Embryonic Senescence and Laminopathies in a Progeroid Zebrafish Model

Background Mutations that disrupt the conversion of prelamin A to mature lamin A cause the rare genetic disorder Hutchinson-Gilford progeria syndrome and a group of laminopathies. Our understanding of how A-type lamins function in vivo during early vertebrate development through aging remains limited, and would benefit from a suitable experimental model. The zebrafish has proven to be a tractable model organism for studying both development and aging at the molecular genetic level. Zebrafish show an array of senescence symptoms resembling those in humans, which can be targeted to specific aging pathways conserved in vertebrates. However, no zebrafish models bearing human premature senescence currently exist. Principal Findings We describe the induction of embryonic senescence and laminopathies in zebrafish harboring disturbed expressions of the lamin A gene (LMNA). Impairments in these fish arise in the skin, muscle and adipose tissue, and sometimes in the cartilage. Reduced function of lamin A/C by translational blocking of the LMNA gene induced apoptosis, cell-cycle arrest, and craniofacial abnormalities/cartilage defects. By contrast, induced cryptic splicing of LMNA, which generates the deletion of 8 amino acid residues lamin A (zlamin A-Δ8), showed embryonic senescence and S-phase accumulation/arrest. Interestingly, the abnormal muscle and lipodystrophic phenotypes were common in both cases. Hence, both decrease-of-function of lamin A/C and gain-of-function of aberrant lamin A protein induced laminopathies that are associated with mesenchymal cell lineages during zebrafish early development. Visualization of individual cells expressing zebrafish progerin (zProgerin/zlamin A-Δ37) fused to green fluorescent protein further revealed misshapen nuclear membrane. A farnesyltransferase inhibitor reduced these nuclear abnormalities and significantly prevented embryonic senescence and muscle fiber damage induced by zProgerin. Importantly, the adult Progerin fish survived and remained fertile with relatively mild phenotypes only, but had shortened lifespan with obvious distortion of body shape. Conclusion We generated new zebrafish models for a human premature aging disorder, and further demonstrated the utility for studying laminopathies. Premature aging could also be modeled in zebrafish embryos. This genetic model may thus provide a new platform for future drug screening as well as genetic analyses aimed at identifying modifier genes that influence not only progeria and laminopathies but also other age-associated human diseases common in vertebrates.


Introduction
The LMNA gene encodes A-type lamins, lamin A and lamin C. The lamins are structural protein components of the nuclear lamina and constitute a class of intermediate filaments, which form a mesh network underlying the inner nuclear membrane that determines nuclear shape and size [1] [2] [3] [4]. Mutations in genes encoding the nuclear lamins and associated proteins cause a wide spectrum of human diseases sometimes called ''laminopa-thies.'' Diseases caused by mutations in LMNA encoding the Atype lamins include more than 10 different clinical syndromes [5]. Major types include diseases of the striated and cardiac muscle, lipodystrophy syndromes, peripheral neuropathies, and premature aging. While mutations and clinical phenotypes of laminopathies have been carefully described, explanations of the underlying pathogenic mechanisms are still emerging. One key question is: how do numbers of different mutations in this single gene produce such phenotypic heterogeneity and so many different diseases.
Heterozygous point mutations in lamin A cause Hutchinson-Gilford progeria syndrome (HGPS) [6] [7]. Most cases of HGPS harbor an identical de novo single-base substitution resulting in a silent Gly-to-Gly change at codon 608 within exon 11. This substitution creates an exonic consensus splice donor site/sequence and activates cryptic splicing, leading to the deletion of 50 codons at the end of prelamin A, called ''progerin'' [8] [9] [10]. The lamin A is synthesized as the precursor molecule prelamin A [11], and the maturation of lamin A involves the removal of 15 residues from the C terminus cleaved by the endoprotease, Zmpste24/FACE1, via isoprenylation and farnesylation involving a C-terminal CAAX (cysteine-aliphatic-aliphatic-any amino acid) box [12]. Progerin retains the CAAX box farnesylated at its C terminus, but lacks the site for endoproteolytic cleavage, and accumulates at the nuclear envelope, causing misshapen nuclei [13]. Importantly, sporadic use of the cryptic splice site in LMNA exon 11 has also been demonstrated during the normal human aging [14].
Notably, lethal neonatal laminopathies have now been identified and are caused by some specific mutations in the LMNA and ZMPSTE24/FACE-1 genes, suggesting that certain developmental abnormalities may be involved in some cases [15] [16] [17]. Whilst the crucial involvement of lamin A and lamina-associated polypeptide (LAP) 2a in the regulation of stem cells has been suggested previously [18] [19] [20] [21] [22] [23] [24], and several developmental studies have been performed in model animals including Drosophila, Xenopus, chicken, cow and mouse [25] [26] [27] [28] [29] [30], the essential roles for this gene and its requirement during early vertebrate development are still poorly understood in vivo. In addition, several gene perturbations identified by microarray profiling have also now been confirmed in HGPS fibroblasts [31] [32] [33] [34], but very few have been validated extensively in animal models.
Previous studies of HGPS fibroblasts have shown that farnesyltransferase inhibitors (FTIs) reduce the nuclear abnormalities associated with prelamin A accumulation and improve the disease phenotypes in mouse models [35]. More recently, it has been reported that a combined drug treatment with statins and aminobisphosphonates, which efficiently inhibits both farnesylation and geranylgeranylation of progerin and prelamin A, markedly improves the aging-like phenotypes and lifespan of mice deficient in Zmpste24 [36]. These findings suggest that the compounds may provide new therapeutic approaches to the treatment of this devastating progeroid syndrome. However, given the evidence of persistent disease in mice expressing a nonfarnesylated version of progerin, FTIs may not represent a ''panacea'' for this disease [37] [38]. Hence, new therapeutic interventions need to be considered, and new vertebrate model systems that facilitate high-throughput screening of chemical compounds to treat HGPS and other laminopathies are also essential. The zebrafish (Danio rerio) system has the potential to emerge as a key vertebrate model in this regard, as zebrafish maintain evolutionally conserved A-type (A and C) as well as Btype (B1 and B2) lamins among vertebrates.
We here describe that two different types of morpholino antisense oligonucleotides (MOs) directed against the zebrafish LMNA gene. The first of these MOs (LMNA-MO1) introduces a knockdown of zebrafish lamin A/C, and the other MO (LMNA-MO2 or -MO2.1) induced an unique cryptic splicing in exon 11 to generate a new truncated form (zebrafish lamin A-D8) in the globular domain of zebrafish lamin A. These MOs caused specific disruptions in the development of zebrafish embryos, particularly in the mesenchymal cell lineages, which are accompanied by either excessive apoptosis or embryonic senescence. We further demonstrate that an FTI can efficiently inhibit the farnesylation of zebrafish progerin (zebrafish lamin A-D37/zProgerin) and suppress the abnormal nuclear morphologies, muscle fiber defects and embryonic senescence phenotypes. Accelerated aging-like symptoms and shortened longevity were also observed in zebrafish expressing progerin/lamin A-D37 or even prelamin A. Our genetic model using zebrafish may thus facilitate a highthroughput approach to the search for novel therapeutics for human progeroid syndromes and laminopathies associated with nuclear-envelope abnormalities.

Transgenic Expression of Zebrafish Progerin and Prelamin A
The zebrafish LMNA gene was assigned to Linkage Group 16 (LG16) by radiation hybrid (RH) mapping of the LN54 RH panel [39]. The zebrafish LMNA region on LG16 is syntenic with human chromosome 1q21-23, which includes the genes adar (adenosine deaminase, RNA-specific), mef2d (MADS box transcription enhancer factor 2), ctsk (cathepsin K), cct3 (chaperonin containing TCP1, subunit 3 (gamma)), and hfe2 (hemochromatosis type 2) in addition to LMNA. This synteny strengthened the validity of the LMNA gene location based on RH mapping and physical linkage, and strengthened the identity of zebrafish LMNA as an ortholog of human LMNA. Upon the molecular cloning of zebrafish LMNA cDNA (Fig. S1), as well as conserved syntenic mapping analysis between the human and zebrafish LMNA genes (Fig. 1A), we developed zebrafish models for HGPS and laminopathies. Reverse transcriptase-PCR (RT-PCR) analysis of adult zlamin A/C revealed ubiquitous expression throughout the tissues, but at higher levels in the brain, heart, liver and muscle (Fig. 1B). We also examined the protein expression of zlamin A/C in adult tissues by western blotting analysis and found its ubiquitous expression as anticipated, although zlamin A was more strongly expressed in heart and higher levels of zlamin C were evident in brain (Fig. 1B, lower panels).
As HGPS is a dominant sporadic disease caused by mutations within exon 11 of LMNA, leading to an in frame deletion of 50 amino acids from the C-terminal tail of prelamin A (Progerin/ lamin A-D50), we first attempted to generate transgenic zebrafish expressing a form of zebrafish lamin A with a deletion of 37 amino acids i.e. a zlamin A-D37 (zebrafish Progerin: zProgerin) expression construct. This was to closely mimic the effects of human Progerin/lamin A-D50 based on the alignment of their amino acid sequence (see Fig. 1C). We microinjected wild-type zebrafish embryos with a zProgerin expression construct driven by the human cytomegalovirus (CMV) promoter at the one-cell stage of development for the generation of founder (F 0 ) fish that showed mosaic expression of the transgene [40]. In contrast to healthy non-transgenic wild-type (wt) siblings (Fig. 1D, upper left picture panel), of the 21 germline-positive F 0 zebrafish injected with enhanced green fluorescent protein (EGFP)-zProgerin, 19 (90%) were symptomatic for early spinal curvature, wasting symptoms, and odd swimming behaviors in young-to mid-adults (Fig. 1D, lower left and upper right picture panels), as were 3 (25%) of 12 germline-positive fish injected with EGFPzebrafish prelamin A (zPrelamin A) (Fig. 1D, lower right picture panel). The affected transgenic zebrafish developed distended curvatures ranging from a mild bend or arch to a mild or moderate distortion, which are similar to manifestations of senescence reported previously [41]. Some fish had protruding ventral or dorsal curvatures with relatively small pectoral fins, whereas others had wavy trunks throughout (Fig. 1D). Many of these fish also showed extensively unbalanced body shapes with age, resulting subsequently in the onset of abnormal swimming behaviors in their later lives. The mean latencies of the phenotype development in zprogerin and zprelamin A fish groups were 332 (range of 221 to 444 days) and 352 days (range of 306 to 431 days), respectively. All of the germline-positive F 0 zprogerin fish died before reaching 2 years of age. On the other hand, nonsymptomatic F 0 zprelamin A fish survived past 3 years although the 3 symptomatic fish died before reaching 2.5 years. In order to determine whether zprogerin and zprelamin A fish were prematurely aging, we used senescence-associated b-galactosidase (SA-b-gal), a hallmark of cellular senescence in vitro as well as a marker of the organismal aging process in vertebrates including zebrafish [42] [43] [44] [45] [46]. Sporadically enhanced SA-b-gal activity was observed in approximately 70% of morphologically still nonsymptomatic (almost normal shaped) transgenic fish (one of zprelamin A fish with high SA-b-gal is shown; Fig. 1E, left picture panel). We further confirmed the appearance of these accelerated aging symptoms in another cohort (18 months of age) comprising uninjected wt sibling (n = 56), F 0 zprogerin (n = 34), or F 0 zprelamin A fish (n = 28) ( Fig. 1D and E, right graphs). Intriguingly, the frequency of abnormal body shapes was higher in the zprogerin transgenic fish group than in the zprelamin A fish group, whereas the SA-b-gal intensities were much higher in the zprelamin A fish. These results suggest that accelerated aging symptoms may develop in most of the mosaicked transgenic fish whose cells carry a functional zprogerin or zprelamin A transgene that is sporadically expressed.

Knockdown of The LMNA Gene in Zebrafish Embryos Showing Muscle and Cartilage Abnormalities
Lamin A/C is ubiquitously expressed in multiple tissues from adult zebrafish ( Fig. 1B and Fig. S2), and we have also detected the expression of the corresponding gene, LMNA, by whole-mount in situ hybridization and semi-quantitative RT-PCR in zebrafish embryos ( Fig. S3) (Supplemental Data S1). Lamin A/C protein expression during zebrafish embryogenesis was found to be relatively lower than that in various adult tissues but this may be an artifact of high yolk protein levels in embryos (Fig. 1B, lower panel as shown for a larval sample at 6 days post fertilization, dpf). LMNA-null mice are born with a normal phenotype [47]. However, it has also been suggested previously that the ectopic expression of prelamin A or exon 9-deleted mutant lamin A affects developmental processes in the early phases of Xenopus embryogenesis [26] [48]. To examine the function of the type A lamins in the early development of zebrafish, we first tested translation-block morpholino antisense oligonucleotides (MO) (zLMNA-MO1) designed to target the translation initiation codon (ATG) and 59-untranslated regions of both lamin A and C. To determine the effectiveness of the resulting knockdowns at the protein level, zLMNA-MO1-transfected zebrafish AB9 fibroblasts were examined by western blotting. Compared with untransfected cells and control 5-mismatch MO1 ('5 mis' for MO1)-transfected cells, a downregulation of lamin A/C protein expression was observed (P,0.01) ( Fig. 2A). We further designed two MOs (zLMNA-MO2 and -MO2.1) that target the splicing junction the 39 end of exon 11 to generate aberrant splicing ( Fig. 2B and Fig. S4). This cryptic splicing of lamin A within exon 11 yields a protein with an 8amino acid deletion (zlamin A-D8) ( Fig. 2B and Fig. 1C). This removes the predicted site of Zmpste24/FACE1-mediated endoproteolytic cleavage [49], whereas the actual cleavage site in zPrelamin A has not yet been tested. The EGFP-fused zPrelamin A, zlamin A-D8 and zlamin A-D37/zProgerin proteins each localized at the nuclear envelope in transgenic embryos (Fig 3, D-F). The specific localization of the zlamin A-D8 and zlamin A-D37 proteins at the nuclear envelope showing abnormally wrinkled misshapen structures was concurrently observed in embryos expressing both constructs ( Fig. 3E and F).
The phenotype induced by the knockdown of zebrafish lamin A/C by different MOs during early development was not particularly evident at the gross morphology level at 24 hours post fertilization (hpf) (Fig. 2C), except an occasional observation of rough skin (Fig. 2D). However, when we analyzed slow skeletal muscle fibers in more detail by staining myosin heavy chains with an F59 antibody (Fig. 4A, e-h), muscle damage was evidenced by a failure in slow muscle fiber migration to the lateral surface along with irregular, gapping (Fig. 4A, f; MO1), and/or wavy (Fig. 4A, h; MO2) muscle fibers in approximately 50% of the zLMNA-morphants (Fig. 4B). In contrast, no evident differences in fast muscle fibers in the morphants were identified via the staining of myosin light chains with an F310 antibody (Fig. 4A, i-l). We further demonstrated the slow muscle phenotype via expression of the zprogerin (zlamin A-D37) transgene. zprogerin transgenic zebrafish embryos (,40%) showed similar slow muscle fiber abnormalities as shown later.
We next examined craniofacial phenotypes in our morphants, as some mutations in the LMNA gene are known to cause craniofacial anomalies and/or disproportion in both humans and mice [50] [51]. In zLMNA-MO1 morphants at 6 dpf, a dorsal displacement of Meckel's cartilage in relation to the palatoquadrate (n = 32 of 65; 49%) (red arrow in Fig. 4C, f) was observed and the jaw appeared to be shorter in the ventral view of the head (Fig. 4C, j). In contrast, this seemed not to be due to a reduction in the size of any elements of the jaw apparatus but rather to a reduced angle between the Meckel's cartilage and the palatoquadrate. Moreover, from the ventral view, retrognathic pharyngeal cartilage defects (a red line in  Table 1). Although obvious pharyngeal cartilage defects in morphology, including retrognathic formation, were not apparent within zLMNA-MO2 morphants ( Table 1), the MO2 molecule induced a relatively hypoplastic formation of cartilage matrix with occasionally reduced Alcian blue staining, indicating a possible alteration of the composition of the matrix. RT-PCR analysis of adult zlamin A revealing strong expression in the brain, heart, liver and muscle (upper panels). Western analysis of 6 dpf embryo lysates and 1-year old adult tissues was preformed using an anti-chicken lamin A polyclonal antibody (ab14309), as well as an anti-mouse actin monoclonal antibody as a loading control (lower panels). (C) Amino acid sequence alignment of the C-terminal portions in human (Homo sapiens) lamin A (Hs. lamin A) and zebrafish (Danio rerio) lamin A (Dr. lamin A) and their deleted forms (D50 in human; D37 and D8 in zebrafish) induced by cryptic splicing events. (D) External features of a wild-type non-transgenic sibling fish (wt; 1.5-year old). Two representative zprogerin transgenic fish with curved trunks (zlamin A-D37; 1.5-year old) are shown. A zprelamin A transgenic fish with a wavy trunk (2.1-year old) is also shown. In these transgenic fish, spinal curvatures and disproportional body shapes are prominently observed. Scale bars, 1 cm. Qualifications of the abnormal body shapes are shown in the right graph (cohorts of 18 months of age). (E) SA-b-gal staining of wild-type sibling and transgenic fish (1.9-year old). Quantifications of the SA-b-gal intensities in the fish are shown in the right graph (cohorts of 18 months of age, *P,0.01). doi:10.1371/journal.pone.0017688.g001

Apoptosis, Senescence, and Lipodystrophy in zLMNA Morphants
Given the morphologies observed in our zebrafish morphants, we assayed for apoptotic cell death in these embryos using the TUNEL assay. Apoptotic cells were predominantly detected in the brain and trunk of the zLMNA-MO1-injected embryos at 24 hpf (Fig. 5A, b). On the other hand, in zLMNA-MO2-injected embryos, we observed only a slight induction of apoptosis (Fig. 5A, d). In the control embryos, no significant apoptosis induction was observed ( Fig. 5A, a, c), although scattered apoptotic cells were occasionally detectable at background levels throughout the embryos.
Given the obvious appearance of apoptotic cells in our morphants, we assessed whether cell proliferation was altered in any way. Although the expression of proliferating nuclear antigen (PCNA) was not substantially affected ( Fig. S5), the cell-cycle progression was prominently disturbed in the morphants. Progression through the S and M phases of the cell cycle, assessed by the incorporation of 5-bromo-2-deoxyuridine (BrdU) and the immunostaining of a mitotic marker, phosphorylated histone H3 (pH 3), respectively, was decreased in both cases in the zLMNA-MO1 morphants (Fig. 5A, f, j, see right graph). Intriguingly, in the zLMNA-MO2 morphants, an accelerated progression through S phase was observed but no significant alterations during M phase were detected (Fig. 5A, h, l, see right graph), implicating the accumulation/arrest in S-phase with increased DNA damage and/or dysregulated DNA repair [52].
Semi-quantitative RT-PCR was used to assay individual morpholino-injected embryos for the expression of p53, p21 WAF1/CIP1 , cyclin D1, cyclin E, and cdk2, with b-actin used as a normalization control. A slight upregulation of p53 mRNA was observed in zLMNA morphants (either MO1 or MO2), when compared with wild-type embryos (data not shown) and control 5mismatch MO-injected embryos (Fig. 5B). In addition, a downstream target of the p53 pathway, cyclin-dependent kinase inhibitor 1, p21 WAF1/CIP1 , was also correspondingly upregulated in the zLMNA morphants (Fig. 5B). Interestingly, neither cyclin D1 nor cyclin E, which are required for progression into G1 and through to the start of S phase, showed any transcriptional changes in the morphants and cdk2, which associates with cyclin E and cyclin A during the transition from G1 to S phase, was slightly decreased only in zLMNA-MO1-injected embryos, suggesting a specific cell-cycle delay resulting from a total lamin A/C decrease. These results are representative of similar analyses on additional single embryos (wild-type, n = 4; MO1-5mis, n = 4; MO1-MO, n = 8; MO2-5mis, n = 4; MO2-MO, n = 8; data not shown) and suggest that, in the reduced lamin A/C expression (by MO1) or a disrupted splicing of this gene (by MO2), p21 WAF1/CIP1 is commonly activated. However, the reason for the observed distinct increases in TUNEL-positive apoptotic cells as well as differential cell-cycle profiles remains unclear. Recently, we have demonstrated that an embryonic (larval) senescence phenotype caused by specific gene mutations and by stress is readily detectable via the SA-b-gal staining of zebrafish embryos and larva within 6 days post-fertilization (dpf) [42]. We therefore examined the zlamin A/C-depletion (by MO1) or zlamin A-disturbance (by MO2) effects in our morphants by staining with SA-b-gal. Intriguingly, the zLMNA-MO2 morphants showed a high intensity of SA-b-gal at 6 dpf, whereas zLMNA-MO1 morphants did not show significantly detectable SA-b-gal activity (Fig. 5C). Lipodystrophy is one of the hallmarks of the laminopathies and HGPS. To thus examine whether neutral fat distribution is affected in our current morphants, we stained whole larvae with Oil Red O (ORO). Significant ORO lipid staining was detectable in the control larvae as reported previously [53] [54] [55]. However, in the zLMNA morphants, a strikingly lower level of lipids was observed in the head, otic vesicle jaw, and yolk (Fig. 6). MO1 was found to induce a more severe lipid loss although MO2 still caused a significant lowering of these levels also. To determine the expression levels of zebrafish peroxisome proliferator-activated receptor-c (PPARc), relative to b-actin as a control, RT-PCR was performed on individual morpholino-injected embryos. The result showed that either a lamin A/C knockdown by MO1 or the induction of cryptic splicing by MO2 is associated with decreased PPARc mRNA levels (Fig. S6), indicating reduced adipogenic differentiation in embryos with lipodystrophic phenotypes.

FTI Treatment of Transgenic Zebrafish Expressing Progerin
To establish the extent of the embryonic phenotypes that arise in lamin A-depleted zebrafish, we further studied the F 1 generations of germline-positive zprogerin transgenic fish since their F 0 parents showed normal reproduction up until severe aging-like manifestations appeared ( Fig. 1D and E). Most of the zprogerin transgenic fish embryos were morphologically normal, similar to the morphants at 24-30 hpf (Fig. 7A). However, all of the examined transgenic fish had the misshapen nuclei ( Fig. 7B, a,  b). Once we analyzed their slow skeletal muscle fibers, striking abnormalities with wavy migrations could be observed in a number of cases at 24 hpf (Fig. 7C, b). The zprogerin transgenic fish also showed obvious SA-b-gal induction even at 3.5 dpf (Fig. 7C, e) To demonstrate the effects of FTI in zebrafish, we then employed a zebrafish fibroblast cell line (AB9). In FTI treated zebrafish cells, the accumulation of prelamin A was detectable in 20 mM, suggesting that the farnesylation of intrinsic prelamin A was prevented as was further processing of the protein (Fig. 7D, right graph).
In zprogerin transgenic embryos, the ratio of nuclei that had abnormal shapes was 54.364.9% (Fig. 7B). Previous studies have shown that when the farnesylation of prelamin A in HGPS cells and in Zmpste24-deficient cells is blocked by FTI, prelamin A does not accumulate at the nuclear rim and the percentage of cells with misshapen nuclei is reduced [56] [10] [57] [58] [8] [9]. To determine whether the nuclear shape abnormalities caused by zProgerin in our transgenic zebrafish embryos could also be ameliorated by FTI, different concentrations of this inhibitor were administrated to 5 hpf embryos which were further incubated for 25 h. As expected, the FTI treatment of zprogerin transgenic embryos reduced the extent of the abnormally misshapen nuclear morphology in a dose-dependent manner (Fig. 7B, right   graph). The frequency of misshapen nuclei expressing zProgerin (54.364.9% in untreated condition) was reduced to 26.3616.2% upon exposure to 100 mM FTI (P,0.05). Moreover, the slow muscle fiber abnormalities, in addition to the onset of embryonic senescence, in these transgenic embryos were significantly blocked by FTI treatment (Fig. 7C, c, f). These findings are consistent with the results in other progeria cells and animals treated with FTI [59].

Shortened Lifespan Resulting from Stable Progerin and Prelamin A Expression in Zebrafish
To measure lifespan, stable zprogerin or zprelamin A transgenic fish were maintained with their wild-type siblings and their survival was measured using Kaplan-Meier survival curves (Fig. 8).
These fish were not sterile as described above, but their observed lifespan was significantly shorter than that of wild-type zebrafish reported previously [41] [42]. We also examined the prelamin A transgenic zebrafish and interestingly, these transgenic fish also suffered from a relatively short lifespan. We maintained populations of these transgenic zebrafish and their wild-type siblings until death, and conducted observational studies on their lifespan. We scored the two cohorts of zprogerin transgenic fish (no genders identified; Cohort 1, n = 102; Cohort 2, n = 54) in comparison with their wild-type siblings (no genders identified; Cohort 1, n = 22 and Cohort 2, n = 46), and found significant decreases in the lifespan of the zprogerin transgenic animals (Cohort 1, P,0.0005; Cohort 2, P,0.0001, log rank test) (Fig. 8). The plots in Figure 8 show Kaplan-Meier survival curves and 95 percentile confidence intervals on the cumulative probability distribution (the survival function).
We additionally found that our zprelamin A fish cohorts (no genders identified; n = 36) manifested a shorter lifespan compared with that of their wild-type siblings (no genders identified; n = 91; P,0.0001, log rank test) (Fig. 8). In both the zprogerin and zprelamin A cases, there were significant differences found between the wildtype siblings and the transgenic fish. These lifespan analyses of zprogerin and zprelamin A transgenic zebrafish in accordance with the embryonic senescence phenotypes suggest that early developmental manifestations can predict a role for the LMNA gene.

Discussion
The zebrafish is a valuable vertebrate model system for the elucidation of novel molecular pathways of aging as well as early development [60]. In particular, this model organism develops an array of senescence symptoms resembling those in humans [61] [62] [63] [44], and is amenable to large-scale genetic screens that can be targeted to conserved aging pathways in vertebrates [42]. However, the establishment of human disease-associated premature senescence-bearing fish had not been pursued previously. In this study, we developed zebrafish progeria and laminopathy models in three different ways: i) exogenous expression of the zebrafish version of 'progerin' (zProgerin) protein, ii) translationalblock MO (zLMNA-MO1)-induced translational downregulation to reduce lamin A/C protein levels, and iii) splice-block MO (zLMNA-MO2)-induced cryptic splicing of LMNA, which deleted 8 amino acid residues containing the predicted site of posttranslational cleavage. First, the exogenous transgenic expression of zProgerin/zlamin A-D37, for which anticipated similar symptoms to MO2 morphants were anticipated, demonstrated a shortened lifespan as well as accelerated aging in adult zebrafish. Second, the translational-block of LMNA by MO1 induced apoptosis, cell-cycle arrest, and craniofacial abnormalities (cartilage defects). In contrast, splice-block generating zlamin A-D8 with  a cryptic splicing of LMNA by MO2 produced embryonic senescence and S-phase accumulation/arrest. Interestingly, a lipodystrophic phenotype was induced by both MO1 and MO2. These phenotypic outcomes are summarized in Table 2 with reference to human progeria and laminopathies. Our data also demonstrated that long-term increased levels of not only progerin but prelamin A in transgenic zebrafish results in a shortened lifespan in these animals. Hence, the expression of normal prelamin A, as well as progerin [14] [64], may be linked to senescence induction during the physiological aging process [65] [66].
Lamin A/C expression is closely correlated with cell differentiation in metazoans, and LMNA mRNA is also developmentally expressed in zebrafish embryos from at least the late gastrula period (9 hpf). Some interesting observations were recently reported for developmentally functional A-type lamins in a comparative study of Drosophila and humans [25] [67]. On the other hand, detailed insights into any functional role of lamin A/C during early vertebrate development have not been obtained to any great extent to date. Because embryonic stem cells appear to express lamin A/C upon differentiation [20], and lamin Aassociated stem cell dysfunction has been linked to disruption of the Wnt and Notch signaling pathways [19] [23] [18] [48], which are essentially involved in the process of early vertebrate development, our current observations of embryonic senescence or apoptosis in LMNA-disturbed zebrafish might be consistent with the dysregulation of somatic stem cells leading to organismal premature aging. In particular, it is plausible that the loss/decrease of plasticity and/or onset of self-renewal defects in mesenchymal stem cells may be responsible for the specificity of accelerated tissue degeneration, as most affected tissues in progeria are of mesenchymal origin [18].
Between the knockdown of the LMNA gene (by zLMNA-MO1) and the induction of mis-splicing within exon 11 of LMNA (by zLMNA-MO2), some phenotypic differences were observed in treated zebrafish embryos in terms of cell-cycle regulation, apoptosis and senescence. These may be attributable to multiple interacting proteins associated with the cell cycle, including cyclin D3, retinoblastoma protein (pRb), and PCNA [ [73], although further investigations will be required in this regard. Clearly, there are some specific cell-cycle alterations due to manipulation of lamin A. The knockdown of A-type lamins by MO1 resulted in an arrest of the cell cycle via both an S and M phase block. On the other hand, induced cryptic splicing within exon 11 by MO2 led to increased S phase entry/accumulation without largely affecting M phase to any greater extent. Any linkage of these cell-cycle regulatory mechanisms to apoptosis versus senescence will need further extensive analyses. In this regard, with the exception of p21 WAF1/CIP1 , no remarkable differences in the mRNA levels of the cyclins and cdks were observed in our current experiments. As a cyclin-dependent kinase inhibitor, p21 binds and inactivates cyclin-cdk2 or cdk4 complexes including cyclin D3/cdk4 which has been shown to be responsible for the phosphorylation of pRb [74]. p21 also interacts with PCNA and blocks its ability to activate S phase DNA replication and DNA damage repair [74]. Presumably at the inner nuclear membrane, lamin A orchestrates a functional network of the three interacting proteins, cyclin D3, pRb and PCNA under the control of p21 in cell cycle. In addition to the cell-cycle control, p21 can play either anti-or pro-apoptotic roles in cells and also mediate cellular senescence [74]. Hence, it is not entirely surprising that the decrease-of-function of lamin A/C by MO1 and the gain-offunction of zlamin A-D8 by MO2 could have such distinct effects on cell cycle, apoptosis and senescence in zebrafish embryos. Since further global gene expression changes as well as epigenetic alterations would be anticipated upon the disruption of lamin A functions [75] [76], the organization and modification of naive chromatin might be worth consideration in future studies.
The phenocopies observed in several zebrafish tissues such as muscle, adipose and skin with their short lifespan as found in human progeria and laminopathies are extremely intriguing. As for the LMNA function in our zebrafish progeria model, it should be noted that the decrease-of-function of LMNA showed specific developmental defects without any significant induction of SA-bgal i.e. embryonic senescence. By contrast, specific gain(s)-offunction(s) of farnesylated forms of progerin and prelamin A could play a crucial role(s) in the pathogenesis of the fish progeroid (senescence) phenotype which was reduced by FTI at least in zProgerin embryos. Therefore, the farnesylated progerin and an excess of farnesylated prelamin A may essentially become pathogenic and induce a progeroid phenotype in addition to other laminopathy phenotypes. This would be consistent with the recent observation that exclusive expression of farnesylatable progerin elicited a relatively more severe progeroid phenotypes than those of its non-farnesylated form in knock-in mice [37]. On the other hand, exclusively expressed non-farnesylated prelamin A still induced prominent cardiomyopathy but not progeria in knock-in mice, whereas the survival of such mice which exclusively produce farnesylatable prelamin A was very similar to that of wildtype mice [38], which is quite possibly due to the lower amount of farnesylated prelamin A. Embryonic senescence and further unforeseen developmental abnormalities may also be occurring in human (or mice) progeria given that severe neonatal lethality has been reported [15] [77] [78].
The power of the zebrafish model lies in its ability to accommodate genetic screens to identify responsible or modifier genes and also facilitate compound screens to identify new chemicals that influence developmental processes and signaling pathways. In our current study, we used this tractable system to analyze both embryonic and adult premature senescence, which we modeled on severe pediatric progeroid disease in humans. By generating the stable transgenic zebrafish lines that expresses zProgerin as well as by transiently manipulating the LMNA-MOs in early developmental stages of zebrafish embryos, premature aging can now be assumed to be adequately modeled in this system. The fairly late onset of progeria after germline transmis-    sion of the transgene made it easier to propagate this stable line. Of note, however, later than fourth generations of transgenic lines onwards showed a gradual decrease in the expression of GFP-zProgerin and no longer maintained a steady-state expression of the transgene further over the multiple generations. Technology to render higher expressing transgenic progerin alleles via induction would also be very valuable for in vivo use in the zebrafish, and we are currently testing this approach for its capacity to generate large numbers of different vector-based and promoter-driven transgenic fish for subsequently stable genetic manipulations. The utility of zebrafish to perform screens for small molecules that perturb or restore normal development has been demonstrated [79] [80]. Our reverse-genetic progeroid zebrafish model may also provide the opportunity to conduct large-scale screens for new drugs for the prevention and treatment of normal aging and ageassociated diseases in addition to progeria and laminopathies.

Zebrafish maintenance and ethics
Zebrafish were maintained with a 14:10 h light/dark cycle and fed living brine shrimp twice daily. Brine shrimp were given using 1 mL pipettes at an amount of about 0.75 mL brine shrimp per 20 fish. Flake food was also given every other day semiquantitatively according to the number of fish in the tanks. A continuously cycling Aquatic Habitats TM system was used to maintain water quality (Apopka, FL) which completely replaces the water in each tank every 6-10 min. Each tank used was a baffle/tank system that eggs water in a circular motion to ensure flushing and water turnover. Ultraviolet (UV) sterilizers (110,000 mW-s cm 22 ) were employed to disinfect the water and prevent the spread of disease in the recirculating system. The water temperature was maintained at 2860.5uC. The system continuously circulated water from the tanks through Siporax TM strainers, through a fiber mechanical filtration system (three different pore-sized filters; one pre-pad filter of 120 mm, two bag filters of 100 and 50 mm, respectively), and finally into a chamber containing foam filters and activated carbon inserts. Water quality was tested daily for chlorine, ammonia, pH, nitrate and conductivity, and also was under real-time computer monitoring with alarms to signal potential fluctuations. The general health of each fish was also observed on a daily basis, and abnormal looking or acting fish were quarantined into isolated tanks that were not connected to the general circulation. The water in these quarantined tanks was treated with methylene blue. If and when fish recovered, they were returned to their original tanks on the general circulation system. The zebrafish embryos were staged by hours post fertilization (hpf) at 28.5uC and also by morphological criteria for experiments [81].
All animal experiments were approved by and conducted in accordance with the guidelines established by the Institutional Animal Care and Use Committee (IACUC) at The Scripps Research Institute, IACUC approval number 09-0009 and The Schepens Eye Research Institute, IACUC approval number S-150-0909 and S-186-1210.

Molecular cloning of zebrafish lamin A/C
To isolate the full length zebrafish lamin A/C cDNA, the zebrafish lamin A mRNA sequence deposited in GenBank was used (accession number AF397016) [82]. mRNA was isolated from zebrafish embryos at 24 hpf using QuickPrep mRNA purification Kit (Amersham Biosciences, Uppsala, Sweden). Poly(A) RNA was then purified with oligo(dT)-resin. Double-stranded cDNA was synthesized using M-MLV reverse transcriptase (Promega, Madison, WI), followed by PCR by ExTaq (Takara Mirus Bio, Madison, WI). The primers used were as follows: forward (F): 59-ATGGAGACTCCAGGTCAGAAAC-39, reverse (R): 59-AAG-CACAGATGGTCAGGGTTTG-39. The PCR program entailed a 30-cycle amplification for 30 sec at 94uC, 30 sec at 60uC, and 4 min at 72uC. The amplified DNA fragments were purified by excision from an agarose gel, and cloned into the pCRII-TOPO cloning vector (Invitrogen, Carlsbad, CA). Nucleotide sequences were determined with an automated sequencer (ABI 3700, AME Bioscience, Norway) using the ABI BigDye V3.1 cycle sequencing system.
To analyze the full-length sequence of the zebrafish lamin C gene, 39-rapid amplification of cDNA ends (39-RACE) was performed with a Generacer kit (Invitrogen). A forward primer (59-CAGAACTCATGGGGCAGCGGTGATTTGT-39; residues 535-544 of zebrafish lamin A) and a forward nested primer  residues 540-549 of zebrafish lamin A) were used to amplify the 39-end portion of the zebrafish cDNA by PCR in conjunction with the adapter primer supplied with the kit (Invitrogen). PCR was performed according to the manufacturer's instructions using 10 ng of cDNA template and 1 ng of each primer in a total volume of 50 ml per sample. The samples were subjected to touchdown PCR (2 min at 94uC, 5 cycles of 30 sec at 94uC and 2 min at 72uC, 5 cycles of 30 sec at 94uC and 2 min at 70uC, and 25 cycles of 30 sec at 94uC, 30 sec at 65uC, and 120 sec at 72uC). The amplified DNA fragments were then subcloned into the pCRII-TOPO cloning vector and sequenced. Multiple sequence alignments were carried out using Megalign software (DNAstar, Madison, WI). All parameters were set at default. A full-length zebrafish lamin A cDNA fragment was subcloned into the KpnI and BamHI sites of the pEGFP-C1 Vector (Clontech, Mountain View, CA).

Whole mount mRNA in situ hybridization of embryos
For in situ hybridization, digoxigenin-labeled antisense RNA probes were synthesized from the 2.0 kb zebrafish lamin A/C insert which had been directionally subcloned into the pCRII-TOPO vector (Invitrogen) using an in vitro transcription kit (Roche Applied Science, Indianapolis, IN). Antisense mRNA was transcribed from the XbaI-linearized plasmid using SP6 polymerase, and sense mRNA from the KpnI-linearized plasmid using T7 polymerase as a control. In situ hybridization analyses and development of whole-mount zebrafish embryos were performed as previously described [83] [84].

RNA isolation and RT-PCR analysis for zebrafish lamin A/C
Total RNA was extracted from 24 hpf embryos using TRIzol reagent according to the manufacturer's protocol (Invitrogen). Double-stranded cDNA was synthesized using M-MLV reverse transcriptase (Promega), followed by PCR with ExTaq (Takara). For RT-PCR analysis of zlamin A/C, the forward and reverse primers used were 59-TCAGCAAAGTGCGTGAGGACTACA-39 and 59-GCGAAGCTGCTCTTCATGTTGGTT-39. Specifically for zlamin A, the forward and reverse primers used were 59-TTTG-TTCCAGACCACCCTCATC-39 and 59-AAGCACAGATGGT-CAGGGTTTG-39: and for zlamin C, 59-TGGCAGGTGAA-GAGGCAGATTG-39 and 59-GACCCGCAACATAAAGGCTA -39. For the detection of the cryptic splicing region in lamin A exon 11 caused by MO (zLMNA-MO2 and zLMNA-MO2.1) injections, the primers used were as follows: forward, 59-TTTGTTCCAGAC-CACCCTCATC-39 and reverse, 59-AAGCACAGATGGT-CAGGGTTTG-39 (as described above). As for the control primers, the forward and reverse primers used to amplify EF1a were 59-ACCACCGGCCATCTGATCTACAAA-39 and 59-ACGGATG-TCCTT GACAGACACGTT -39, respectively. The forward and reverse primers used to amplify b-actin were 59-CCCAGACAT-CAGGGAGTGAT-39 and 59-CACCGATCCAGACGGAGTAT-39, respectively [85]. For amplification by PCR, the initial denaturing step at 94uC for 5 min was followed by 25 amplification cycles of 30 sec at 94uC; 30 sec at 60uC; 60 sec at 72uC, and a final extension period of 10 min at 72uC. Each amplification reaction was separated on a 1.5% agarose gel stained with ethidium bromide-stain and the bands were visualized and recorded using a Multi Image Light Cabinet (Cell Bioscience, Santa Clara, CA).

Western blot analysis
The zebrafish cells, embryos, and the tissues of adults were homogenized in RIPA buffer supplemented with one tablet of protease inhibitor cocktail (Roche Applied Science)/10 ml on ice. The protein extracts were then fractionated on 10% SDSpolyacrylamide gels and transferred to PVDF membranes (Bio-Rad, Hercules, CA). The blot was incubated with the primary antibody diluted in TBST (1:2000 dilution anti-lamin A antibody, ab14309; Abcam, Cambridge, MA) overnight at 4uC. After washing, the blot was then incubated with the secondary anti-IgY antibody (1:2000 dilution, ab6753; Abcam) at room temperature for 1 h and visualized using an ECL kit (Perkin Elmer, Waltham, MA) in accordance with the manufacturer's instructions. In addition, reactivity with an actin antibody (AC-40; Sigma-Aldrich, St. Louis, MO) at a dilution of 1:1000 was used to normalize the loading per lane.
Senescence-associated b-gal activity (SA-b-gal) assay of zebrafish embryos Zebrafish larvae at 3.5-6 dpf were washed in phosphate buffer saline (PBS) three times and fixed overnight in 4% paraformaldehyde (PFA) with PBS at 4uC. Adult zebrafish were also washed in PBS 3 times and fixed for 3 days in 4% PFA in PBS. After fixation, samples were washed 3 times in PBS, pH 7.5, twice again in PBS, pH 6.0 at 4uC, and then incubated at 37uC (in the absence of CO 2 ) for 12-16 h with SA-b-gal staining solution (5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM MgCl 2 in PBS at pH 6.0). Staining intensity was quantified by pixel analysis using with Adobe Photoshop CS software as described previously [42].

Apoptosis detection and Immunohistochemistry
Embryos at 24-48 hpf, were fixed overnight in 4% PFA with PBS at 4uC and stored in 100% methanol at 220uC until samples were used. Samples were then incubated in 100% acetone at 220uC for 20 min. Following fixation, the embryos were rinsed three times with PBS containing 0.1% Tween-20 (PBST; 5 min each). Samples were then permeabilized by treatment with 0.5% Triton X-100 and 0.1% sodium citrate in PBS for 15 min, then with 5 to 50 mg/ml proteinase K (Invitrogen) for 5 to 25 min depending on embryo stages. Embryos were subjected to TUNEL assay by using the ApopTag Red in situ Apoptosis Detection Kit (Chemicon, Temecula, CA) according to the manufacture's instruction.
Anti-PCNA FITC-labeled antibody (BD-Biosciences-Pharmingen,) (1:100) staining was performed by the same procedure done in pH3 antibody staining as described above, except the secondary antibody.

Detection and quantitation of visible and fluorescent images
All animals were photographed under the same conditions using reflected light (for SA-b-gal) or fluorescence (for BrdU and pH 3) under a macro microscope, AZ100 (Nikon). SA-b-gal activity in each animal was quantitated using a selection tool in Adobe Photoshop for a color range that was chosen by 25 additive blue color selections of regions that showed visually positive SA-b-gal staining. For analyses of embryos, these regions were selected in each embryo proper only and not in the yolk. This was to eliminate variability due to differences in initial yolk volume and yolk consumption over time. Since the yolk stains much more intensely blue for SA-b-gal at all stages of development than any other embryonic tissues, it was desirable to exclude this as a source of variability. Following pixel selection, a fuzziness setting of 14 was used, and the chosen pixel number was calculated using the image histogram calculation. Incorporated BrdU and pH 3 positive cells in each animal were quantitated in the same way except that a fuzziness setting of 64 was used.

Oil red O staining
Zebrafish larvae at 5 dpf were PBS three times and fixed overnight in 4% PFA with PBS at 4uC. After fixation, samples were washed twice in PBS, followed by oil red O staining in freshly filtered 0.3% oil red O in 60% 2-propanol for 30 min to 2 h [54]. Samples were quickly washed in 60% 2-propanol for a few times, and then finally washed once in PBS.

Semi-quantitative RT-PCR
First strand cDNA was generated using total RNA that had been isolated from LMNA MO-and 5-base mismatch control (5mis) MO-injected 24 hpf embryos with Trizol (Invitrogen). The linear amplification ranges were then determined for each of the primer sets as described previously [85]. Primers used to amplify the zebrafish cdk2 and PPARc genes were designed using Genetyx mac software (cdk2, forward primer: 59-AGTTACAGGA-GAAACTGTCGCGCT-39, reverse primer: 59-AGCGTTCT-TTGCCGAGATCCTCTT-39; PPARc, forward primer: 59-AAGAGCCGCAACAAGTGCCAATAC-39; reverse primer: 59-AGAACTCGAACTTGGGCTCCATCA-39). All other primer sequences had been reported previously by Liu et al. [85]. PCR conditions were as follows: an initial denaturing step at 94uC for 5 min was followed by 25 amplification cycles of 30 sec at 94uC; 30 sec at 60uC; 60 sec at 72uC, and a final extension period of 10 min at 72uC.

Generating transgenic zebrafish
To generate transgenic zebrafish expressing EGFP-tagged zProgerin/zlamin A-D37, the corresponding expression construct was generated using the QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA) in accordance with the manufacturer's instructions. This vector was then injected into one-cell stage embryos with meganuclease I-SceI [88] [89]. Germline transgenic founders (F 0 ) were identified by screening F 1 embryos via PCR. Four founder transgenic fish carrying the EGFP-zProgerin DNA construct were thus identified. F 1 embryos of these transgenic founders were then allowed to grow to maturity. Adult transgenic fish of the F 1 generation were screened individually by PCR using DNA extracted from their caudal fins. For genotyping, EGFP-specific primers (EGFP-sense: 59-TCTTCTTCAAGGAC-GACGGCAACT-39 and EGFP-antisense: 59-TGTGGCGGA-TCTTGAAGTTCACCT-39) and control PCR primers (Wnt5asense: 59-CAGTTCTCACGTCTGCTACTTGCA-39 and Wnt5aantisense: 59-ACTTCCGGCGTGTTGGAGATTTC-39) were mixed and 1 mM of each was used in a PCR reaction with a final volume of 25 ml.
For FTI treatment, cells were cultured to 70% confluence, and treated with 5, 10, and 20 mM of the inhibitor in cell culture medium for 72 h until harvesting. Embryo FTI treatments (20,40, and 100 mM) were performed from 5 through 30 hpf in E3 medium with 1% DMSO at 28.5uC.

Quantitative analysis and statistics
Data processing and statistical analyses were performed using Statistical Package for the Social Sciences (SPSS) version 14.0. This software was used to generate each of the scatter plots, tables and graphs shown in the text and perform statistical tests where appropriate. Additional statistical analyses were performed using a custom MATLAB script to analyze survival estimates by the method of Kaplan and Meier, and to compare of survival between mutant fish and their wild type controls using the log rank test [90]. Figure S1 The sequence and genetic organization of zebrafish lamin A/C is conserved with human lamin A/ C. The zebrafish lamin A/C gene (zlamin A/C) comprises 12 exons, encoding two globular domains and a rod domain; zlamin C is encoded by exons 1 to 9.