient with p.A126fs, p.E119G and p.L138P gave the following results: two clones with p.A126fs, a clone with p.E119G + p.L138P and another clone with the wt sequence. This result shows that p.A126fs mutation occurs in a different allele as p.L138P, which appears to be linked with p.E116V or p.E119G.

In the light of these findings, we re-analyzed the cardiac malformations in all patients ( n = 5 in hypoplastic hearts and n = 2 in septal defect groups) harboring p.L138P, in combination with p.A126fs, p.E116V or p.E119G (Fig.  5 D). In the five hypoplastic hearts, we also found the following malformations among others: hypoplastic left or right ventricles including a single ventricle, hypoplastic ascending aorta (in two), aortic valve atresia (in two), preductal aortic isthmus stenosis (in three), tricuspid valve stenosis (in two). Of the two septal hearts, F13AVSD had also aortic isthmus stenosis, whereas F24ASD was affected additionally with partial anomalous venous return, a defect in which a few of the pulmonary veins return to the right atrium instead of the left atrium. Altogether, all but one (C18HRHS) with p.L138P had defects involving the left heart (Fig.  5 D).

p.K102E impaired transactivation activities of HAND1. It may be potent in homozygous condition, and being located in the basic region, is predicted to affect protein function as well (Table  1 ). Indeed, it will change the charge of the protein. In both yeast and mammalian assays, p.K102E exhibited loss or reduced transactivation function. Notably, p.K102E was found in both homozygous and heterozygous conditions after direct sequencing and PCR-RFLP with Sty I (Fig.  5 E). This mutation was identified in two patients with ASDs. C75ASD who was homozygous for p.K102E, had septal defects, hypoplastic left ventricle, mitral valve stenosis, preductal aortic isthmus stenosis, whereas F20ASD with the heterozygous genotype had only septal defect. p.K102E was so far only detected in affected tissues. Interestingly, C75ASD with the homozygous genotype for p.K102E had hypoplastic ventricle, suggesting a possible role of p.K102E when homozygous in hypoplasia of the heart.

Besides p.L138P and p.K102E, other mutations, i.e. p.L28H, p.S112G, p.V149A and p.L150P were wild-type like or exhibited only subtle effects in the functional assays. Notably, p.L28H is a highly frequent mutation, and after direct sequencing and PCR-RFLP with Mbo II, homozygous as well as heterozygous genotypes were obtained (Fig.  5 F). We observed five VSDs, one ASD and one AVSD homozygous for the p.L28H genotype, which is suggestive of LOH events. Remarkably, p.L28H was also detected in 15 of 31 hypoplastic hearts ( 13 ). Although p.L28H is predicted to affect protein function (Table  1 ), functional assays in yeast and mammalian showed only wild-type like or subtle effects. Being frequently detected, this mutation may also contribute to cardiac malformations in concert with other HAND1 mutations or perhaps with mutations from other transcription factor genes such as NKX2-5 or GATA4 .

DISCUSSION

This study aimed for an improved understanding of the importance of HAND1 mutations in septation defects of the human heart. On the basis of studies in mammalian cells and yeast functional assays, we were able to determine a possible contribution of these mutations to the disease phenotype. Although different mutations in HAND1 were detected, none were frequent except p.L28H. It is interesting to note, however, that the location of HAND1 mutations in patients was not randomly distributed. Notably, mutations were more frequently detected in the N-terminal region, and the bHLH domain, which is essential for DNA binding and protein–protein interactions of HAND1 (Fig.  1 ). Indeed, in affected tissues, a mutation in the bHLH domain occurred 18 times, whereas in the N-terminal region 39 times. These mutations in the N-terminal were mostly within the first 48 amino acid residues, a region identified previously for strong transcription repression activity ( 10 ). Owing to technical difficulties, we did not have data on 11 amino acids (i.e. amino acids 181–191), but on the remaining 54 amino acids on the C-terminal mutations were rare. This observation seems to support an earlier finding that amino acid residues from 150 to 215 have little role in DNA binding or transcriptional transactivation ( 10 ). Obviously, the identified HAND1 mutations affected important regions of HAND1, even though we were unable to determine a specific hotspot in HAND1 .

It is of considerable importance that in the case of NKX2-5 and GATA4 common mutations in the DNA binding domains were identified in the same collection of hearts with complex septation defects. This contrast findings recently published by us where a frequent frameshift mutation (p.A126fs) was observed in another group of patients with hypoplastic hearts ( 13 ). Notably with NKX2-5 , we found in 22 of 23 AVSDs a single p.K183E mutation in the third helix of homeodomain, which resulted in transcriptional inactivation. None of the VSDs had this mutation ( 17–19 ). For GATA4 , we found a zinc finger mutation (p.C292R) in 19 of 29 VSDs, 6 of 16 ASDs and 6 of 23 AVSDs. This mutation, which is located in the C-terminal zinc finger, affects one of the four zinc coordinating cysteines and is predicted to disrupt the secondary structure of the C-finger ( 20 ). The C-terminal zinc finger of GATA4 is required for its many interactions, notably with the third helix (homeodomain) of NKX2-5 resulting in transcriptional synergy of regulated genes (reviewed in 21 ).

Although little is known about HAND1-targeted genes, the cardiac atrial natriuretic factor ( ANF ) promoter can be directly activated by HAND1, making it the first known HAND1 transcriptional target ( 14 ). This action of HAND1 does not rely on the classical model requiring heterodimerization with class A bHLH factors or DNA binding through E-box elements. Instead, HAND1 appears to be recruited to the ANF promoter through physical interaction with MEF2 proteins resulting in synergistic activation. Furthermore, the MEF2 binding site in the ANF promoter, unlike the E-box, was necessary to mediate this synergy. We conducted experiments in yeast to evaluate whether MEF2C RE is a necessary and sufficient cis -acting mediator of HAND1–MEF2C functional interaction, but we could not recapitulate the cofactor activity of HAND1 on MEF2C-dependent transactivation. The failure to reproduce this phenotype could be dependent on the requirement of additional cis -acting elements in the proximal ANF promoter and/or additional transcription factors/cofactors and will be investigated in a follow-up study. This present result impaired our ability to evaluate the impact of HAND1 mutations in the context of MEF2C-dependent transcription. Nonetheless, we were able to demonstrate that certain HAND1 mutations could indeed affect HAND1 transcriptional activities, and there is basically an agreement of results from the yeast and mammalian reporter systems.

Taken collectively, we show that HAND1 was affected by mutations in a cohort of patients with septation defects of the heart, but unlike our study with another cohort of patients with hypoplastic hearts, no frequent loss-of-function mutation was identified in the bHLH domain. Here, we report infrequent and so-called ‘private’ mutations that do not appear to be linked to a specific disease phenotype, i.e. ASD, VSD or AVSD. Nonetheless, many of the mutations clustered in the DNA-binding bHLH domain and in the N-terminal region shown to be capable of repression activities, indicating the importance of these regions in human heart development. Furthermore, the spectrum of cardiac malformations of patients with HAND1 mutations seems to recapitulate those found in mice neonates and embryos that survived after conditional knock-out of Hand1 ( 11 ). Indeed, we detected more HAND1 mutations in patients with VSDs, outflow tract and AV defects, and our findings suggest a broader role of HAND1 not only confined to hypoplasia, but in septation defects of the human heart as well. Since the patients in the septal defects group had mutations affecting binding domains in other cardiac transcription factors, particularly NKX2-5 and GATA4, we propose sequence alterations in several transcription factor genes as an underlying cause of the same structural heart malformations. Consequently, a simple genotype–phenotype correlation appears less likely as a cause of CHD, and the present study provides further evidence for combinatorial interactions of mutated cardiac transcription factors affecting heart development.

MATERIALS AND METHODS

Heart and blood samples

The characterization of malformed hearts with septal defects and genomic DNA isolation have been described previously ( 17 ). Briefly, we investigated 68 malformed hearts from unrelated individuals, who mostly died at early infancy due to complex cardiac malformations. Hearts were collected from 1954 to 1982 and stored in formalin at the Institute of Anatomy, University of Leipzig, Germany. After detailed morpho-pathological characterization, hearts were classified according to their septal defects, e.g. 29 VSD, 16 ASD and 23 AVSD for genetic analysis. We also investigated 10 formalin-fixed, normal hearts from the same collection, as well as five frozen normal hearts. Furthermore, we analyzed blood samples of 12 patients affected by CHD (e.g. defects included VSD, ASD, hypoplastic left heart syndrome, transposition of the great arteries, sub-pulmonary stenosis and heterotaxy) and blood samples of 100 unrelated normal individuals. In the formalin-fixed malformed hearts, affected tissues in the vicinity of the septal defects were analyzed. Matched unaffected heart tissues from the same malformed hearts were likewise investigated. Materials used in this study were obtained in accordance to an approved protocol. J.B. has obtained approval to conduct genetic studies involving human materials from the Hannover Medical School. The formalin-fixed hearts which were collected more than 40 years ago and cadavers were donated voluntarily by patients’ relatives.

Genomic DNA isolation, mutation analysis

Genomic DNA including formalin-fixed material was isolated with NucleoSpin Tissue or Blood Kit (Macherey-Nagel, Dueren, Germany). HAND1 fragments within the coding region were amplified using standard procedure. Basically, a PCR reaction consisted of 20–50 ng of genomic DNA, 1× PCR buffer, 1 U of Hot Star Taq DNA polymerase (Qiagen, Hilden, Germany), 0.2 m m dNTPs, 5 pmol of each primer pair, to a volume of 25 µl with distilled water. After an initial 15 min activation step at 95°C, a typical PCR program included 35 cycles of 10 s at 94°C denaturation, 30 s at 60°C annealing and 2 min at 68°C, elongation; a final extension of 10 min, 68°C followed by indefinite 4°C. PCR reactions were carried out on Biometra thermocyclers (Biometra, Goettingen, Germany).

The human HAND1 gene was mapped to chromosome 5q32, contains two exons spanning 3.3 kb and codes for a protein of 215 amino acids. We amplified four fragments within the two exons of HAND1 using the primer pairs EHDx1-1F 5′-agaagggttaaacaggtctttgg-3′/EHDx1-1R 5′-gaagtagggcctttcctgatgac-3′; EHDx1-3F 5′-tcggtccggcctcgcgctgtcatc-3′/EHDx1-3R 5′-ctcccttttccgcttgctctcacg-3′; EHDF2-2 5′-tgggcccagtcgagaagaggat-3′/EHD2-2 R 5′-gggaaatggagatagggctgaggt-3′ and EHD2-1F 5′-tatctggctctttctctcttgtcc-3′/EHD2-1R 5′-acacagacttgaggtagaaaagggg-3′. The primers EHDx1-1F/R results in 345 bp, EHDx1-3F/R in 455 bp, EHDx2-2F/R in 444 bp and EHDx2-1F/R in 394 bp fragments. These fragments spanned the entire coding region of HAND1 , except for 31 (c.541–571) nucleotides affecting 11 (amino acids 181–191) amino acids in the carboxy region and also include most of the untranslated regions.

PCR-amplified fragments were sequenced directly with HAND1 -specific primers using BigDyeTerminator v3.1 Kit (Applied Biosystems, Darmstadt, Germany) and Applied Biosystems 3100 Genetic Analyzer. Sequences were analyzed using SeqScape 2.0 (Applied Biosystems) or DNASTAR Lasergene 7.0 (Madison, WI, USA). The numbering of mutations within the coding region of HAND1 starts with nucleotide A of the first codon ATG. Sequence variations were verified by double-strand sequencing, independent PCR, PCR-RFLP or TOPO-TA cloning (Invitrogen, Karlsruhe, Germany) of heterozygous genotypes followed by sequencing of clones, allowing the identification of variant alleles.

Furthermore, to validate our methods on genetic analysis of formalin-fixed material, we analyzed genes with or without prior implication to heart development in the same collection of malformed hearts, as well as in formalin-fixed normal mouse hearts ( 22 ). For instance, we did not observe artificially induced sequence alterations in Hand1, Gata4 and Nkx2-5 after storage of whole mice hearts in 4% buffered formalin for 1.5 months. Unless reported as NCBI dbSNPs, we refer to nucleotide changes as sequence variations or mutations interchangeably, which simply mean deviations from the reference HAND1 sequence (NM_004821) and may or may not be disease causing.

Prediction of the effect of bHLH mutations on the secondary structure of protein

To predict effects of mutations on the secondary structure, we used a model for HAND1 bHLH domain as template (ModBase, http://modbase.compbio.ucsf.edu/modbase-cgi/search_form.cgi ). Protein modeling was carried out using Swiss-PdbViewer ( http://expasy.org/spdbv ) and Protein Explorer ( http://proteinexplorer.org ). We analyzed p.K96E, p.K102E, p.S112E, p.E116V, p.E119G, p.L138P, p.V149A and p.L150P. The effect of p.A126fs was determined previously. Moreover, for prediction of the functional consequences of mutations on protein sequence, we used PMut ( http://mmb2.pcb.ub.es:8080/PMut/ ), PolyPhen ( http://genetics.bwh.harvard.edu/pph/ ) and SIFT ( http://sift.jcvi.org/ ).

Yeast assay

The identified HAND1 mutations were introduced into the previously developed pTSG-transactivation domain (TAD)-HAND1 plasmid ( 13 ).This centromeric TRP1 selective vector provides for galactose-inducible expression of HAND1 alleles with a chimeric Nter acidic TAD which was derived from the human p53 protein and previously shown to be functional in the yeast S. cerevisiae . We also generated two N-ter HAND1 deletion mutants lacking the first 48 (Δ48) or 78 (Δ78) amino acids using a PCR-based approach followed by gap repair assay in yeast ( 23 ).Wild-type HAND1 cDNA from pTSG-TAD- HAND1 plasmid was PCR amplified with primers: HAND1 Δ48-Fw, 5′- gctgctccccccgtggcccctgcagctcctacaccggcg ccggctgacgctgccccggac-3′ or HAND1 Δ78-Fw, 5′- gctgctccccccgtggcccctgcagctcctacaccggcg gggcagagccccgggcggct-3′, each one paired with HAND1 Δ-Rv, 5′- aactaattacatgatggtggcggccgctctagaactagtgg atcctcaaatcctcttctcgactgggc -3′(underlined are the homology tails for gap repair). Unpurified PCR products were cotransformed into the yeast cells using the LiAc method ( 24 ) together with pTSG-TAD-HAND1 vector double-digested with Sgr aI– Sma I endonucleases to expose the homologies with the PCR product. Transformant clones were selected on plates lacking tryptophan and then cultured overnight in rich liquid YPDA medium followed by genomic/plasmid DNA extraction performed using a modified protocol ( 25 ). Briefly, cells were harvested and resuspended in sorbitol containing buffer and incubated for 30 min at 37°C with lyticase solution (500 U; Sigma, Aldrich, Milan, Italy) to remove the cell wall. Spheroplasts were resuspended in TE buffer and lysed in 1% SDS solution at 65°C for 20 min followed by addition of potassium acetate and ice incubation for 30 min and subsequent clearing of cellular debris by centrifugation. The recovered DNA-containing solution was precipitated in isopropanol/sodium acetate solution. Chemically competent Escherichia coli cells were transformed (KCM method) and plasmid DNA was purified from ampicillin resistant clones (QiaQuick spin column, Qiagen, Milan, Italy) and the correct construction of HAND1 deletion mutants verified by restriction pattern and DNA sequencing.

The MEF2C gamma minus isoform was cloned into the pLSG-TAD yeast expression vector starting from total RNA extracted from normal heart tissue, using RNAeasy kit and RT–PCR with Omniscript RT Kit, according to the manufacturer's procedure (Qiagen, Hilden, Germany). cDNA was PCR amplified using the following primers: MEF2C-Fw, 5′-gttaactcgag caccggcg atggggagaaaaaagattcagatta-3′and MEF2C-Rv, 5′-actagt ggatcc gtc gactcatgttgcccatccttcagaaagt-3′. PCR products were inserted into the pCR®4-TOPO®vector, using the TOPO-TA cloning kit, according to the manufacturer's protocol (Invitrogen) and subsequently cloned into pLSG-TAD vector using Sgr aI– Bam HI restriction sites (present in the vector downstream the TAD sequence and introduced in the MEF2C clone with the PCR primer—underlined). This vector has the same characteristics of pTSG-TAD except for the LEU2 selectable marker replacing TRP1, to enable selection of double or triple transformants with HAND1 (TRP1) and E47 (URA3) expression vectors.

In addition to the previously described D-box (12 repeats) and E-box (6 repeats) HAND1 REs ( 13 ), we constructed a panel of D-box REs characterized by different copy number of the ‘CGTCTG’ sequence, using the delitto perfetto in vivo mutagenesis system ( 26 ). Strains where the consensus was repeated two, three and five times, respectively (named in the text as D-box #2, D-box #3 and D-box #5) were obtained starting from the yLFM-ICORE, as previously described ( 27 ). In the resulting isogenic strains, the various REs are placed upstream of a minimal promoter driving the luciferase gene located at the ade2 locus on chromosome XV of yeast. Using the same approach, we also developed a yeast reporter strain containing two copies of the MEF2 RE derived from the ANF promoter region. All primers contained tails of homology allowing targeting immediately upstream of the minimal promoter. In the case of the MEF2C reporter, two MEF2 RE sequences separated by 4 bp were introduced using primer: MEF2-ANF-RE, 5′- cggaattgactttttcttgaataatacat tactaaaaaataacttactaaaaaata ctgcagatccgccaggcgtgtatatagcgtgg -3′ (underlined are the homology tails). Correct oligonucleotide targeting in each yeast strain was verified by colony PCR and DNA sequencing.

Yeast reporter strains were transformed with the pTSG-TAD-HAND1 or the specific mutant derivatives along with available pUSG-E12/E47 expression vector ( 13 ) or the pLSG-TAD-MEF2C, following the LiAc-based protocol ( 24 ). Extracts of soluble proteins were obtained from mechanical lysis of yeast transformant cells in GLO lysis buffer (Promega, Milan, Italy) and were quantified using the BCA assay (Pierce Biotechnology, Milan, Italy). The activity of the luciferase reporter was then measured using the BrightGlo assay reagent (Promega) on a multilabel Mithras LB-940 plate reader (Berthold Technologies, Milan, Italy), as described previously ( 28 ).

Protein expression in yeast was examined by western blot. E12/47, MEF2C as well as wt or mutant HAND1 expression was induced in liquid cultures containing galactose for 6 h. Soluble protein fractions were prepared by mechanical cell disruption using 0.5 mm acid-washed glass beads (Sigma-Aldrich, Milan, Italy) in lysis buffer (100 m m NaCl, 20 m m Tris–HCl pH 7.2, 5 m m EDTA, 2 m m DTT, 0.1% NP40, 10% Glycerol, 2% SDS, 100 μg/ml PMSF, 1 μg/ml Leupeptin, 2 μg/ml Aprotinin, 1 μg/ml PepstatinA) after washing in NET buffer (100 m m NaCl, 50 m m Tris–HCl pH 8, 5 m m EDTA). Extracts were quantified (BCA assay, Pierce Biotechnology) and loaded on 12% acrylamide gels. Electrophoresis was performed at 180 V constant voltage using a miniprotein apparatus according the manufacturer's protocol (BioRad, Milan, Italy), followed by liquid protein transfer protocol onto PVDF membrane (Amersham, Milan, Italy). Immunodetection was performed using the primary antibody sc-126 directed against the TAD domain (Santa Cruz Biotechnology, Milan, Italy) and revealed using the ECL plus detection kit (Amersham).

Mammalian assay

P19CL6 cells were obtained from the Cell Bank, RIKEN BioResource Center, Japan. Cells were cultured in 96-well plates containing 100 µl of α-minimal essential medium (Invitrogen), supplemented with heat-inactivated 10% fetal calf serum, penicillin (100 U/ml) and streptomycin (100 µg/ml) at 37°C. Growth medium was changed every 2 days. On the day of transfection, cells were ∼90–95% confluent. For transfection into P19CL6 cells, plasmid DNA comprised: (i) HAND1_wt or mutant cloned in expression vector, pIRES2-DsRed2 (100 ng); (ii) E12 or E47 also in pIRES2-DsRed2 (100 ng); (iii) D-box or E-box in firefly luciferase vector, pGL4.26 luc2/minP/hygro (300 ng) and (iv) control renilla luciferase vector, pGL4.7/hRlucTK (50 ng). Total plasmid DNA concentration was kept constant with empty pIRES2-DsRed2. The HAND1_wt or mutants as well as E12 and E47 were re-cloned from yeast vectors into the mammalian vector. The D-box or E-box consisted of three copies of CGTCTG or CATCTG, respectively, as described previously ( 10 ).

Transfection was carried out using Lipofectamine 2000 (Invitrogen) according to manufacturer's protocol. Briefly, prior to transfection (1 h), cells were washed with phosphate-buffered saline, and growth medium was replaced with 100 µl of Opti-MEM I without antibiotics. Each well was transfected with 50 µl complexes obtained from combining 550 ng of endotoxin-free plasmid DNA in Opti-MEM I (25 µl) and 0.5 µl of Lipofectamine also in Opti-MEM I (25 µl). After 4–6 h, Opti-MEM I was replaced with growth medium. Fluorescence microscopy and luciferase assays were carried out after 48 h. Luciferase assays were performed using Dual-Glo™ Luciferase Assay System according to recommended protocol (Promega). Luciferase activities were measured using Wallac Victor3 1420 Multilabel Counter (Perkin Elmer) in a 96-well microtiter plate. Luciferase activity for firefly or renilla consisted of five measurements per well carried out at an interval of 2 min, and the final values are the average of three transfections. Relative light values were obtained from the ratio of firefly to renilla luciferase. Statistical analysis was performed using t -test.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online .

FUNDING

Ministry of Science and Culture, Lower Saxony, Germany; Grant number: 25A.5-76251-99-3/00 to J.B.

ACKNOWLEDGEMENTS

We thank the Institute of Anatomy, University of Leipzig for providing the heart collection; C. Murre, UCSD for the E12/E47 clones; A. Roskowetz and A. Hiemisch, for the excellent technical support. The financial support of the Lower Saxony Ministry of Science and Culture, Germany to J.B. is greatly appreciated.

Conflict of Interest statement . None declared.

REFERENCES

1
Riley
P.
Anson-Cartwright
L.
Cross
J.C.
The Hand1 bHLH transcription factor is essential for placentation and cardiac morphogenesis
Nat. Genet.
 , 
1998
, vol. 
18
 (pg. 
271
-
275
)
2
Firulli
A.B.
McFadden
D.G.
Lin
Q.
Srivastava
D.
Olson
E.N.
Heart and extra-embryonic mesodermal defects in mouse embryos lacking the bHLH transcription factor Hand1
Nat. Genet.
 , 
1998
, vol. 
18
 (pg. 
266
-
270
)
3
McFadden
D.G.
McAnally
J.
Richardson
J.A.
Charite
J.
Olson
E.N.
Misexpression of dHAND induces ectopic digits in the developing limb bud in the absence of direct DNA binding
Development
 , 
2002
, vol. 
129
 (pg. 
3077
-
3088
)
4
Fernandez-Teran
M.
Piedra
M.E.
Rodriguez-Rey
J.C.
Talamillo
A.
Ros
M.A.
Expression and regulation of eHAND during limb development
Dev. Dyn.
 , 
2003
, vol. 
226
 (pg. 
690
-
701
)
5
Barbosa
A.C.
Funato
N.
Chapman
S.
McKee
M.D.
Richardson
J.A.
Olson
E.N.
Yanagisawa
H.
Hand transcription factors cooperatively regulate development of the distal midline mesenchyme
Dev. Biol.
 , 
2007
, vol. 
310
 (pg. 
154
-
168
)
6
Martinez
H.J.
Ferraro
A.
Sacchetti
S.
Keller
S.
De Martino
I.
Borbone
E.
Pallante
P.
Fedele
M.
Montanaro
D.
Esposito
F.
, et al.  . 
HAND1 gene expression is negatively regulated by the High Mobility Group A1 proteins and is drastically reduced in human thyroid carcinomas
Oncogene
 , 
2008
, vol. 
28
 (pg. 
876
-
885
)
7
Firulli
A.B.
A HANDful of questions: the molecular biology of the heart and neural crest derivatives (HAND)-subclass of basic helix-loop-helix transcription factors
Gene
 , 
2003
, vol. 
312
 (pg. 
27
-
40
)
8
Firulli
B.A.
Hadzic
D.B.
McDaid
J.R.
Firulli
A.B.
The basic helix-loop-helix transcription factors dHAND and eHAND exhibit dimerization characteristics that suggest complex regulation of function
J. Biol. Chem.
 , 
2000
, vol. 
275
 (pg. 
33567
-
33573
)
9
Hill
A.A.
Riley
P.R.
Differential regulation of Hand1 homodimer and Hand1-E12 heterodimer activity by the cofactor FHL2
Mol. Cell. Biol.
 , 
2004
, vol. 
24
 (pg. 
9835
-
9847
)
10
Knofler
M.
Meinhardt
G.
Bauer
S.
Loregger
T.
Vasicek
R.
Bloor
D.J.
Kimber
S.J.
Husslein
P.
Human Hand1 basic helix-loop-helix (bHLH) protein: extra-embryonic expression pattern, interaction partners and identification of its transcriptional repressor domains
Biochem. J.
 , 
2002
, vol. 
361
 (pg. 
641
-
651
)
11
McFadden
D.G.
Barbosa
A.C.
Richardson
J.A.
Schneider
M.D.
Srivastava
D.
Olson
E.N.
The Hand1 and Hand2 transcription factors regulate expansion of the embryonic cardiac ventricles in a gene dosage-dependent manner
Development
 , 
2005
, vol. 
132
 (pg. 
189
-
201
)
12
Togi
K.
Kawamoto
T.
Yamauchi
R.
Yoshida
Y.
Kita
T.
Tanaka
M.
Role of Hand1/eHAND in the dorso-ventral patterning and interventricular septum formation in the embryonic heart
Mol. Cell. Biol.
 , 
2004
, vol. 
24
 (pg. 
4627
-
4635
)
13
Reamon-Buettner
S.M.
Ciribilli
Y.
Inga
A.
Borlak
J.
A loss-of-function mutation in the binding domain of HAND1 predicts hypoplasia of the human hearts
Hum. Mol. Genet.
 , 
2008
, vol. 
17
 (pg. 
1397
-
1405
)
14
Morin
S.
Pozzulo
G.
Robitaille
L.
Cross
J.
Nemer
M.
MEF2-dependent recruitment of the HAND1 transcription factor results in synergistic activation of target promoters
J. Biol. Chem.
 , 
2005
, vol. 
280
 (pg. 
32272
-
32278
)
15
Murre
C.
McCaw
P.S.
Baltimore
D.
A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins
Cell
 , 
1989
, vol. 
56
 (pg. 
777
-
783
)
16
van der Heyden
M.A.
Defize
L.H.
21 years of P19 cells: what an embryonal carcinoma cell line taught us about cardiomyocyte differentiation
Cardiovasc. Res.
 , 
2003
, vol. 
58
 (pg. 
292
-
302
)
17
Reamon-Buettner
S.M.
Hecker
H.
Spanel-Borowski
K.
Craatz
S.
Kuenzel
E.
Borlak
J.
Novel NKX2-5 mutations in diseased heart tissues of patients with cardiac malformations
Am. J. Pathol.
 , 
2004
, vol. 
164
 (pg. 
2117
-
2125
)
18
Reamon-Buettner
S.M.
Borlak
J.
Somatic NKX2-5 mutations as a novel mechanism of disease in complex congenital heart disease
J. Med. Genet.
 , 
2004
, vol. 
41
 (pg. 
684
-
690
)
19
Inga
A.
Reamon-Buettner
S.M.
Borlak
J.
Resnick
M.A.
Functional dissection of sequence-specific NKX2-5 DNA binding domain mutations associated with human heart septation defects using a yeast-based system
Hum. Mol. Genet.
 , 
2005
, vol. 
14
 (pg. 
1965
-
1975
)
20
Reamon-Buettner
S.M.
Borlak
J.
GATA4 zinc finger mutations as a molecular rationale for septation defects of the human heart
J. Med. Genet.
 , 
2005
, vol. 
42
 pg. 
e32
 
21
Pikkarainen
S.
Tokola
H.
Kerkela
R.
Ruskoaho
H.
GATA transcription factors in the developing and adult heart
Cardiovasc. Res.
 , 
2004
, vol. 
63
 (pg. 
196
-
207
)
22
Reamon-Buettner
S.M.
Borlak
J.
HEY2 mutations in malformed hearts
Hum. Mutat.
 , 
2006
, vol. 
27
 pg. 
118
 
23
Ishioka
C.
Frebourg
T.
Yan
Y.X.
Vidal
M.
Friend
S.H.
Schmidt
S.
Iggo
R.
Screening patients for heterozygous p53 mutations using a functional assay in yeast
Nat. Genet.
 , 
1993
, vol. 
5
 (pg. 
124
-
129
)
24
Gietz
R.D.
Schiestl
R.H.
Willems
A.R.
Woods
R.A.
Studies on the transformation of intact yeast cells by the LiAc/SS-DNA/PEG procedure
Yeast
 , 
1995
, vol. 
11
 (pg. 
355
-
360
)
25
Sherman
F.
Fink
G.R.
Hicks
J.B.
Methods in Yeast Genetics
 , 
1986
Cold Spring Harbor, NY
Cold Spring Harbor Laboratory
26
Storici
F.
Durham
C.L.
Gordenin
D.A.
Resnick
M.A.
Chromosomal site-specific double-strand breaks are efficiently targeted for repair by oligonucleotides in yeast
Proc. Natl Acad. Sci. USA
 , 
2003
, vol. 
100
 (pg. 
14994
-
14999
)
27
Tomso
D.J.
Inga
A.
Menendez
D.
Pittman
G.S.
Campbell
M.R.
Storici
F.
Bell
D.A.
Resnick
M.A.
Functionally distinct polymorphic sequences in the human genome that are targets for p53 transactivation
Proc. Natl Acad. Sci USA
 , 
2005
, vol. 
102
 (pg. 
6431
-
6436
)
28
Resnick
M.A.
Inga
A.
Functional mutants of the sequence-specific transcription factor p53 and implications for master genes of diversity
Proc. Natl Acad. Sci. USA
 , 
2003
, vol. 
100
 (pg. 
9934
-
9939
)

Author notes

The authors wish it to be known that, in their opinion, the last two authors should be regarded as joint Senior Authors.