Downregulation of N1 gene expression inhibits the initial heartbeating and heart development in axolotls

Tissue & Cell 36 (2004) 71–81

Downregulation of N1 gene expression inhibits the initial heartbeating and heart development in axolotls C. Zhang a , F. Meng b,1 , X.P. Huang a , R. Zajdel c , S.L. Lemanski a , D. Foster b , N. Erginel-Unaltuna d , D.K. Dube c,e , L.F. Lemanski a,∗ b

a Department of Biomedical Science, Florida Atlantic University, Boca Raton, FL 33431, USA Department of Medical Physiology, Texas A & M University System HSC, College Station, TX 77843, USA c Department of Cell and Development Biology, Upstate Medical University, Syracuse, NY 13210, USA d Department of Genetics, University of Istanbul, Istanbul, Turkey e Department of Medicine, Upstate Medical University, Syracuse, NY 13210, USA

Received 10 March 2003; received in revised form 29 September 2003; accepted 8 October 2003

Abstract Recessive mutant gene c in the axolotl results in a failure of affected embryos to develop contracting hearts. This abnormality can be corrected by treating the mutant heart with RNA isolated from normal anterior endoderm or from endoderm conditioned medium. A cDNA library was constructed from the total conditioned medium RNA using a random priming technique in a pcDNAII vector. We have previously identified a clone (designated as N1) from the constructed axolotl cDNA library, which has a unique nucleotide sequence. We have also discovered that the N1 gene product is related to heart development in the Mexican axolotl [Cell Mol. Biol. Res. 41 (1995) 117]. In the present studies, we further investigate the role of N1 on heartbeating and heart development in axolotls. N1 mRNA expression has been determined by using semi-quantitative RT–PCR with specifically designed primers. Normal embryonic hearts (at stages 30–31) have been transfected with anti-sense oligonucleotides against N1 to determine if downregulation of N1 gene expression has any effect on normal heart development. Our results show that cardiac N1 mRNA expression is partially blocked in the hearts transfected with anti-sense nucleotides and the downregulation of N1 gene expression results in a decrease of heartbeating in normal embryos, although the hearts remain alive as indicated by calcium spike movement throughout the hearts. Confocal microscopy data indicate some myofibril disorganization in the hearts transfected with the anti-sense N1 oligonucleotides. Interestingly, we also find that N1 gene expression is significantly decreased in the mutant axolotl hearts. Our results suggest that N1 is a novel gene in Mexican axolotls and it probably plays an important role in myofibrillogenesis and in the initiation of heartbeating during heart development. © 2003 Elsevier Ltd. All rights reserved. Keywords: Heart development; Novel gene; Anti-sense oligonucleotides; Myofibrils; Axolotl

1. Introduction The Mexican axolotl (Ambystoma mexicanum) provides an excellent model for studying heart development since it carries a cardiac lethal mutation in gene c. Homozygosity for the lethal mutant allele of gene c results in a failure of affected embryos to develop contracting hearts due to an absence of organized myofibrils. This abnormality can be corrected by co-culturing early mutant hearts with normal anterior endoderm, in medium “conditioned” ∗ Corresponding

author. Tel.: +1-561-297-0475; fax: +1-561-297-0422. E-mail address: [email protected] (L.F. Lemanski). 1 Present address: Department of Molecular and Cell Biology, University of Texas at Dallas, Richardson, TX, USA. 0040-8166/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.tice.2003.10.003

by normal tissue, or by RNA isolated from “conditioned” medium. Additionally, RNA isolated from normal anterior endoderm/mesoderm conditioned medium corrects the mutant hearts in a dose-dependent manner (Lemanski et al., 2001). A cDNA library was constructed previously in our laboratory by using the RNAs from conditioned medium. On the basis of sequence analysis on this cDNA library, it was estimated that 56% of the total RNA present in the conditioned medium are rRNAs, while 44% are non-ribosomal RNAs. One of the non-ribosomal RNAs that showed no significant homology with other known sequences in the Genebank was examined further. An RT–PCR analysis showed that this RNA (designated “N1”) was expressed in juvenile skeletal muscle, brain, and heart in significant amounts, less in the

72

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

lungs and not at all in the liver, kidney, and other organs (Erginel-Unaltuna et al., 1995). Anti-sense oligonucleotides have been used in vitro and in vivo as tools for selective expression blockade of inducible transcription factor genes, and developed as a promising approach to generate novel treatments to inhibit specific gene function in cancer, AIDS, cardiovascular, and other diseases, where gene sequences are known. The mechanisms by which anti-sense oligonucleotides inhibit expression of target genes are not completely understood, but they are thought to involve some interference with translation of the target mRNA and degradation of DNA–RNA hybrids by ribonuclease H (RNase H). In the present study, we have applied cationic liposomes to introduce the N1 anti-sense oligonucleotides into whole embryonic hearts at pre-heartbeat stages to determine whether the downregulation of N1 gene expression can adversely affect the initial heartbeat process and organized myofibril formation. Our results show that cardiac N1 mRNA expression is partially blocked in the hearts transfected with anti-sense nucleotides and the downregulation of N1 gene expression results in a decrease of heartbeating in normal embryos, although the hearts remain alive as indicated by calcium spike movement throughout the hearts. The results suggest that N1 is a novel gene in Mexican axolotls and it probably plays an important role in myofibrillogenesis and in the initiation of heartbeating during heart development. 2. Material and methods 2.1. Procurement and maintenance of axolotls All experiments were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals (National Academy Press, Washington, DC, USA, 1996) and were approved by the Animal Care and Use Committee of Texas A&M University and Florida Atlantic University. Normal and cardiac mutant axolotl embryos were obtained from matings between homozygous normal (+/+ × +/+) or heterozygous (+/c × +/c) animals from our colony at the Texas A&M University Health Science Center at College Station and at Florida Atlantic University. The animals were maintained in aquaria in diluted Holtfreter’s solution, fed commercial salmon pellets, and supplemented occasionally with raw beef liver and live earthworms. The embryos were staged according to the methods of Bordzilovskaya et al. (1989). 2.2. Whole-heart bioassay Embryos from stages 30 to 31, 34 to 35, and 38 to 39 were washed four times in filter-sterilized Steinberg’s buffered salt solution (58 mM NaCl, 0.67 mM KCl, 0.9 mM CaCl2 , 0.2 mM MgSO4 , 4.6 mM HEPES, pH 7.4), containing 1% antibiotic/antimycotic at a final concentration of penicillin G sodium 100 units/ml, 0.1 g/ml streptomycin sulfate, and 0.25 ␮g/ml amphotericin B (Gibco Life Technologies, Grand

Island, NY, USA). Embryos were killed after being anesthetized with MS-222 (Tricaine methanesulfonate), and the hearts were extirpated with heat-sterilized watchmaker’s forceps under a dissecting microscope and placed in 100 ␮l of Steinberg’s solution without antibiotics. 2.3. Molecular cloning of N1 gene The original partial sequence of N1 (223 bp) was obtained by sequencing the clone with the longest insert from the conditioned medium library, which could promote myofibrillogenesis. In this study, we have screened all the inserts in pcDNAII expression vectors (Invitrogen, Carlsbad, CA, USA) with the help of RT–PCR. 2 ␮g of the total embryonic RNA (stages 15–17) was used for the first strand generation using hexamer oligo dT and Thermoscript reverse transcriptase (Invitrogen). RNA and oligo dT were denatured at 65 ◦ C for 5 min in 10 ␮l of volume and placed on ice. Reactions were continued by adding 10 ␮l of cDNA synthesis buffer and elongated at 55 ◦ C for 10 min and then 50 ◦ C for another 50 min. DNA–RNA hybrids were digested by RNase H. 2 ␮l of the total reaction solution (20 ␮l) was taken for subsequent PCR reaction. The oligonucleotide primers were made in the Gene Technology Laboratory at Texas A&M University, College Station, TX, USA. N1-RT-5 : 5 -GGCATCTGGAGGTATCAATGTC-3 ; N1-RT-3 : 5 -AGAGCTAGTTGGCACACAGACC-3 . The PCR reactions were carried out using plantinum Taq according to the manufacture’s protocols (Invitrogen). The amplification was carried out for 35 cycles in a thermocycler. All amplified DNA were subjected to Southern blot hybridization with random-labeled DNA probes from original 223 bp N1 templates. The PCR amplified DNA were cloned into a pGEM-T easy vector (Promega, Madison, WI, USA) following the manufacturer’s instructions. Subsequent transformations, colony hybridizations to pick up the desired clones and preparation of DNA templates were performed. Fluorescence DNA sequencing was performed afterwards. After comparison of the two sequences, a deduced ∼200 amino acids partial coding sequence is located at the 5 -end of the N1-cDNA. In the meantime, the phage library was being plotting and filter lifts were being screening by DIG labeling and detecting Starter Kit II (Boehringer-Mannheim, Indianapolis, IN, USA). Long distance PCR (Herculase from Stratagene, California) was performed with an N1-specific primer and a vector-specific primer to amplify the genomic and cDNA sequences. All pre- and post-PCR analyses were done in separate rooms with different sets of equipments. 2.4. Cationic liposome transfection Phosphothioate-modified anti-sense oligonucleotides specific to N1-cDNA (5 -ACATTGATACCTCCAGATGCCATAGAG-3 ) were transfected into normal embryonic hearts at stages 30–31, 3 days before the initial heartbeat. Liposome reagents (0.1–0.16 ␮g/␮l) and oligonucleotides (5–10 ng/␮l)

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

containing either sense or anti-sense N1 sequence were diluted independently in Steinberg’s solution without antibiotics for 30–45 min at room temperature. The two solutions were then mixed for 15 min according to the manufacturer’s instructions (Lipofectin; Gibco Life Technologies). Drops of transfection solution (1 ml) were placed on tissue culture plates. Normal embryos from stages 31–32 were washed four times in filter-sterilized Steinberg’s buffered salt solution. Prior to the operation the embryos were anesthetized by MS-222. Hearts were extirpated and placed directly into the transfection solution. After 48 h in a 17 ◦ C incubator, the embryos were placed in fresh Steinberg’s solution without lipofectin and oligonucleotides and cultured for an additional 5 days. The viability of the hearts after treatment of the oligonucleotides were confirmed by detection of calcium spikes in the cardiac cells. Calcium indicator Fluro-4 (F-14201) obtained from Molecular Probes, Eugene, OR, USA, was dissolved in DMSO at a concentration of 5 mM and stored at −20 ◦ C. Right before the experiments, the Fluro-4 solution was diluted to 5 ␮M with 75% Holtfreter’s solution. The normal or mutant heats after the treatment with the oligonucleotides were incubated with the diluted Fluro-4 solution for 1 h, then transferred to calcium-free Holtfreter’s solution (75%) before microscope examination. The whole-heart calcium spikes were observed under an Olympus inverted fluorescence microscope with a GFP filter set. Digital images were recorded with an attached QIMAGING RETIGA-1300 CCD monochrome camera. 2.5. RNA and cDNA preparation The method of isolation of RNA was based on a rapid one-step procedure of Chomczynski and Sacchi (1987) using a total RNA/mRNA isolation kit RNA STAT-60 (TEL-TEST, INC., Friendwood, TX, USA). Total RNA was extracted from freshly obtained embryonic hearts by using the phenol/chloroform procedure. The sample obtained was quantified by absorbance at 260 nm. RNA integrity was assessed by electrophoresis on a 1.8% agarose gel. cDNA was synthesized from 0.5 to 1 ␮g of total RNA using 0.2 U of murine leukemia virus reverse transcriptase (Promega), 2.5 ␮M random hexamers (Boehringer-Mannheim), 1 mM dNTP (Boehringer-Mannheim), and 2 U RNase inhibitor (Promega) in a total volume of 20 ␮l. The reaction was performed at 37 ◦ C for 60 min. 2.6. Semi-quantitative PCR The procedure of PCR was based on Barnes technique. The Reverse Transcriptase and Expand High Fidelity PCR System were ordered from Boehringer-Mannheim. RT–PCR was performed using 2 ␮g aliquots of total cellular RNA, as previously described (Normanno and Ciardiello, 1993). Serial dilutions of this reaction were subsequently used for PCR amplification using specific primers based on the partial sequence of the N1 gene. PCR was performed for

73

30 cycles. All PCR results were repeated three times. The PCR products were analyzed by electrophoresis on 1.5% agarose gels containing ethidium bromide (0.5 ␮g/ml) and photographed under ultraviolet light. The images were further analyzed by scanning densitometry (ChemilmagerTM 4400 System with AlphaEaseTM Software, Alpha Innotech Co.). The relative amounts of N1 and ␤-actin mRNA in anti-sense and sense oligo-treated hearts were assessed. Several dilutions of cDNA were assayed for each test group, to reach non-saturating concentrations of amplification product in the RT–PCR reaction. The average ratio between the transcript levels of stages 30–31, 35–36, and 38–39 group hearts in anti-sense versus sense hearts was assessed, following normalization by ␤-actin levels. PCR primers (Gene Technology Lab, Texas A&M University) were as follows: N1, 5 -CTC TAT GGC ATC TGG AGG TAT CAA TGT-3 ; 3 -TTC GAA ACC GAA GGT ATA ACC AAT CTA-5 ; ␤-actin, 5 -CGA GGC CCA GAG CAA GAG A-3 ; 3 -CGT GAC ATT AAG GAG AAG CTG TG-5 . All RT–PCR products were confirmed by Southern blotting assays with a radioactive-labeled probe specific for N1 gene (the sequence of the probe: 5 -GAA GCA GCG ACA GCC CAT CAC TGC ATA CAC AAG CAA GAA GAT-3 ). 2.7. Confocal microscopy The cardiac tissues were fixed according to a previously described method (Bell et al., 1987). All steps took place at room temperature with gentle agitation on an orbital shaker. Hearts were placed into 1 mM dithiosulfylpropionate (Sigma, St. Louis, MO, USA) dissolved with 1% dimethylsulfoxide in Steinberg’s solution for 15 min. Then, the hearts were further permeabilized by 0.5% Nonidet P-40 in Steinberg’s solution for 15 min. The reaction was quenched by two 10-min washes in 0.1 M glycine. After blocking with 3% BSA in Steinberg’s solution containing 0.05% Tween-20 for 1 h, the hearts were incubated in purified monoclonal anti-tropomyosin antibody (Lin et al., 1985) diluted to 1:75 in Steinberg’s solution for 1 h, and washed several times. Hearts were then incubated in Oregon green-conjugated goat anti-mouse secondary antibody (Molecular Probes) at a 1:75 dilution for 45 min. The hearts were rinsed in several changes of BSA in Steinberg’s solution and then post-fixed in 2% paraformaldehyde for 30 min. The fixation was quenched by 0.1 M glycine for 30 min and then placed in Steinberg’s solution. The hearts were mounted on slides in 50% glycerol/phosphate buffered saline containing 2% n-propyl gallate. Three layers of fingernail polish were used to support the glass cover slips to prevent crushing of the whole hearts. The specimens were viewed on a Meridian Instruments Ultima Z Confocal Laser Scanning Microscope. Fluorescence was excited at 488 nm. A 30-section z-series was made for each sample and system software was used to display the final image on a monitor. Color photographs of these images were then printed on Ilfochrome Classic Deluxe CLM 1k paper.

74

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

3. Results 3.1. The partial sequence of the N1 gene The extended nucleotide sequence of the N1 gene is shown in Fig. 1. Within the sequence of the total 709 bp, there is a 3 -end coding region (506 bp in Fig. 1) and a 200 bp 3 -untranslated region without polyadenylation sequence. The sequence has been examined for its homology with other known genes in the databases available to the Genebank using BLAST search. No significant homology was found, suggesting that it is a novel gene in Mexican axolotls. However, the extended N1 is still not the full sequence of the N1 gene, which misses the 5 -end of the coding region. The 5 -RACE experiments and the library screening are still on going in our laboratory. Moreover, our studies have shown that the original N1 sequence (223 bp) is a critical part for the gene, which shows bioactive properties. Therefore, in most of our experiments, we still applied the 223 bp N1 fragment for the studies. 3.2. The expression pattern of N1 in the heart during heart development Total RNAs were isolated from normal hearts at stages 27–28 (initial heart-forming stage), 30–31, and 32–33 (pre-heartbeat stages), 35–36 (initial heartbeat stage), and 38–39 (well-formed heartbeat stage). By using reverse transcription and PCR amplification using primer pairs specifically designed from the extended nucleotide sequence of the N1-cDNA as described in Section 2, we have found that

1 51 101 151 201 251 301 351 401 451 501 551 601 651 701

N1 expression is significantly increased in normal hearts after the initial heartbeat stage (stages 35–36) and reaches a plateau in stages 35–39 (Fig. 2). However, the expression of N1 in mutant hearts is significantly reduced compared with the normal hearts at the same stage (Fig. 3). 3.3. Anti-sense oligonucleotide inhibits the N1 expression in the heart Normal axolotl hearts at stages 30–31 can continuously develop in vitro from pre-heartbeat stage to heartbeat stage. By using liposome as a mediator, we have successfully delivered N1 anti-sense oligonucleotides into the normal hearts at stages 30–31 and maintained those hearts in culture for 4 days. As shown in Table 1, the results have shown that 48 out of 50 normal axolotl hearts have developed into beating hearts after 4-day culture (Table 1). N1 anti-sense nucleotide treatment resulted in a significant decrease of beating hearts in the normal axolotl hearts after 4-day incubation. N1 sense oligonucleotide-transfected hearts have not shown any significant decrease in beating heart rate (46 beating hearts out of 54 sense nucleotide-treated hearts) (Table 1). Fig. 4a shows a whole heart after lipofectin-mediated delivery of the anti-sense oligonucleotide labeled with FITC. The staining is located throughout the heart. The labeled oligonucleotides have entered the myocytes. The confocal micrograph of control heart treated with lipofectin and FITC solution alone shows a little non-specific fluorescent staining in the heart (Fig. 4b). The introduction of N1 anti-sense nucleotides has resulted in reduced N1 mRNA levels in normal hearts at stages

Primer-1 (+) GCAGGGGAAT TTCACAGACC ATTTCTACTG CCCCCATGC C TCTATGGCAT CTGGAGGTAT CAATGTCCAA AACTCCAATG TAGCACTGAT TCTAGATGAT Detector TTGATGGGGT CTGACGCCCA GAAGCAGCG A CAGCCCATCA CTGCATACAC AAGCAAGAAG ATTCCCGGGT GGGGAAGATG TAAATGCTCT CTCTACAATG Primer-2 (-) ATCTTGCCCC CACAGTTACA GTTCTCCCAA GGCCATCTAA CCAATATGGA AGCCAAAGCT TGCTAATCTC CACAGTCATG CATACACATC ACAAGGCAAT ACCAGGACTC AGGGGTAAAA CAAACCCAAG AATTTTCCTC TTCGGCAATG AAGGGAGTCA CAGGTGGTTG TCTAATGCCC CTCCAAAACA GCAAAACCAG TCCAAAGCAC CAGTGCACTG CATGTTTGGA TTGGCTTCTA ATCATGGGAA TCCGACAAGC TCAGGCACCA AGGGTGTACT CCCTCCTCTA AAGGTATTGT CTCTGTAGAG ACCTCTGGAA CCTATGGCAG GAGAGAAGCT GTCCGAGAGG GTGGACAAAC CAAGCCAAGT ATAACTTCAG GAGGAACACC ACTTCCCGGA GCACCCAGCT TCTCCAGTTT GACCCATAAC TGGGGACATT ATACCCATGC TAATAATCAA TGAGAATTCC TAGAAAGCTG GGTCTGTGTG CCAACTAGCT CTACATTAT

Fig. 1. The extended nucleotide sequence of N1-cDNA. Within the 709 bp total sequence, there is a 3 -end of N1 coding region (the stop codon TAG is in bold). The sequence of the primer pair used for RT–PCR analysis is indicated as “Primer-1 (+)” and “Primer-2 (−)”. The sequence used as a probe for Southern blotting is indicated as “Detector”.

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

75

Fig. 2. N1 expression pattern in axolotl hearts. N1 expression in normal embryonic hearts from initial heart-forming stage (stages 27–28) to well-formed heartbeat stage (stages 38–39) has been detected by RT–PCR. Band intensities were quantified by a laser densitometer and were corrected for the intensity of ␤-actin mRNA, a housekeeping gene. The ratio of the intensity of N1 mRNA band over the intensity of ␤-actin mRNA at stages 32–33 was arbitrarily designated as 1.0. N1 expression increased significantly after the initial heartbeat stage (stage 35).

Fig. 3. N1 expression in normal and mutant axolotl hearts. The total RNA has been isolated from the normal and mutant axolotl hearts and RT–PCR is carried out as described in Section 2. The ethidium bromide-stained bands were confirmed by Southern blotting assays with an N1-specific probe.

30–31, 35–36, and 38–39 (Fig. 5). Levels of N1 mRNA in the hearts are determined by RT–PCR analysis and are normalized by ␤-actin values, which showed a similar expression level in samples from different heart developmental stages (Fig. 5). Anti-sense oligonucleotide-transfected hearts showed a significant reduction in the levels of N1 Table 1 Inhibition of initial heartbeat in normal axolotl hearts by N1 anti-sense oligonucleotides Treatment

Rate of beating hearts (%) (beating hearts/total hearts treated)

N1 anti-sense oligo-treated group N1 sense-treated oligo-treated group No treatment group (Steinberg’s solution)

22 (11/51)∗ 85 (46/54) 98 (48/50)

∗

P < 0.01.

mRNA after the initial heartbeat stage. In particular, 43 and 61% reductions of, respectively, N1 mRNA levels at initial heartbeat stage and well-formed heartbeat stage were observed in anti-sense-transfected normal hearts when compared with control untreated hearts. Introduction of sense oligonucleotides into the normal hearts at the same stages did not significantly affect N1 mRNA expression in the hearts (Fig. 5). The expression was still increased sharply after the initial heartbeat stage (Fig. 5). 3.4. Anti-sense oligonucleotide inhibits the heartbeating and myofibrillogenesis in the heart Normal axolotl hearts can continuously develop in vitro from pre-heartbeat stage to heartbeat stage. In general, the normal hearts from axolotl at stages 30–31 develop into a beating heart after 3- to 4-day incubation in vitro. The

76

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

Fig. 4. Liposome-mediated delivery of N1 oligonucleotides into normal axolotl hearts. (a) Normal heart at stages 36–37, 4 days after transfection with lipofectin and FITC-labeled oligonucleotides. The FITC staining is located throughout the heart, indicating that the labeled oligonucleotides have entered into the myocytes. (b) Confocal micrograph of control heart treated with lipofectin and FITC solution only. Little specific labeling is found.

results have shown that 48 out of 50 normal axolotl hearts have developed into beating hearts after 4 days in culture (98% beating rate). N1 anti-sense nucleotide treatment has resulted in a significant decrease of beating hearts in normal axolotl hearts after 4 days incubation (11 beating hearts out of 51 treated hearts, 22% beating rate). N1 sense oligonucleotide-transfected hearts has not shown any significant decrease in beating hearts (46 beating hearts out of 54 sense nucleotide-treated hearts, 85% beating rate). We further examined myofibril structure under a con-

focal microscope in those heart samples decorated with anti-tropomyosin antibody. The results have shown that the normal hearts at stages 37–38 have shown the organized myofibril structure inside of myocytes or in the peripheral circumference of the cells. Moreover the sarcomeric myofibrils have been observed throughout the hearts (Fig. 6a). However, transfection of N1 anti-sense oligonucleotides into normal hearts drastically disrupted the organization of the myofibrils within the myocytes. Tropomyosin localization in those hearts is primarily in amorphous collections

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

77

Fig. 5. Effect of anti-sense oligonucleotide treatment on N1 expression in normal axolotl hearts. N1 expression levels were assessed by semi-quantitative RT–PCR as described in Section 2. The levels of the specific transcripts in treated and untreated hearts were normalized to ␤-actin content. Data represent the average of the ratio between densitometric values of several corresponding dilutions of cDNA from treated and untreated hearts. The values are summarized from three separate experiments.

throughout the hearts (Fig. 6b). Furthermore, the transfection of sense N1 oligonucleotide causes very little disruption of sarcomeric myofibril formation in the hearts (Fig. 6c). The results of calcium spike detection confirm that the

oligonucleotide-treated hearts, either the normal or mutant hearts, remain alive and have the normal electrophysiological properties indicted by the calcium spike movement throughout the whole heart (Fig. 7).

78

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

4. Discussion In 1968, Dr. Humphrey first reported the discovery of a recessive mutant gene in a dark stock of axolotls, A. mexicanum, imported from Mexico (Humphrey, 1972). The effect of homozygosity for the mutant allele is the absence of a heartbeat, even though initially heart development appears normal. Cardiac mutant hearts of Mexican axolotls have been characterized by a severe reduction in tropomyosin, a myofibrillar protein present in the thin filament, and a lack of organized myofibrils. Our previous biochemical studies have confirmed the reduction of tropomyosin in mutated hearts (Moore and Lemanski, 1982; Starr et al., 1989). The tropomyosin deficiency is believed to be a secondary effect of the mutation. However, mutant hearts can be rescued by co-culturing them with anterior endoderm/mesoderm tissue, or by culturing them in a medium “conditioned” by this normal tissue (Lemanski et al., 2001). There are three factors involved in regulating heart formation and differentiation. The principal inductor tissue is the anterior endoderm. Secondly, a general stimulatory effect of the epidermis increases the frequency of heart development. The third factor is the inhibitory effect of the neural plate tissue. The N1 gene is mainly expressed in the heart and in the two major inductive tissues: epidermis and endoderm; N1 is virtually absent in tissues that are inhibitory to cardiogenesis, i.e. neural plate and brain tissues in embryos (Erginel-Unaltuna et al., 1995). The clone which contains the insert for the N1 gene is selected from among the various clones of the cDNA library that were hybridized with 32 P-labeled conditioned RNA. This initial screening of the library with the labeled conditioned medium RNA was done to select the most abundant species of RNA present in the conditioned medium. A majority of the clones we obtained contained inserts that belong to ribosomal RNAs. N1 was derived from one of the remaining clones, which was nonribosomal. Thus, it is likely that the RNA belonging to the N1 gene is one of the most abundant RNA species in the conditioned medium. N1 clone was initially obtained from RNAs originating from the normal anterior endoderm regions in the embryos at tailbud stage. The gene is expressed mainly in the heart or “heart inductive” tissues. Moreover, a significant decrease of N1 mRNA expression has been observed in cardiac mutant axolotl hearts. All these make it tempting to speculate that the gene might be involved in heart induction and the initial heartbeat process. In the present studies, we have used FITC-tagged oligonucleotides for transfection assays in the whole hearts. Transfection of anti-sense oligonucleotides have been used

Fig. 6. Three-dimensional confocal microscopy images of axolotl hearts decorated with CH1 anti-tropomyosin antibody. (a) Normal hearts at stages 36–37 with no treatment. (b) Normal hearts at stages 36–37, 4 days after anti-sense oligonucleotide transfection. (c) Normal hearts at stages 36–37, 4 days after sense nucleotide transfection.

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81 Fig. 7. A single cycle of the whole-heart calcium spikes. The whole-heart calcium spikes are measured as described in Section 2. The whole process of the calcium spike movement from the atrium to the ventricle takes about 1.5 s (the images from the left to the right). The calcium spike transmittance remains at a constant speed throughout the whole heart. The quiescent heart remains dark with only a few cells showing fluorescence (indicated by the arrows, left panel 1). The calcium spike starts from the tip of the atrium (A) (potential sino-atrium node, left panel 2, arrow) and transmitted through the atrium to the ventricle (V) (left panel 3, the dashed arrows) and also from outward to inward. The spikes continually transmit through ventricle (left panel 4) to the outflow tract (O) (left panel 5) with the brightest signal emitted.

79

80

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81

to alter expression of a specific gene. For example, an anti-sense oligonucleotide transfected into skeletal muscle was creatively used to skip a mutant exon and create a novel in-frame dystrophin transcript (Dunckely et al., 1998). Many hypotheses have been proposed to explain the mechanisms by which anti-sense oligonucleotides inhibit gene expression. Oligonucleotides targeting the AUG region are generally believed to act through a “translation arrest” mechanism, in which the oligonucleotide masks the ribosome-binding site and prevents the formation of translation complex (Neckers et al., 1992). However, RNase H-mediated cleavage is considered the major mechanism of action of the anti-sense oligonucleotides targeting internal coding regions (Carter and Lemosine, 1993; Duroux et al., 1995; Reichert et al., 2000). RNase H probably recognizes the mRNA–oligonucleotide complex as a substrate and cleaves the target mRNA strand. Our RT–PCR results support this mechanism, but the exact process remains to be elucidated. Recent studies have confirmed that it is practical to use cationic liposomes for transferring fluorescein-labeled oligonucleotide into the target cells (Noguchi et al., 1998). In addition, it has been observed that cationic liposomes can stabilize oligonucleotides within cardiac tissues (Aoki et al., 1997). Fluorescence labeling with oligonucleotides was visible in the nucleus of the cells for at least 1 week after transfection, whereas the FITC dye alone shows a little of fluorescence 1 day after transfection. In our experimental system, FITC-tagged anti-sense oligonucleotides have been successfully introduced into the heart cells throughout the whole heart. Kinetic examination of the oligonucleotides in the cells 1 day after transfection has shown that fluorescein staining is presented in the hearts in a diffused pattern, but 4 days after transfection, the fluorescence staining had been observed in the nuclear area. Transfection of N1 anti-sense oligonucleotide into normal hearts at stages 30–31 resulted in a disorganization of the myofibrils and stoppage of the initial heartbeat in a statistically significant number of hearts 4 days after transfection. Inhibition of N1 mRNA at the same time by this transfection suggests a possible pathway from N1 gene expression, myofibril formation to initial heartbeat development. By using lipofectin-mediated transfection assays, we have introduced several labeled proteins, RNAs and DNAs into the normal and mutant axolotl hearts to study the function of these molecules. The calcium spike data confirm that the hearts transfected with oligonucleotides for several days in vitro in the presence of lipofectin remain alive. This technique provides us a useful tool for investigating functional genomics, contractile proteins in heart development (Zajdel et al., 1998, 1999; Meng et al., 2003). Our present results indicate that the expression of N1 gene is essential for normal heart development and that the alteration of N1 gene expression results in a specific abnormality during heart development which leads to a failure in myofibril formation and a decrease of heartbeat rate in axolotls. Since it is difficult to create a gene knockout model in amphibian systems,

the application of anti-sense oligonucleotide technique is able to block a specific gene expression and provides us a useful tool for investigating the gene expression and its function during heart development in axolotls.

Acknowledgements This study was supported by NIH Grants HL-58435 and HL-061246 and by a Christine E. Lynn American Heart Association grant-in-aid to L.F.L.

References Aoki, M., Morishita, R., Higaki, J., Moriguchi, A., Kida, I., Hayashi, S., Matsushita, H., Kaneda, Y., 1997. In vivo transfer efficiency of anti-sense oligonucleotides into the myocardium using HVJliposomes method. Biochem. Biophys. Res. Commun. 231, 540– 545. Bell, P.B., Rundquist, I., Svensson, I., Collins, V.P., 1987. Use of cytofluorotomy to evaluate binding of antibodies to the cytoskeleton of cultured cells. J. Histochem. Cytochem. 35, 1381–1388. Bordzilovskaya, N.P., Detlaff, T.A., Duhon, S.T., Malacinski, G.M., 1989. Developmental stage series of axolotl embryos. In: Armstrong, J.B., Malacinski, G.M. (Eds.), Developmental Biology of the Axolotl. Oxford Press, Oxford, pp. 201–219. Carter, G., Lemosine, N.R., 1993. Anti-sense technology for cancer therapy: does it make sense? Br. J. Cancer 67, 869–876. Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolation by acid guanidinum thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162, 156–159. Dunckely, M.G., Manoharan, M., Villiet, T., Eperon, I.C., Dickson, G., 1998. Modification of splicing in the dystrophin gene in cultured Mdx muscle cells by antisense oligoribonucleotides. Hum. Mol. Genet. 7, 1083–1090. Duroux, I., Godard, G., Boidot-Forget, M., Schwab, G., Helene, C., Saison-Behmoaras, T., 1995. Rational design of point mutationselective anti-sense DNA targeted to codon 12 of Ha-rasmRNA in human cells. Nucleic Acids Res. 23, 3411–3418. Erginel-Unaltuna, N., Dube, D.K., Salsbury, K., Lemanski, L.F., 1995. Confocal microscopy of a newly identified protein associated with heart development in the Mexican axolotl. Cell Mol. Biol. Res. 41, 117–130. Humphrey, R.R., 1972. Genetic and experimental studies on a mutant gene c determining absence of heart action in embryos of the Mexican axolotl (Ambystoma mexicanum). Dev. Biol. 27, 365–375. Lemanski, F.L., Huang, X.P., Zajdel, R.W., Lemanski, S.L., Zhang, C., Meng, F., Foster, D., Li, Q., Dube, D.K., 2001. Cellular, molecular, and developmental studies on heart development in normal and cardiac mutant axolotls, Ambystoma mexicanum. In: Dube, D. (Ed.), Myofibrillogenesis. pp. 207–222. Lin, J.J.C., Chou, C.S., Lin, J.L.C., 1985. Monoclonal antibodies against chicken tropomyosin isoforms: production, characterization, and application. Hybridoma 4, 223–242. Meng, F., Huang, X.P., Zajdel, R.W., Foster, D., Dawson, N., Lemanski, S.L., Zawiejia, D., Dube, D.K., Lemanski, L.F., 2003. Relationship between cardiac protein tyrosine phosphorylation and myofibrillogenesis during axolotl heart development. Tissue Cell 35, 133– 142. Moore, P.B., Lemanski, L.F., 1982. Quantitation of tropomyosin by radioimmunoassay in developing hearts of cardiac mutant axolotls, Ambystoma mexicanum. J. Muscle Res. Cell Motil. 3, 161–167.

C. Zhang et al. / Tissue & Cell 36 (2004) 71–81 Neckers, L., Whitesell, L., Rosolen, A., Geselowitz, D.A., 1992. Antisense inhibition of oncogene expression. Crit. Rev. Oncog. 3, 175–231. Noguchi, A., Furuno, T., Kawaura, C., Nakanishi, M., 1998. Membrane fusion plays an important role in gene transfection mediated by cationic liposomes. FEBS Lett. 433, 169–173. Normanno, N., Ciardiello, F., 1993. Expression of amphiregulin, cripto1, and heregulin ␣ inb human breast cancer cells. Int. J. Oncol. 2, 903–911. Reichert, T.E., Nagashima, S., Kashii, Y., Stanson, J., Gao, G., Dou, Q.P., Whiteside, T.L., 2000. Interleukin-2 expression in human carcinoma cell lines and its role in cell cycle progression. Oncogene 19, 514–525.

81

Starr, C., Diaz, J.G., Lemanski, L.F., 1989. Analysis of actin and tropomyosin in hearts of cardiac mutant axolotls by two-dimensional gel electrophoresis, Western blots, and immunofluorescent microscopy. J. Morphol. 201, 1–10. Zajdel, R.W., Mclean, M., Lemanski, S.L., Muthuchamy, M., Wieczorek, D., Lemanski, L.F., Dube, D.K., 1998. Ectopic expression of tropomyosin promotes myofibrillogenesis in mutant axolotl hearts. Dev. Dyn. 213, 412–420. Zajdel, R.W., Dube, D.K., Lemanski, L.F., 1999. The cardiac mutant Mexican axolotl is a unique animal model for evaluation of cardiac myofibrillogenesis. Exp. Cell Res. 248, 557–566.