Nuclear Myosin I
Nuclear myosin I (NM1) belongs to the group of class I myosins, which are monomeric, nonprocessive, slow-rate, and low-duty ratio molecular motors transforming free chemical energy stored in ATP into mechanical force. Nuclear myosin I was discovered by testing antibodies to adrenal myosin 1. The antibody was staining a 120 kDa nuclear protein with ATPase activity, and the protein was ATP-, actin-, and calmodulin- binding, which are the typical features of unconventional myosins. At that time, there were no myosins known to be present in the cell nucleus, hence the discovered protein was called nuclear myosin I (Pestic-Dragovich et al. 2000). The mass spectrometric analysis of the NM1 showed a high homology to the Myosin 1c (Myo1c) protein, the first single-headed myosin isolated from mammals, also known as mammalian myosin I, or myosin 1β. However, with the increasing numbers of myosins discovered, there was a need to unify the myosin nomenclature. Therefore, we now recognize NM1 as an isoform of Myosin 1c protein (Gillespie et al. 2001).
Historically, Myo1C was considered to be purely cytoplasmic, while NM1 purely nuclear. N-terminal extension as the only difference between cytoplasmic and nuclear isoform was believed to provide a nuclear localization signal. However, there is rising evidence that these isoform cooperate together in both compartments and serve similar functions. Dzijak et al. showed that nuclear localization signal which directs these myosins to the cell nucleus is localized in the neck region of the molecule and is common for all three isoforms (Dzijak et al. 2012). Moreover, the proteomic studies of fractionated lysates showed that there is no compartment-specific enrichment of one of the isoforms. Instead, relatively small fraction of both isoforms is present in the nucleus (∼20%) in comparison to the cytoplasm (∼80%) (Venit et al. 2016). Therefore it seems plausible that rather than having separate function, the overall amount of myosin molecules together is more important in concrete cellular processes, and that they might even replace one another in these functions. This was shown in NM1 knockout mice, where the loss of NM1 was compensated by Myo1C (Venit et al. 2013). However, we cannot exclude the possibility, that these isoform has also some specific functions as the identified N-terminal extension of NM1 is highly conserved across metazoans and expression profiles of both isoforms somewhat differ in various tissues (Kahle et al. 2007).
NM1 Structure and Function
NM1 in Transcription and Chromatin Remodeling
NM1 in Gene Movements
The cell nucleus is highly compartmentalized structure, where chromosomes occupy discreet radially organized territories. Chromosomes with more active genes are localized to the center of the nucleus while gene-poor chromosomes are localized more to the nuclear periphery. However, this organization is not static and there is always some intermingling of genes which seem to be actively moved toward the transcription machineries. In the cytoplasm, the movement of cargos is achieved by different molecular motors which move vesicles along the different kinds of filaments. The similar mechanism has been suggested to occur also in the cell nucleus. It was shown that NM1 and actin are necessary for long-range migration of chromosome site from nuclear periphery to the interior after transcription activation. Mutation in NM1 motor domain or blocking of actin polymerization blocks this movement, however the direct mechanism is still not known and results are not conclusive yet (Chuang et al. 2006; Dundr et al. 2007). Apparently, there are no long actin filaments present in the nucleoplasm, although they can form there under some specific, mostly unphysiological, circumstances (Kalendova et al. 2014). New tools, especially molecular probes, will be needed to understand actin-myosin functions in chromosome movements.
NM1 in the Cytoplasm
As it was shown that “cytoplasmic” Myo1c can compensate for the NM1 loss in the nucleus and that the two isoforms are interchangeable in the transcription, the question whether NM1 protein can function in the cytoplasm was raised up. The study by Venit and Kalendova demonstrated that NM1 is tethered to the plasma membrane by interaction with plasma membrane-associated phosphoinositides, and NM1 and Myo1c isoforms are evenly distributed next to each other (Venit et al. 2016).
Myo1C has been shown to maintain effective cell membrane tension and regulate osmotic state of the cell by linking of plasma membrane to the underlying actin cytoskeleton. Atomic force microscopy measurements of plasma membrane of cultured fibroblasts with a deletion in NM1 showed that these cells have higher membrane elasticity, supporting the idea that NM1 regulates plasma membrane tension via its lineage to the underlying actin cytoskeleton similarly to Myo1C. This was proven also functionally, as upon deletion of NM1, the cells exhibit higher ability to tolerate strong hypotonic conditions, suggesting a deleterious effect of myosin linkage between plasma membrane and actin cytoskeleton under rapid membrane rearrangements caused by changes in osmotic conditions. Therefore, the deletion of NM1 leads to the reduction of the overall number of myosin molecules on the plasma membrane, which afterward binds to the cytoskeleton less tightly, and cells are more susceptible to swelling. On the other side, NM1 KO cells did not show any difference in adhesion, spreading, or motility, suggesting that the basic level of membrane tension is sufficiently maintained by other class I myosins. However, the regulation and dynamics between different myosin I members on the plasma membrane are not yet known. For example, the N-terminal extension of NM1 does not affect its membrane localization since the amount of NM1 and Myo1c isoforms at the plasma membrane is the same, but as a part of the motor head domain it could affect the binding of NM1 to the actin cytoskeleton. It is also possible that multiple pathways contribute to the phenotype observed under stress conditions – because NM1 is widely spread all over the plasma membrane, not only bound to actin filaments.
Summary and Perspectives
There is an increasing evidence of specific nuclear functions of several myosin classes. However, the interplay between different myosin isoforms and classes is hardly known. In case of NM1 and Myo1C, the general idea of having one isoform in the nucleus and one in the cytoplasm is still present in recent publications. However, strictly speaking, no studies have been published directly proving isoform-specific functions of these myosins. This is especially important in the experimental design and data explanation, as most of the techniques based on the gene targeting or antibody usage affect both myosin isoforms. Therefore, one of the questions which has not been answered yet is what is the functional significance of N-terminal extension of NM1 protein? Current advances in gene knockout technologies by using of CRISP/Cas system combined with exogenous expression of single isoforms and high-throughput analysis could answer this question.
Secondly, there is an emerging evidence about different phosphoinositides in the cell nucleus. Phosphoinositol-4,5-bis-phosphate (PIP2) have been shown to be in complex with nascent RNA transcript affecting Pol I transcription by binding to UBF and fibrillarin in the nucleolus (Yildirim et al. 2013). NM1 binds to PIP2 by its PH domain in the cytoplasm and therefore similar interaction is predicted also in the cell nucleus. However, the functional significance of such interaction needs to be elucidated.
Finally, MYO1C gene is localized in a genomic locus commonly deleted in a variety of human tumors, and therefore it is one of the candidates as a novel tumor suppressor gene (Hedberg Oldfors et al. 2015). This is further supported by the increasing number of studies reporting that cancer cells show significantly higher elasticity and deformability than control cells derived from the healthy tissues. While healthy control cells display a broad distribution of the elasticity modulus, cancer cells show a significantly narrower Gaussian distribution with a significantly lower standard deviation of the Young’s modulus (Suresh 2007). These data resemble elastic phenotype of the WT and NM1 KO cells, which show very similar relative patterns of the Young’s modulus value distribution. Therefore, deletion, mutation, or change of the relative amounts of NM1 or other class 1 myosins might lead to an increased metastatic potential of the cells. Understanding of myosins dynamics at the plasma membrane could therefore bring new insight into a cancer research and therapy.
- Hedberg Oldfors C, Dios DG, Linder A, Visuttijai K, Samuelson E, Karlsson S, et al. Analysis of an independent tumor suppressor locus telomeric to Tp53 suggested Inpp5k and Myo1c as novel tumor suppressor gene candidates in this region. BMC Genet. 2015;16:80. doi: 10.1186/s12863-015-0238-4.PubMedPubMedCentralCrossRefGoogle Scholar
- Kalendova A, Kalasova I, Yamazaki S, Ulicna L, Harata M, Hozak P. Nuclear actin filaments recruit cofilin and actin-related protein 3, and their formation is connected with a mitotic block. Histochem Cell Biol. 2014;142:139–52. doi: 10.1007/s00418-014-1243-9.PubMedPubMedCentralCrossRefGoogle Scholar
- Sarshad A, Sadeghifar F, Louvet E, Mori R, Bohm S, Al-Muzzaini B, et al. Nuclear myosin 1c facilitates the chromatin modifications required to activate rRNA gene transcription and cell cycle progression. PLoS Genet. 2013;9:e1003397. doi: 10.1371/journal.pgen.1003397.PubMedPubMedCentralCrossRefGoogle Scholar
- Sarshad AA, Corcoran M, Al-Muzzaini B, Borgonovo-Brandter L, Von Euler A, Lamont D, et al. Glycogen synthase kinase (GSK) 3beta phosphorylates and protects nuclear myosin 1c from proteasome-mediated degradation to activate rDNA transcription in early G1 cells. PLoS Genet. 2014;10:e1004390. doi: 10.1371/journal.pgen.1004390.PubMedPubMedCentralCrossRefGoogle Scholar
- Sielski NL, Ihnatovych I, Hagen JJ, Hofmann WA. Tissue specific expression of Myosin IC Isoforms. Bmc Cell Biol. 2014;15. doi: 10.1186/1471-2121-15-8.
- Venit T, Dzijak R, Kalendova A, Kahle M, Rohozkova J, Schmidt V, et al. Mouse nuclear myosin I knock-out shows interchangeability and redundancy of myosin isoforms in the cell nucleus. PLoS One 2013;8. doi: 10.1371/journal.pone.0061406.