SEARCH

SEARCH BY CITATION

Keywords:

  • Embryonic stem cells;
  • Differentiation;
  • Pluripotency;
  • Lamin A;
  • Lamin B;
  • Nuclear lamina

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Nuclear lamins comprise the nuclear lamina, a scaffold-like structure that lines the inner nuclear membrane. B-type lamins are present in almost all cell types, but A-type lamins are expressed predominantly in differentiated cells, suggesting a role in maintenance of the differentiated state. Previous studies have shown that lamin A/C is not expressed during mouse development before day 9, nor in undifferentiated mouse embryonic carcinoma cells. To further investigate the role of lamins in cell phenotype maintenance and differentiation, we examined lamin expression in undifferentiated mouse and human embryonic stem (ES) cells. Wide-field and confocal immunofluorescence microscopy and semiquantitative reverse transcription–polymerase chain reaction analysis revealed that undifferentiated mouse and human ES cells express lamins B1 and B2 but not lamin A/C. Mouse ES cells display high levels of lamins B1 and B2 localized both at the nuclear periphery and throughout the nucleoplasm, but in human ES cells, B1 and B2 expression is dimmer and localized primarily at the nuclear periphery. Lamin A/C expression is activated during human ES cell differentiation before downregulation of the pluripotency marker Oct-3/4 but not before the downregulation of the pluripotency markers Tra-1-60, Tra-1-81, and SSEA-4. Our results identify the absence of A-type lamin expression as a novel marker for undifferentiated ES cells and further support a role for nuclear lamins in cell maintenance and differentiation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Nuclear lamins are intermediate filament proteins that were originally identified as components of the nuclear lamina [1, 2], a 10- to 50-nm-thick meshwork of proteins that underlies the inner nuclear membrane [3]. It had traditionally been thought that the nuclear lamina functions primarily to provide structural support and organization to the nuclear envelope [410], but recent reports have documented that lamins exist throughout the nucleoplasm, suggesting structural contributions to a nuclear matrix [1113] and roles in DNA replication [14], transcription [15], and RNA processing [1618]. Mutations in lamins have also been assigned a causative role in the group of diseases collectively called laminopathies [19, 20].

In mammals, seven lamin proteins, which are the products of three genes, have been identified and can be divided into two subtypes. B-type lamins include B1, encoded by the LMNB1 gene, and B2 and B3, splice variants of the LMNB2 gene [2123]. A-type lamins A, AΔ10, C, and C2 are all splice variants of the LMNA gene [2426]. Lamins B3 and C2 are expressed only in male germ cells [22, 25]. The remaining B-type lamins are present ubiquitously in embryonic and adult cell types. At the cellular level, lamins B1 and B2 are essential for cell growth and survival [27]. lmnB1 knockout mice exhibit abnormal lung and bone development and die at birth [28]. Contrastingly, A-type lamins are primarily found in differentiated cells. During mouse embryonic development, expression of A-type lamins is first detected on day 9 in extraembryonic tissues and on day 12 in the embryo itself [2933]. Embryonic carcinoma cells generally express little or no A-type lamins [30, 34]. Lastly, certain adult cell types that are not fully differentiated also express little or no A-type lamins [29, 35]. Consistent with these findings, lamin A/C is not essential for cell growth and survival [27]. Although lmna knockout mice exhibit normal development at birth, they soon develop retarded postnatal growth and a muscular dystrophy phenotype that results in death by 8 weeks of age [4]. The detection of A-type lamins primarily in differentiated cells suggests that they might function to limit developmental plasticity by maintaining a cell's differentiated phenotype [29].

Embryonic stem (ES) cells are obtained from the inner cell mass of blastocyst-stage embryos and exhibit unlimited proliferative capacity and pluripotency [36, 37]. The ability of ES cells to differentiate into all cell types of the embryo suggests that they are ideal candidates for therapeutic use. However, the specific differentiation of ES cells toward desired lineages remains ineffective, resulting in mixtures of differentiated cells. Further, due to the proliferative capacity of ES cells and their tendency to form teratomas when transplanted into animals [38, 39], it is important to ensure that differentiated cells do not revert back to an ES cell phenotype after transplantation. Such reversion has been reported for mouse muscle–derived stem cells after differentiation into hematopoietic lineages [40] and for drosophila spermatogonial stem cells that have differentiated into spermatogonia [41]. Thus, before clinical applications, it is necessary to understand the mechanisms controlling ES cell differentiation and the maintenance of the differentiated state. Given the possible role of nuclear lamins in cell maintenance and differentiation, we decided to characterize their expression in mouse ES (mES) and human ES (hES) cells. We report that 129/SVEV mES cells and HSF-6 hES cells express lamins B1 and B2 but not lamin A/C and that as HSF-6 hES cells begin to differentiate into neuronal lineages and cardiomyocytes, lamin A/C expression is activated. Our results demonstrate that lamin A/C expression is a reliable marker of ES cell differentiation.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Cell Culture

mES Cells

129/SVEV mES cells (Specialty Media, Phillipsburg, NJ, http://www.specialtymedia.com) were maintained in an undifferentiated state by coculture with mitomyocin-treated (Sigma-Al-drich, St. Louis, http://www.sigmaaldrich.com) SNL mouse embryonic fibroblast (MEF) feeder cells (gift from Dr. Richard Chaillet, University of Pittsburgh) in Dulbecco's modified Eagle's medium (DMEM) with 15% fetal bovine serum, 1% non-essential amino acids, 1 mM L-glutamine, 0.1 mM β-mercaptoethanol, 1% nucleosides (all from Specialty Media), and 105 U/ml leukemia inhibitory factor (Chemicon, Temecula, CA, http://www.chemicon.com). Cells were passaged every 2 to 3 days, and media were changed daily.

hES Cells

HSF-6 hES cells (from University of California San Francisco, under license) were maintained in an undifferentiated state by coculture with mitomyocin-treated CF-1 MEF feeder cells in DMEM medium with 20% knockout serum replacer, 1% non-essential amino acids, 1 mM L-glutamine, and 4 ng/ml fibroblast growth factor-2 (all from Invitrogen, Carlsbad, CA, http://www.invitrogen.com). Cells were passaged every 5 to 7 days, and media were changed daily. WI-38 human embryonic lung fibroblasts were cultured in DMEM with 15% fetal bovine serum and 2 mM L-glutamine. Neuronal lineages and cardiomyocytes were obtained by culture of HSF-6 hES for 5 weeks on CF-1 MEF feeder cells without passage. Media were changed every 2 days.

RNA Isolation and Reverse Transcription–Polymerase Chain Reaction

The centers of undifferentiated HSF-6 colonies, cultured for 5 days under optimal conditions, were gently scraped with pulled Pasteur pipettes (Figs. 1A, 1B). Whole colonies of HSF-6 cells, which had begun to naturally differentiate after 10 days in culture, were scraped as well. WI-38 cells were trypsinized. All cells were spun at × 200 g for 5 minutes and then washed twice with phosphate-buffered saline (PBS). Cell pellets were immediately frozen in liquid nitrogen and stored at −80°C. RNA was isolated using RNAqueous 4-PCR (Ambion, Austin, TX, http://www.ambion.com), and cDNA was generated using the Reverse Transcription System (Promega, Madison, WI, http://www.promega.com) following the manufacturer's instructions. Spectrophotometry was used to verify quantity and quality of RNA and cDNA. Semiquantitative polymerase chain reaction (PCR) was performed with 50 ng of cDNA in a final concentration of 0.02 U Biolase DNA polymerase, 1 × NH4 buffer, 2.5 mM MgCl2 (Bioline, Randolph, MA, http://www.bioline.com), 0.2 mM dNTPs (Roche Diagnostics, Basel Switzerland, http://www.roche-applied-science.com), and 0.2 μM primers (Invitrogen). PCR conditions were 95°C for 3 minutes followed by 35 cycles at 95°C for 35 seconds, 55°C for 40 seconds, and 72°C for 45 seconds and a final extension at 72°C for 7 minutes. Human-specific primers used for each gene were as follows:

thumbnail image

Figure Figure 1.. Detection of lamin mRNA in human embryonic stem (hES) cells. HSF-6 hES cells were cultured for 5 days after passage, and the centers of colonies were scraped to obtain a pure population of undifferentiated HSF-6 hES cells for RNA isolation (A, B). Whole HSF-6 hES cell colonies cultured for 10 days after passage, consisting of undifferentiated and differentiated cells, were also scraped for RNA isolation (not shown). (C): Reverse transcription–polymerase chain reaction of WI-38 fibroblasts (lane 1), HSF-6 center colony scrapings (CCS) (lane 2), and HSF-6 whole colony scrapings (WCS) (lane 3).

Download figure to PowerPoint

Lamin A/C forward: AATGATCGCTTGGCGGTCTAC, Lamin A/C reverse: CTTCTTGGTATTGCGCGCTTT (255 bp);

Lamin B1 forward: AAAAGACAACTCTCGTCGCAT, Lamin B1 reverse: CCGCTTTCCTCTAGTTGTACG (256 bp);

Lamin B2 forward: ATTCAGAATCCAGGCGTCGAC, Lamin B2 reverse: TTATTGTTGTGACAGGTCTTACGA CG (339 bp);

B-actin forward: TGGCACCACACCTTCTACAATGAGC, B-actin reverse: GCACAGCTTCTCCTTAATGTCACGC (400 bp);

Oct-4 forward: CGRGAAGCTGGAGAAGGAGAAGCTG, Oct-4 reverse: AAGGGCCGCAGCTTACACATGTTC (230 bp).

Wide-Field and Confocal Microscopy and Immunocytochemistry

hES and mES cells were passaged 4 to 5 days and 2 days, respectively, before fixation. Cells were fixed with 100% methanol for 20 minutes at −20°C or 2% formaldehyde for 45 minutes. Cells were subsequently washed twice with PBS, per-meabilized with 0.1% Triton X-100, and incubated for 1 hour at 25°C with combinations of the following primary antibodies: monoclonal mouse anti-human Oct-4 (1:50), polyclonal rabbit anti-human lamin A/C (1:100) (both from Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com), monoclonal mouse anti-human lamin B1 (1:50) and B2 (1:50) (both from Zymed Laboratories, San Francisco, http://www.zymed.com), monoclonal mouse anti–β-III tubulin (1:250), mouse antinestin (1:200) (both from Covance, Berkeley, http://www.covance.com), mouse anti-sarcomeric α-actinin (EA-53) (1:100) (from Sigma), mouse anti–SSEA-4, mouse anti–TRA-160, mouse anti–TRA-180, and mouse anti-light meromyosin (MF-20) (1:3) (all obtained from the Developmental Studies Hybridoma Bank under the auspices of the NICHID, University of Iowa, Department of Biological Sciences, Iowa City). Cells were washed three times with PBS, and the primary antibodies were detected with Alexa Fluor 488, Alexa Fluor 546, and Alexa Fluor 568 secondary antibodies (all 1:100) (all from Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com) for 1 hour at 25°C. DNA was visualized with TOTO-3 (1:1,000) (Molecular Probes), which was added to the secondary antibody mix. Cells were washed three times in PBS and mounted onto glass slides with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, http://www.vectorlabs.com). Confocal images were captured with a Leica TCS SP2 using proprietary Leica software.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Undifferentiated mES and hES Cells Do Not Express A-Type Lamins

Semiquantitative reverse transcription (RT)-PCR analysis with human-specific primers was used to examine lamin A/C expression levels in HSF-6 hES cells. The centers of several HSF-6 hES cell colonies were scraped 5 days after passage to isolate undifferentiated cells and exclude peripheral cells that might be differentiating (Figs. 1A, 1B). Whole colonies were also scraped 10 days after passage to examine whether lamin A/C expression is activated in some peripheral cells. Using this method, we show that HSF-6 hES cells, obtained by center colony scraping and that are undifferentiated by morphological appearance (Fig. 1A), express Oct-4 but little or no detectable lamin A/C (Fig. 1C, lane 2). In contrast, differentiated WI-38 control cells express lamin A/C (Fig. 1C, lane 1). However, lamin A/C was detected in whole colony scrapings (Fig. 1C, lane 3), suggesting that lamin A/C expression is activated in some peripheral ES cells that are, presumably, differentiating.

Immunocytochemistry experiments corroborated the RT-PCR results, demonstrating a lack of lamin A/C in HSF-6 hES cells and 129SV/EV mES cells (Fig. 2). The pluripotency markers Oct-4, [37, 38, 42], TRA-1-60 [37, 43], TRA-1-81 [37, 43], and SSEA-4 [37, 44] were used to identify undifferentiated cells along with morphological observations (undifferentiated ES cells are relatively small [8- to 10-μm diameter] with a high nucleus/cytoplasm ratio). Undifferentiated HSF-6 hES cells, which are positive for Oct-4 (Fig. 2B), TRA-1–60 (Fig. 2E), TRA-1-81 (Fig. 2H), and SSEA-4 (Fig. 2K), lack lamin A/C (Figs. 2C, 2F, 2I, 2L), but adjacent CF-1 MEF feeder cells that are negative for each of the pluripotency markers are lamin A/C positive. Similar results were obtained with the H1 hES cell line (not shown). Furthermore, 129/SVEV mES cells, as identified by Oct-4 labeling (Fig. 2N), do not express lamin A/C (Fig. 2O). In contrast, differentiated SNL MEF feeder cells adjacent to the mES cells that are negative for Oct-4 labeling contain lamin A/C. Collectively, these observations indicate that undifferentiated ES cells do not express lamin A/C.

thumbnail image

Figure Figure 2.. Human embryonic stem (hES) and mouse embryonic stem (mES) cells do not contain lamin A/C. HSF-6 hES cells were fixed with 100% methanol (A–C) or 2% formaldehyde (D–L) 4 to 5 days after passage. 129/SVEV mES cells were fixed with 100% methanol (M–O) 2 days after passage. Cells were colabeled for lamin A/C (C, F, I, L, O) and the pluripotency markers Oct-4 (B, N), TRA-160 (E), TRA-180 (H), and SSEA-4 (K). Undifferentiated HSF-6 hES cells growing in a colony are positive for Oct-4 (B), TRA-160 (E), TRA-180 (H), and SSEA-4 (L) but negative for lamin A/C (C, F, I, L). Conversely, adjacent differentiated CF-1 mouse embryonic fibroblast (MEF) feeder cells are negative for the pluripotency markers but positive for lamin A/C. Undifferentiated 129/SVEV mES cells within a colony are positive for Oct-4 (N) but negative for lamin A/C (O), whereas adjacent differentiated SNL MEF feeder cells are negative for Oct-4 but positive for lamin A/C. DNA was visualized with TOTO-3 (A, D, G, J, M). Representative MEF feeder cells are identified with arrows. Images were captured using confocal microscopy. Scale bar = 50 μm.

Download figure to PowerPoint

Undifferentiated mES and hES Cells Express B-Type Lamins

Next, we used RT-PCR analysis to investigate the expression of lamins B1 and B2 in HSF-6 hES cells and report that HSF-6 hES cells express lamin B1 (Fig. 1C, lane 2), but under the conditions used, we could not detect the presence of lamin B2 mRNA (Fig. 1C, lane 2). Immunocytochemistry experiments showed that undifferentiated HSF-6 hES cells and 129/SVEV mES cells contain lamins B1 and B2 (Fig. 3). Because we previously demonstrated that undifferentiated HSF-6 hES cells and 129/SVEV mES cells do not express lamin A/C, we used this feature along with morphological characteristics to identify undifferentiated ES cells. Undifferentiated HSF-6 hES cells that are lamin A/C negative (Figs. 3B, 3E) express lamins B1 (Fig. 3C) and B2 (Fig. 3F). Similarly, undifferentiated 129/SVEV mES cells that are lamin A/C negative (Figs. 3H, 3K) express lamin B1 (Fig. 3I) and lamin B2 (Fig. 3L). Interestingly, lamins B1 and B2 levels are low in HSF-6 hES cells relative to the differentiated CF-1 MEF feeder cells. Further, lamin B2 is localized primarily at the nuclear envelope (Fig. 4F), whereas lamin B1 demonstrates a peripheral and punctate labeling pattern throughout the nucleoplasm (Fig. 4C). These results indicate that mouse and human ES cells express both B-type lamins, although expression and/or protein levels may be lower than in differentiated cell types and localization may differ.

thumbnail image

Figure Figure 3.. Human embryonic stem (hES) and mouse embryonic stem (mES) cells contain lamin B1 and B2. HSF-6 hES cells were fixed with methanol 5 days after passage. Cells were colabeled for lamin A/C (B, E, H, K) in conjunction with either lamin B1 (C, I) or lamin B2 (F, L). Undifferentiated HSF-6 hES cells are negative for lamin A/C (B, E) but positive for lamin B1 (C) and B2 (F). Adjacent differentiated CF-1 MEF feeder cells are positive for lamin A/C, lamin B1, and lamin B2. Lamin B1 (F) and B2 (I) levels are low in hES cells relative to the MEF feeder cells. DNA was visualized with TOTO-3 (A, D, G, J). Representative MEF feeder cells are identified with arrows. Images were captured using confocal microscopy. Scale bar = 50 μm (A–C, G–J) or 5 μm (D–F, J–L).

Download figure to PowerPoint

thumbnail image

Figure Figure 4.. Lamins B1 and B2 localize primarily to the nuclear envelope of undifferentiated human embryonic stem (hES) cells. HSF-6 hES cells were colabeled for lamin A/C (B, E) and lamin B1 (C) or lamin B2 (F). Undifferentiated HSF-6 hES cells that are lamin A/C-negative (B, E) exhibit low levels of lamins B1 (C) and B2 (F). Lamin B2 is present primarily at the nuclear periphery (F). Lamin B1 is present at the periphery as well as in punctate structures throughout the nucleoplam. DNA was visualized with TO-TO-3 (A, E). Images were captured with confocal microscopy. Scale bar = 5 μm.

Download figure to PowerPoint

hES Cells Express A-Type Lamins Upon Differentiation to Neuronal Lineages and Cardiomyocytes

Our RT-PCR results from whole HSF-6 hES cell colony scrapings suggested that lamin A/C expression is activated in some peripheral hES cells, suggesting that expression is activated in differentiating cells. Further, some HSF-6 hES cell colonies that we examined exhibited peripheral cells that labeled for both Oct-4 (Fig. 5B) and lamin A/C (Fig. 5C) and probably represent cells that are beginning to differentiate. To investigate this further, we used immunocytochemistry to examine if lamin A/C expression is activated in neuronal lineages and cardiomyocytes derived from HSF-6 hES cells. Neuronal cells, as identified by either nestin (Fig. 5E) or β-III tubulin (Fig. 5F), labeled for lamin A/C (Figs. 5F, 5I). Similarly, beating cardiomyocytes, as identified by α-actinin (Fig. 5K) or light meromyosin (MF 20) (Fig. 5N), also demonstrated lamin A/C expression (Figs. 5L, 5O). Our results indicate that, upon differentiation to neuronal lineages and cardiomyocytes, hES cells begin to express lamin A/C. Furthermore, the detection of cells at the periphery of some HSF-6 hES cell colonies that label for both Oct-4 and lamin A/C suggests that lamin A/C expression is activated before downregulation of Oct-4 so that there is a transient overlap in expression. We did not observe similar colabeling of lamin A/C with Tra-1-60, Tra-1-81, and SSEA-4.

thumbnail image

Figure Figure 5.. Lamin A/C is activated upon human embryonic stem (hES) cell differentiation to neuronal lineages and cardiomyocytes. Some HSF-6 hES cell colonies contained peripheral cells that colabeled for Oct-4 (B) and lamin A/C (C) with varying intensity. HSF-6 hES cells were cultured for 5 weeks on CF-1 MEF feeder cells without passage to obtain neuronal lineages (D–I) and cardiomyocytes (J–O). Differentiated cells were labeled for lamin A/C (F, I, L, O) in conjunction with either nestin (E), β-III tubulin (H), α-actinin (K), or light meromyosin (MF-20) (N). DNA was labeled with TOTO-3 (A, D, G, J). Neuronal lineages, as identified by nestin (E) or β-III tubulin (H), are positive for lamin A/C (F, I). Cardiomyocytes, as identified by α-actinin (K) or light meromyosin (MF-20) (N), are positive for lamin A/C (L, O). Images were captured with confocal microscopy. Scale bars = 25 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Nuclear lamins are components of the nuclear envelope that are involved in nuclear structural support and organization [410], DNA replication [14], transcription [15], and RNA processing [16, 17]. B-type lamins are present in most cell types from the first zygotic cell division through adulthood [29]. Contrastingly, A-type lamins are detected primarily in cells that are differentiated, as demonstrated by the lack of expression for most mouse embryonic development [29], in various adult cells that are not fully differentiated [29] and in EC cell lines [30, 34]. The observation that B-type lamins are present in all cell types but that A-type lamins are present primarily in differentiated cells suggests that A-type lamins could function to maintain the differentiated state. To begin investigating this possibility, we examined lamin expression in embryonic stem cells and differentiated derivatives. Because ES cells are isolated from blastocyst-stage embryos and are pluripotent [36], we hypothesized that undifferentiated ES cells should express B-type but not A-type lamins and that as ES cells differentiate, lamin A/C expression would be activated.

Using semiquantitative RT-PCR with human-specific primers, we show that undifferentiated HSF-6 cells express lamin B1 (Fig. 1, lane 2), but lamin B2 (Fig. 1, lane 2) could not be detected using two different sets of primers (only one shown). Interestingly, immunocytochemistry analysis identified the presence of both lamin B1 (Figs. 3C, 3I) and B2 (Figs. 3F, 3L) in HSF-6 hES and 129/SVEV mES cells. However, HSF-6 hES cells seem to have lower B-type lamin levels relative to differentiated CF-1 MEF feeder cells. Further lamin B2 localization seems to be primarily peripheral, with little or no labeling throughout the nucleoplasm (Fig. 4F), whereas lamin B1 seems to be localized both to the periphery and in nuclear foci (Fig. 4C). This suggests that although B-type lamins are present in HSF-6 hES cells, the nuclear lamina structure at the periphery and throughout the nucleoplasm is relatively underdeveloped.

Our ability to detect lamin B2 protein using immunocytochemistry but inability to detect lamin B2 mRNA using RT-PCR in HSF-6 hES cells seems contradictory. Cross-reactivity of the lamin B2 antibody with lamin B1 is highly unlikely because it has been shown that the antibody does not detect lamin B1 in Western analysis [45], so we hypothesize two alternative explanations. First, lamin B2 mRNA might have a very short half-life, but sufficient protein accumulates for immunofluorescent detection. Second, hES cells might express a splice variant of lamin B2 that is recognized by the primary antibody but lacks the region amplified by the PCR primers. For example, lamin B3 is a significantly smaller splice variant of lamin B2 that so far has only been detected in male germ cells [22]. However, if such a splice variant is present in hES cells, it is unlikely to be lamin B3 because the region of lamin B2 amplified by the PCR primers is shared by lamin B3 [22]. Therefore, it is possible that hES cells express a unique, previously unidentified splice variant of lamin B2.

Both RT-PCR and immunofluorescence analysis demonstrate that HSF-6 hES cells (Fig. 1C, lane 2; Figs. 2C, 2F, 2I, 2L) and 129/SVEV mES cells (Fig. 2O) do not express lamin A/C. Further, our results show that lamin A/C expression is activated in neuronal lineages (Figs. 5F, 5I) and cardiomyocytes (Figs. 5L, 5O) differentiated from HSF-6 hES cells. This is consistent with previous reports that have detected lamin A/C only in differentiated cell types [2935]. Interestingly, cells at the periphery of some HSF-6 hES cell colonies, where spontaneous differentiation sometimes occurs (our observations), express both Oct-4 (Fig. 5B) and lamin A/C (Fig. 5C). The intensity of Oct-4 and lamin A/C labeling varied within individual peripheral cells that had both present, but no statistically significant relationship could be established; we observed cells that had high intensity labeling for both, low intensity labeling for both, or high intensity for one and low intensity for the other. The different combinations of Oct-4 and lamin A/C expression in individual cells suggests that either protein is not directly influenced by the other. Interestingly, different ratios of Oct-4 to lamin A/C could represent cells undertaking alternate differentiation pathways. Thus, it seems that lamin A/C is activated before complete Oct-4 downregulation during the in vitro differentiation process, suggesting that lamin A/C expression is an earlier indicator of ES cell differentiation. However, we did not observe similar colabeling of lamin A/C with Tra-1-60, Tra-1-81, or SSEA-4, suggesting that these pluripotency markers are downregulated before the activation of lamin A/C expression in vitro.

Taken together, our results identify the absence of A-type lamins as a novel marker of undifferentiated ES cells in culture and positive expression as an indicator of differentiation. The presence of B-type lamins but absence of A-type lamins in ES cells demonstrates another similarity between ES cell lines and the inner cell mass cells from which they are derived. However, during murine development, Oct-4 is downregulated in somatic tissues well before lamin A/C expression is detected in any cell type. We describe lamin A/C expression before Oct-4 down-regulation in HSF-6 hES cells, suggesting that in vitro differentiation of ES cell lines might not mimic the consecutive steps that lead to tissue specification during embryonic development in vivo. Lastly, our results are in accordance with a possible role for A-type lamins in the maintenance of the differentiated state.

How could A-type lamins maintain cells in a differentiated state? Once expressed, A-type lamins could “lock” the nucleus of differentiated cells in particular gene expression patterns. This would be a simple way to maintain transcriptional and phenotypic stability in specialized, terminally differentiated cells. A-type lamins could accomplish this by increasing nuclear rigidity, which could constrain chromatin remodeling. Several lines of evidence support this possibility. First, cells from patients with various LMNA mutations have an abnormally plastic nuclear morphology, marked by lobes and invaginations [9, 10]. Second, it has been directly shown that lamin A/C deficiency reduces nuclear structural integrity under mechanical strain [5]. Lastly, lamin A/C [46, 47] and lamin-associated proteins [48] can interact directly with chromatin by binding DNA directly or via DNA-associated proteins. Alternatively, lamin A/C could regulate the expression of specific genes that are involved in the maintenance of a differentiated state, as evidenced by observations that lamin A/C can interact with transcription factors such as MOK2 [49] and SREBP1 [50], as well as retinoblastoma, which is involved in regulating cell-cycle arrest and apoptosis [5153]. Lamin A/C might also have widespread effects on gene expression because proper lamin A/C organization may be essential for polymerase II–dependent transcription [15, 18] and has been found to colocalize with splicing factor speckles [1618]. The possibility should be considered that laminopathies arise because cells are unable to maintain their differentiated state as a result of a disorganized nuclear lamina caused by lamin A mutations. Cellular dysdifferentiation, previously postulated as a cause of cancer and aging [5456], would result in abnormal gene expression, reduced proliferation, and/or apoptosis. The devastating premature aging disease Hutchinson-Gilford Progeria Syndrome could be the ultimate example of such dysdifferentiation.

If A-type lamins maintain the differentiated state by “locking” cells in a specific gene expression pattern, whether directly through structural modification of the nuclear envelope or indirectly by influencing the expression of other genes, their absence in undifferentiated ES cells might be essential for pluripotency. The presence of a flexible nuclear envelope could be necessary for proper chromatin remodeling, which occurs during differentiation. Interestingly, the low levels and/or altered organization of B-type lamins observed in HSF-6 hES cells (Figs. 3C, 3F, 4C, 4F) could also contribute to nuclear envelope flexibility. In such a scenario, forced expression of lamin A/C or increased B-type lamin expression might “lock” ES cells in their natural undifferentiated state.

In conclusion, our results identify the lack of lamin A/C as a novel marker for undifferentiated ES cells and lamin A/C expression as an early indicator of differentiation. These observations support the hypothesis that lamin A/C has a functional role in the maintenance of the differentiated state. Further, our results and the observation that undifferentiated ES have few nuclear pores (unpublished data) suggest that there are multiple structural differences between the nuclear envelopes of undifferentiated and differentiated cells that may be critical to the differentiation process.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Esther Jane, Patti Petrosko, Amy Mucka, and Hina Qidwai for maintaining the 129/SVEV mouse and HSF-6 human ES cell lines; Richard Chaillet for donating immortalized SNL MEF feeder cells; Leah Kauffman for editing the manuscript; and Chris Navara and Vanessa Y. Rawe for helpful discussion.

Disclosures

The authors indicate no potential conflicts of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References