|Home | About | Journals | Submit | Contact Us | Français|
Hutchinson-Gilford progeria syndrome (HGPS), a rare disease that results in what appears to be premature aging, is caused by the production of a mutant form of prelamin A known as progerin. Progerin retains a farnesyl lipid anchor at its carboxyl terminus, a modification that is thought to be important in disease pathogenesis. Inhibition of protein farnesylation improves the hallmark nuclear shape abnormalities in HGPS cells and ameliorates disease phenotypes in mice harboring a knockin HGPS mutation (LmnaHG/+). The amelioration of disease, however, is incomplete, leading us to hypothesize that nonfarnesylated progerin also might be capable of eliciting disease. To test this hypothesis, we created knockin mice expressing nonfarnesylated progerin (LmnanHG/+). LmnanHG/+ mice developed the same disease phenotypes observed in LmnaHG/+ mice, although the phenotypes were milder, and mouse embryonic fibroblasts (MEFs) derived from these mice contained fewer misshapen nuclei. The steady-state levels of progerin in LmnanHG/+ MEFs and tissues were lower, suggesting a possible explanation for the milder phenotypes. These data support the concept that inhibition of protein farnesylation in progeria could be therapeutically useful but also suggest that this approach may be limited, as progerin elicits disease phenotypes whether or not it is farnesylated.
Hutchinson-Gilford progeria syndrome (HGPS) is a rare pediatric progeroid syndrome associated with the production of a mutant form of prelamin A (1–5). Affected children appear normal at birth but soon develop multiple disease phenotypes resembling premature aging, including slow growth, loss of adipose tissue, alopecia, osteoporosis, and occlusive vascular disease (1, 3, 6). HGPS is caused by a LMNA point mutation that alters mRNA splicing and leads to the production of a mutant prelamin A, progerin, that contains a 50–amino acid internal deletion within the carboxyl-terminal domain of the protein (1, 7). Progerin, like wild-type prelamin A (8), contains a carboxyl-terminal CaaX motif, which triggers protein farnesylation. The farnesyl lipid is added to the thiol group of the cysteine within the CaaX motif by a cytosolic enzyme, protein farnesyltransferase (FTase). Although the CaaX motif of progerin is intact, the 50–amino acid deletion eliminates the cleavage site for the endoprotease ZMPSTE24 (7, 9, 10); consequently, progerin cannot undergo the final endoproteolytic processing step that would normally clip off the carboxyl-terminal portion of the protein (including the farnesylated cysteine) (1, 2, 7). Thus, farnesylated progerin accumulates in cells (7). Progerin is targeted to the nuclear rim and causes misshapen nuclei, almost certainly by interfering with the integrity of the nuclear lamina (1, 2, 11–13).
Yang et al. (13) recently created a gene-targeted HGPS allele, LmnaHG, which exclusively yielded progerin. Heterozygous mice (LmnaHG/+) exhibit a variety of phenotypes similar to those in humans with HGPS (e.g., normality at birth followed by slow growth, osteoporosis, alopecia, partial lipodystrophy) (14). Like fibroblasts from children with HGPS (1, 2), LmnaHG/+ mouse embryonic fibroblasts (MEFs) have misshapen nuclei (13). We hypothesized that protein farnesylation could be important for disease pathogenesis (15) and further hypothesized that inhibiting protein farnesylation with a FTase inhibitor (FTI) might reduce the frequency of misshapen nuclei and ameliorate the disease phenotypes in LmnaHG/+ mice. Indeed, an FTI lowers the frequency of misshapen nuclei in LmnaHG/+ MEFs (13). Also, in 2 independent studies (14, 16), a highly selective FTI reduces disease phenotypes in LmnaHG/+ mice. An FTI also reduces disease phenotypes in ZMPSTE24-deficient mice (17), where the farnesylated form of wild-type prelamin A accumulates (9, 10). While an FTI was unequivocally efficacious in all 3 studies (14, 16, 17), the amelioration of disease was far from complete; all of the FTI-treated mice ultimately developed severe phenotypes and succumbed to the disease.
The improved disease phenotypes in LmnaHG/+ mice treated with an FTI have prompted a clinical trial of an FTI treatment regimen in children with HGPS (18, 19). Given the merciless nature of this disease, we hope that the FTI treatment will be efficacious. However, key questions remain about the scientific underpinnings of the FTI strategy. An FTI will leave HGPS cells with a distinctly abnormal lamin protein, nonfarnesylated progerin, and it is unclear whether that protein might retain the ability to elicit disease. An optimist might contend that nonfarnesylated progerin would be entirely nontoxic, given that FTIs can markedly reduce the frequency of misshapen nuclei in HGPS fibroblasts (13, 20, 21). On the other hand, a pessimist would point to the less-than-complete amelioration of disease phenotypes in FTI-treated LmnaHG/+ mice (14, 16) and suggest that nonfarnesylated progerin might retain the capacity to cause disease.
The standard approach for assessing the functional importance of protein farnesylation is to examine the properties of a nonfarnesylated mutant protein in which the cysteine of the CaaX motif has been replaced with a serine (22–24). Serine is identical to cysteine, except that a side-chain sulfur atom is replaced with an oxygen. In the current study, we applied this approach and created a new line of HGPS mice expressing a nonfarnesylated version of progerin. Aside from the point mutation in the CaaX motif, the new gene-targeted HGPS mice (LmnanHG/+) were identical to the original mouse model (LmnaHG/+). We went on to compare the phenotypes of genetically identical LmnanHG/+ and LmnaHG/+ mice.
We constructed a gene-targeting vector to create a mutant Lmna allele, LmnanHG, designed to produce nonfarnesylated progerin (Figure (Figure1A).1A). The LmnanHG allele is identical to the LmnaHG/+ allele created earlier (13), except that the cysteine within the CaaX motif was replaced with a serine (which prevents protein farnesylation). Two targeted ES cell clones were identified by Southern blotting (Figure (Figure1B)1B) and used to create male chimeric mice, which were bred with C57BL/6 females to produce LmnanHG/+ mice. LmnanHG/+ MEFs produced progerin from the LmnanHG allele and lamins A and C from the wild-type allele (Figure (Figure1C).1C). The electrophoretic mobility of progerin in LmnanHG/+ MEFs was slightly slower than in LmnaHG/+ MEFs, which was predicted from the difference in farnesylation (Figure (Figure1C).1C). The point mutation in the LmnanHG allele was verified by DNA sequencing (Figure (Figure1D). 1D).
The absence of farnesylation in the progerin of LmnanHG/+ MEFs was verified by metabolically labeling with anilinogeranyl diphosphate (AGPP), a farnesyl pyrophosphate analog (25–27). The incorporation of the AGPP into proteins can be monitored by Western blotting with an antibody against the anilinogeranyl moiety of AGPP (Figure (Figure2A).2A). As expected, farnesylated progerin was detected in LmnaHG/+ but not LmnanHG/+ MEFs (Figure (Figure2A).2A). Lamins B1 and B2 are farnesylated proteins with an electrophoretic mobility identical to that of progerin (intermediate between lamins A and C). Thus, it was not surprising to observe a farnesylated protein between lamin A and lamin C in all of the MEFs (Figure (Figure2A).2A). However, following immunoprecipitation of the A-type lamins (i.e., lamin A, lamin C, and progerin), farnesylated progerin was observed only in LmnaHG/+ MEFs (Figure (Figure2A). 2A).
The farnesylation of progerin in LmnaHG/+ MEFs was abolished by treating the cells with an FTI (Figure (Figure2A).2A). Western blots with an antibody against HDJ-2 provided further evidence that the FTI was active (Figure (Figure2B). 2B).
LmnanHG/+ mice developed the same spectrum of disease phenotypes observed in LmnaHG/+ mice, but the disease phenotypes were less severe (14). Body-weight curves were abnormal in both male and female LmnanHG/+ mice, compared with wild-type control mice (P < 0.0001 for both males and females) (Figure (Figure3,3, A and B). However, the growth of LmnanHG/+ mice was substantially better than that of LmnaHG/+ mice (P < 0.0001 for both males and females) (Figure (Figure3,3, A and B).
Kaplan-Meier survival curves revealed that LmnanHG/+ mice lived longer than LmnaHG/+ mice (P < 0.0001 for both males and females) (Figure (Figure3C).3C). LmnaHG/+ mice develop progressive rib fractures near the costovertebral joints with increasing age (14); these same bone abnormalities were observed in LmnanHG/+ mice. However, the number of rib fractures at the time of death was lower in LmnanHG/+ mice than in LmnaHG/+ mice (P < 0.0001 for both male and female mice) (Figure (Figure3D),3D), even though the LmnanHG/+ mice were significantly older at the time of death (36.1 ± 0.69 weeks vs. 21.3 ± 0.84 weeks in LmnaHG/+ mice). LmnanHG/+ mice had more body fat than LmnaHG/+ mice (P < 0.0001 for both males and females) (Figure (Figure3E). 3E).
The reduced number of rib fractures and reduced kyphosis of the spine in LmnanHG/+ mice were evident on μCT scans (Figure (Figure4,4, A–F). The bone density and cortical thickness in the ribs were greater in LmnanHG/+ mice than in LmnaHG/+ mice (Figure (Figure4,4, G and H).
Intercrossing LmnanHG/+ mice yielded LmnanHG/nHG mice, which appeared normal for the first 2 weeks of life. By 4–8 weeks of age, however, male and female LmnanHG/nHG mice began to lose weight and exhibited reduced amounts of subcutaneous and abdominal fat. The LmnanHG/nHG mice (n = 11) had multiple rib fractures by 12 weeks of age, and all died by 17 weeks of age. In contrast, LmnaHG/HG mice die by 4 weeks of age (28). LmnanHG/nHG mice were more severely affected than either LmnaHG/+ or LmnanHG/+ mice.
A hallmark of LmnaHG/+ fibroblasts is misshapen nuclei. We suspected that misshapen nuclei might be less frequent in LmnanHG/+ MEFs than in LmnaHG/+ MEFs. To test this possibility, we scored nuclear shape abnormalities in primary MEFs from LmnanHG/+, LmnaHG/+, LmnanHG/nHG, LmnaHG/HG, and Lmna+/+ embryos (n = 3–8 cell lines/genotype) (Figure (Figure5A).5A). These experiments involved scoring nuclear shape in 4,000 cells per genotype. In 2 independent experiments, the percentage of misshapen nuclei was lower in LmnanHG/+ MEFs than in LmnaHG/+ MEFs (P < 0.0001) (Figure (Figure5A).5A). Similarly, the percentage of misshapen nuclei was lower in LmnanHG/nHG MEFs than in LmnaHG/HG MEFs (P < 0.0001) (Figure (Figure5A).5A). The spectrum of nuclear shape abnormalities differed in LmnanHG/+ and LmnaHG/+ MEFs; nuclei containing folds accounted for a greater percentage of the total number of misshapen nuclei in LmnanHG/+ and LmnanHG/nHG MEFs (P < 0.0001) (Figure (Figure5A). 5A).
Nuclear shape abnormalities in LmnanHG/+, LmnanHG/nHG, LmnaHG/+, and Lmna+/+ primary MEFs were examined in the presence or absence of an FTI (n = 4 cell lines/genotype; 4,000 cells/genotype counted) (Figure (Figure5B).5B). As expected, the FTI reduced the frequency of nuclear shape abnormalities in LmnaHG/+ MEFs (P < 0.0001), but it had no effect on LmnanHG/+ or LmnanHG/nHG MEFs (Figure (Figure5B). 5B).
Normally, mature lamin A in wild-type MEFs is concentrated at the rim of the nucleus (Figure (Figure5C).5C). In LmnanHG/nHG MEFs, we never observed a concentration of progerin at the nuclear rim (Figure (Figure5D).5D). In many cells, the nonfarnesylated progerin was highly concentrated in nuclear folds and blebs (Figure (Figure5,5, D and E).
Disease phenotypes were less severe in LmnanHG/+ mice than in LmnaHG/+ mice. To explore the potential explanation for this difference, we used Western blots to gauge progerin levels in LmnaHG/+ and LmnanHG/+ mice. The levels of progerin relative to actin were lower in LmnanHG/+ tissues (heart, liver, kidney, skull) than in LmnaHG/+ tissues (P < 0.0001) (Figure (Figure6,6, A–D and F). Similarly, progerin levels relative to actin were lower in LmnanHG/+ MEFs than in LmnaHG/+ MEFs (P < 0.0001) (Figure (Figure6,6, E and F). The lower levels of progerin in LmnanHG/+ MEFs and tissues raised the possibility that FTI treatment might reduce the steady-state levels of progerin in LmnaHG/+ MEFs and LmnaHG/+ mice. Indeed, this was the case. Treatment of LmnaHG/+ MEFs (n = 4 independently isolated cell lines) with an FTI reduced progerin levels relative to actin (P < 0.0001) (Figure (Figure7,7, A and B). The FTI led to the appearance of nonfarnesylated prelamin A, as expected, and also reduced the levels of mature lamin A. The reduced levels of mature lamin A in FTI-treated MEFs were not surprising; previously, Toth et al. (20) showed that 2 chemically distinct FTIs markedly reduced the levels of mature lamin A in wild-type MEFs.
We also analyzed progerin levels in the hearts of LmnaHG/+ mice that had been treated with an FTI (ABT-100) for 3 or more months. The levels of progerin (relative to actin) were lower in hearts of FTI-treated LmnaHG/+ mice than in the hearts of mice treated with vehicle alone. This difference was significant (P < 0.0001) (Figure (Figure7,7, C and D) and was observed in 2 independent experiments.
To examine the mechanism for the lower levels of progerin in LmnanHG/+ MEFs and tissues, we measured progerin transcript levels in LmnanHG/+ and LmnaHG/+ tissues. Progerin mRNA levels, as judged by RT-PCR, were not different in the 2 mouse models (Figure (Figure8,8, A and B). The fact that progerin levels were lower in LmnanHG/+ MEFs and tissues (Figure (Figure6)6) despite similar progerin transcript levels implied that the absence of farnesylation might accelerate progerin turnover. To test this possibility, we examined the disappearance of progerin in LmnanHG/+ and LmnaHG/+ MEFs after labeling with [35S]methionine. In the span of 4 days after labeling of the cells, the progerin/lamin C ratio in LmnanHG/+ MEFs fell by 55.6%, while the progerin/lamin C ratio in LmnaHG/+ MEFs fell by 29.2%. We also examined by Western blot the progerin/actin ratios in LmnanHG/+ and LmnaHG/+ MEFs before and after inhibition of Lmna expression with a mouse prelamin A–specific antisense oligonucleotide (29). Two days after the second oligonucleotide transfection, the progerin/actin ratio fell by 39.4% in LmnaHG/+ MEFs and by 58.9% in LmnanHG/+ MEFs. These observations are consistent with the lower steady state levels of progerin in LmnanHG/+ MEFs and tissues (Figure (Figure6). 6).
In 2 independent studies (14, 16), an FTI ameliorated disease phenotypes in LmnaHG/+ mice. However, the amelioration of disease was far from complete, leading us to consider the possibility that the nonfarnesylated form of progerin might itself be toxic and capable of eliciting disease. In the current study, we created a new line of HGPS-knockin mice expressing a nonfarnesylated version of progerin (LmnanHG/+). Aside from a cysteine-to-serine substitution in the CaaX motif, LmnanHG/+ mice were genetically identical to LmnaHG/+ mice (13). The absence of progerin farnesylation in LmnanHG/+ cells was verified with metabolic labeling experiments. Interestingly, the LmnanHG/+ mice developed the same spectrum of disease phenotypes found in LmnaHG/+ mice (i.e., slow weight gain, rib fractures, loss of body fat, reduced survival). Thus, the principal message of this study is that a nonfarnesylated version of progerin is still toxic. However, the extent of disease phenotypes in LmnanHG/+ mice was milder than in LmnaHG/+ mice. Also, the disease phenotypes in LmnanHG/nHG mice were milder than those in LmnaHG/HG mice (14). Thus, the second message of this study is that disease phenotypes are milder in the knockin mice expressing a nonfarnesylated version of progerin. The finding of less severe phenotypes in LmnanHG/+ mice is consistent with the results of earlier FTI treatment studies (14, 16).
Both LmnanHG/+ and LmnaHG/+ mice produced progerin, and they were genetically identical — except at a single codon. However, the severity of the phenotypes was reduced in LmnanHG/+ mice. We reasoned that one explanation might be lower steady-state levels of progerin in LmnanHG/+ mice. Indeed, the levels of progerin, both in MEFs and tissues, were significantly lower in LmnanHG/+ mice than in LmnaHG/+ mice despite similar levels of progerin transcripts, and follow-up studies in MEFs pointed to a more rapid disappearance of nonfarnesylated progerin. Reduced levels of progerin in the absence of farnesylation were not entirely surprising, as there are precedents for the absence of prenylation causing reduced steady-state levels of other CaaX proteins (30, 31). It is quite plausible that the different steady-state levels of progerin in LmnanHG/+ and LmnaHG/+ mice could explain the phenotypic differences in the 2 models. Higher levels of progerin expression are associated with more nuclear shape abnormalities (11), and rare HGPS mutations yielding higher-than-usual progerin levels are associated with particularly severe forms of progeria (32). Also, we found that LmnaHG/LCO mice (compound heterozygotes with 1 LmnaHG allele and 1 “lamin C–only” allele) have lower steady-state levels of progerin than LmnaHG/+ mice, despite identical levels of progerin transcripts. The reduced progerin levels in LmnaHG/LCO mice were accompanied by reduced disease phenotypes and a reduced frequency of misshapen nuclei (28).
Given the very positive effects of FTIs on nuclear shape in HGPS cells (13, 20, 21, 33, 34), many will probably be quite surprised by the fact that nonfarnesylated progerin is toxic. However, viewed from the perspective of the human genetics of progeroid syndromes, one could argue that the findings are not particularly shocking. Very rare missense mutations in lamin A and lamin C — abnormalities that do not affect the carboxyl-terminal processing of prelamin A — can cause progeroid disorders in humans. For example, E145K and S143F mutations in lamins A/C clearly elicit progeria-like disease phenotypes (1, 35), and compound heterozygosity for T528M and M540T substitutions in lamins A/C causes a progeroid syndrome that is similar to HGPS (36). Why these subtle structural variations cause progeroid syndromes rather than other “lamin A/C diseases” such as lipodystrophy or muscular dystrophy is not known, but at face value these genetic observations suggest that protein farnesylation is not “the entire story” in the pathogenesis of progeroid syndromes. Our current studies with nonfarnesylated progerin reinforce that view.
In our study, we eliminated protein farnesylation with the standard approach, which is to replace the cysteine of the CaaX motif with a serine (22–24). The potential impact of a single amino acid substitution in lamin A or lamin C must be considered carefully, given that single amino acid substitutions in these proteins can elicit disease in humans (4). Indeed, one could argue that our conclusions require a caveat. The nonfarnesylated progerin in LmnanHG/+ mice contains a missense mutation. One might conceivably argue — and it is clearly a formal possibility — that nonfarnesylated progerin with a cysteine in the CaaX motif (as would occur during FTI treatment of LmnaHG/+ mice) is 100% innocuous but that a nonfarnesylated progerin with the cysteine-to-serine substitution (as in LmnanHG/+ mice) is toxic and elicits disease phenotypes similar to those associated with farnesylated progerin in untreated LmnaHG/+ mice. According to this reasoning, it is the point mutation in the LmnanHG allele and not the 50–amino acid deletion that causes disease. There is no good experimental way to exclude this possibility, but this seems quite unlikely. First, the 50–amino acid deletion is a far more substantial structural alteration than a cysteine-to-serine substitution. Second, there is no precedent for a heterozygous Lmna missense mutation eliciting disease phenotypes in mice. Heterozygosity for L530P, H222P, and N195K mutations causes severe disease in humans (37–39) but no disease phenotypes in mice (40–42).
Based on the efficacy of FTI treatment in LmnaHG/+ and Zmpste24–/– mice (14, 16, 17), a clinical trial of FTI therapy is under way in children with HGPS (18, 19). What are the implications of our current studies for the treatment of HGPS? On the positive side, our current studies are consistent with our FTI treatment results (14, 16, 17) and suggest that reducing progerin farnesylation might be useful. On the other hand, our current studies suggest that the impact of an FTI could be limited. Despite a total absence of progerin farnesylation, LmnanHG/+ mice still developed severe disease phenotypes. Thus, while FTIs may well ameliorate some phenotypes in children with HGPS, we doubt that FTIs will be a panacea, and we are skeptical that other strategies to interfere with protein prenylation, such as bisphosphonates (20), will be any better. We believe that it will be important to continue to develop other therapeutic strategies, such as approaches to reduce the alternative splicing event that lies at the root of the disease (12), or to eliminate prelamin A transcripts with antisense approaches (29, 43).
The targeting vector for the LmnanHG allele is identical to that of the LmnaHG allele produced previously (13), except for a point mutation in exon 12 (in the 5′ arm of the vector) that changes the cysteine of the CaaX motif to a serine. The mutation was introduced into a plasmid containing the 5′ arm of the vector by site-directed mutagenesis with the QuikChange kit (Stratagene) with primer 5′-CAGAGCTCCCAGAACAGCAGCATCATGTAATCT-3′ (the single-base change is underlined) and a complementary reverse primer. The integrity of the completed gene-targeting vector was verified by restriction endonuclease digestion (the point mutation eliminates a PstI site) and DNA sequencing (13).
Targeted 129/OlaHsd ES cell clones carrying the LmnanHG allele were identified by Southern blotting with EcoRI-cleaved genomic DNA and a 348-bp 5′ flanking probe (13). The probe detects a 10.4-kb band in the wild-type Lmna allele and a 9.3-kb band in the LmnanHG allele. To produce chimeric mice, targeted ES cells were microinjected into C57BL/6 blastocysts. Chimeras were bred with C57BL/6 females to generate LmnanHG/+ mice. Chimeras generated from LmnaHG/+ ES cells were used to generate LmnaHG/+ mice. Because LmnanHG/+ and LmnaHG/+ mice were bred from chimeras, they were genetically identical (with 1 C57BL/6 chromosome and 1 129/OlaHsd chromosome) except for the point mutation in exon 12 of the LmnanHG allele. LmnanHG/+ mice were intercrossed to obtain LmnanHG/nHG mice.
Genotyping of mice was performed by PCR with genomic DNA from tail biopsies and oligonucleotide primers 5′-TGAGTACAACCTGCGCTCAC-3′ and 5′-CAGACAGGAGGTGGCATGT-3′. The wild-type allele yields a 582-bp product, whereas the mutant allele yields a 186-bp product (13).
In some experiments, LmnaHG/+ mice were treated with an FTI ABT-100 (14). ABT-100 was kindly provided by David J. Frost (Abbott). The FTI was mixed in drinking water containing 0.4% hydroxy propyl methyl cellulose and 1.0% ethanol at a concentration of 0.3 mg/ml, so as to deliver a dose of approximately 39 mg/kg/d. The vehicle control consisted of drinking water with 0.4% hydroxy propyl methyl cellulose and 1.0% ethanol (14).
All mice were fed a chow diet and housed in a virus-free barrier facility with a 12-hour light/12-hour dark cycle. Body weights were assessed weekly. All animal studies were approved by UCLA’s Animal Research Committee.
Primary MEFs were prepared from E13.5 embryos (10). Early-passage MEFs were plated in 6-well plates and grown to 75% confluency. In some experiments, protein farnesylation was blocked by incubating the cells with an FTI (10 μM ABT-100) for 48 hours.
MEFs were incubated for 36 hours with an analogue of farnesol, 8-anilinogeraniol (AG) (30 μM in DMSO) (25), in the presence and absence of an FTI (ABT-100, 5 μM). After entering cells, AG is incorporated into AGPP, which is used by protein FTase as a substrate for the modification of CaaX proteins such as progerin. The AG moiety of AGPP is transferred onto cellular proteins by FTase, competitive with endogenous pools of farnesyl diphosphate (25). AG incorporation into endogenous cellular proteins was detected by Western blotting with a mouse monoclonal antibody specific for AG, diluted 1:5000 (25).
MEFs were washed with PBS and harvested by scraping, then pelleted by centrifugation. The pellet was resuspended in 0.3 ml ice-cold lysis buffer (50 mM Tris, 0.15 M NaCl, 1 mM EDTA, 1% NP-40, 2.5 mg/ml deoxycholate, 1 mM PMSF, and 0.1% SDS in PBS) containing Complete Mini Protease Inhibitors (Roche) at 1× concentration, and incubated on ice for 10 minutes. After removing insoluble material by centrifugation, cell extracts were precleared by incubation with 25 μl of protein G–agarose (50% suspension in lysis buffer) (Roche) at 4°C for 30 minutes. Goat anti–lamin A/C antibody (15 μl) (Santa Cruz Biotechnology Inc.) was added to the precleared supernatant fluids and incubated at 4°C overnight with constant rotation. Immune complexes were isolated by the addition of protein G–agarose and incubation at 4° C for 2 hours, followed by centrifugation. The agarose gel was washed 5 times with ice-cold lysis buffer, and the bound proteins were released by boiling for 3 minutes in 2× sample loading buffer (Invitrogen). The agarose gel was removed by centrifugation, and the supernatant fluids were analyzed by SDS-polyacrylamide gel electrophoresis and Western blotting.
LmnanHG/+ and LmnaHG/+ MEFs were grown in methionine-deficient medium (Invitrogen) and 10% fetal bovine serum (dialyzed against PBS) at 37°C overnight and then labeled with [35S]methionine (200 μCi/ml; MP Biomedicals) for 24 hours. After labeling, cells were grown in standard DMEM/10% fetal calf serum. MEFs were harvested after the labeling period and daily for 4 days. Lamin A/C, progerin, and actin were immunoprecipitated and size fractionated by SDS-PAGE. The intensities of the progerin, lamin C, and actin bands were quantified with a Typhoon 9410 Phosphorimager (GE Healthcare), and the progerin/lamin C ratio was calculated.
Urea-soluble extracts were prepared from early-passage MEFs and tissues as described previously (44). Freshly harvested mouse tissues (~150 mg) were snap-frozen in liquid nitrogen and then ground into a powder with a mortar and pestle; the powder was resuspended in 0.5 ml of urea-solubilization buffer (44) and ground in a glass tissue grinder for 2 minutes. Finally, the samples were sonicated and centrifuged at 14,000 g for 10 minutes. The extracts were size fractionated on 4%–12% gradient polyacrylamide Bis-Tris gels (Invitrogen) and the separated proteins transferred to nitrocellulose membranes for Western blotting. The membranes were incubated in Odyssey blocking solution for 1 hour at room temperature before the addition of primary antibodies. The antibody dilutions were 1:400 for anti-lamin A/C rabbit IgG (sc-20681; Santa Cruz Biotechnology Inc.), 1:500 anti–HDJ-2 mouse IgG (NeoMarkers), 1:5000 for anti-AG mouse IgG, and 1:1000 for anti-actin goat IgG (sc-1616, Santa Cruz Biotechnology Inc.). The membranes were washed 3 times for 10 minutes each time in PBS containing 0.1% Tween 20 and then incubated with 1:5000 IRDye 700 anti-mouse IgG antibody and 1:5000 IRDye 800 anti-goat IgG antibody (Rockland Immunochemicals) in Odyssey Blocking buffer. The IR-coupled antibodies were detected on an Odyssey infrared imaging scanner and quantified according to the manufacturer’s instructions.
Western blots were performed on LmnanHG/+ and LmnaHG/+ MEFs after partial inhibition of Lmna expression with a mouse prelamin A–specific antisense oligonucleotide (29). Primary MEFs were plated on 6-well tissue culture plates and allowed to adhere overnight. The cells were washed with reduced serum Opti-MEM I medium (Invitrogen) and transfected with 30 nM of the mouse prelamin A–specific antisense oligonucleotide (ISIS 359445) or a control oligonucleotide (29) using 10 μl/ml Lipofectin (Invitrogen) at 37°C for 5 hours. The medium was then replaced with fresh culture medium (DMEM/10% FBS). The transfection was repeated the next day, and the cells were cultured for an additional 48 hours and then processed for Western blotting. The amount of progerin relative to lamin C was quantified with a LI-COR Odyssey scanner.
Early-passage MEFs (matched for passage number) were grown on coverslips, fixed in 3% paraformaldehyde, permeabilized with 0.2% Triton X-100, and blocked with PBS containing 10% fetal bovine serum and 0.2% bovine serum albumin (15). Cells were incubated for 1 hour with antibodies against lamin A (1:200, sc-20680; Santa Cruz Biotechnology Inc.) and LAP2β (1: 200, BD Biosciences — BD Transduction Laboratories). After washing, cells were incubated with 1:800 Cy3-conjugated anti-rabbit antibody (Jackson ImmunoResearch Laboratories) and 1:600 Alexa Fluor 488–conjugated anti-mouse antibody (A21202; Molecular Probes) and DAPI to visualize DNA. Images were obtained on an Axiovert 200 MOT microscope (Zeiss) with a ×63/1.25 oil immersion objective and processed with AxioVision 4.2 software (Zeiss). Nuclear shape (4,000 cells per genotype) was assessed in blinded fashion as described previously (13, 20, 28, 29).
Lmna+/+, LmnaHG/+, and LmnanHG/+ mice were euthanized, and the major fat pads (reproductive, inguinal, and mesenteric) were isolated and weighed.
The isolation of RNA, cDNA synthesis, and RT-PCR was performed exactly as described previously (28). The band intensities of the PCR fragments generated from the wild-type Lmna allele and the LmnanHG/+ and LmnaHG/+ alleles were quantified with the Odyssey Infrared Imaging System and the results normalized to Arbp.
Lmna+/+, LmnanHG/+, LmnanHG/nHG, and LmnaHG/+ mice were examined by compact cone-beam tomography (μCT 40 scanner; SCANCO Medical). Scans were performed in the axial plane mounted in a cylindrical sample holder at medium resolution with a current of 143 μA and a voltage of 55 kVp.
Bone density of isolated rib segments was determined by μCT scanning in a cylindrical sample holder tube with a voltage of 55 kVp and current of 143 μA. Three-dimensional images were reconstructed in 2048 × 2048 pixel matrices with a standard convolution back-projection procedure with a Shepp-Logan filter, and a cortical volume of interest was defined for each measurement. The mineralized tissue was segmented with a threshold of 120 for all samples. A manufacturer-provided hydroxyapatite phantom of known density was used to calibrate the mean density of bone volume and the cortical thickness.
Body weight curves were compared with repeated-measures ANOVA and the log-rank test. The bone density, number of rib fractures, and fat-pad weights were compared with the 2-tailed Student’s t test. Means and SEM are shown. Differences in the percentages of misshapen nuclei were assessed by the χ2 test. The number of surviving male and female mice was recorded weekly and expressed as a percentage of the total number of mice. The significance of differences was determined by the Kaplan-Meier method.
We thank B. Davies, C. Coffinier, and E. Farber for helpful discussions. We thank D. Frost for ABT-100. This study was supported by NIH grants HL76839, CA099506, and AR050200 (to S.G. Young), HL086683 (to L.G. Fong), and GM66152 to (H.P. Spielmann) and a grant from the Kentucky Lung Cancer Research Program. S.H. Yang was supported by a postdoctoral fellowship grant from the American Heart Association, Western States Affiliate, and a fellowship from the Vascular Biology Program at UCLA (supported by the NIH).
Nonstandard abbreviations used: AG, 8-anilinogeraniol; AGPP, anilinogeranyl diphosphate; FTase, farnesyltransferase; FTI, FTase inhibitor; HGPS, Hutchinson-Gilford progeria syndrome; MEF, mouse embryonic fibroblast.
Conflict of interest: The authors have declared that no conflict of interest exists.
Citation for this article: J. Clin. Invest. 118:3291–3300 (2008). doi:10.1172/JCI35876