Protein folding, cellular stress and cancer.

Proteins are acknowledged as the phenotypical manifestation of the genotype, because protein-coding genes carry the information for the strings of amino acids that constitute the proteins. It is widely accepted that protein function depends on the corresponding "native" structure or folding achieved within the cell, and that native protein folding corresponds to the lowest free energy minimum for a given protein. However, protein folding within the cell is a non-deterministic dissipative process that from the same input may produce different outcomes, thus conformational heterogeneity of folded proteins is the rule and not the exception. Local changes in the intracellular environment promote variation in protein folding. Hence protein folding requires "supervision" by a host of chaperones and co-chaperones that help their client proteins to achieve the folding that is most stable according to the local environment. Such environmental influence on protein folding is continuously transduced with the help of the cellular stress responses (CSRs) and this may lead to changes in the rules of engagement between proteins, so that the corresponding protein interactome could be modified by the environment leading to an alternative cellular phenotype. This allows for a phenotypic plasticity useful for adapting to sudden and/or transient environmental changes at the cellular level. Starting from this perspective, hereunder we develop the argument that the presence of sustained cellular stress coupled to efficient CSRs may lead to the selection of an aberrant phenotype as the resulting adaptation of the cellular proteome (and the corresponding interactome) to such stressful conditions, and this can be a common epigenetic pathway to cancer.

Is cancer a disease set up by cellular stress responses?

For several decades, the somatic mutation theory (SMT) has been the dominant paradigm on cancer research, leading to the textbook notion that cancer is fundamentally a genetic disease. However, recent discoveries indicate that mutations, including "oncogenic" ones, are widespread in normal somatic cells, suggesting that mutations may be necessary but not sufficient for cancer to develop. Indeed, a fundamental but as yet unanswered question is whether or not the first step in oncogenesis corresponds to a mutational event. On the other hand, for some time, it has been acknowledged the important role in cancer progression of molecular processes that participate in buffering cellular stress. However, their role is considered secondary or complementary to that of putative oncogenic mutations. Here we present and discuss evidence that cancer may have its origin in epigenetic processes associated with cellular adaptation to stressful conditions, and so it could be a direct consequence of stress-buffering mechanisms that allow cells with aberrant phenotypes (not necessarily associated with genetic mutations) to survive and propagate within the organism. We put forward the hypothesis that there would be an inverse correlation between the activation threshold of the cellular stress responses (CSRs) and the risk of cancer, so that species or individuals with low-threshold CSRs will display a higher incidence or risk of cancer.

The epicenter of chromosomal fragility of Fra14A2, the mouse ortholog of human FRA3B common fragile site, is largely attached to the nuclear matrix in lymphocytes but not in other cell types that do not express such a fragility.

Common fragile sites (CFSs) correspond to chromosomal regions susceptible to present breaks, discontinuities or constrictions in metaphase chromosomes from cells subjected to replication stress. They are considered as genomic regions intrinsically difficult to replicate and they are evolutionary conserved at least in mammals. However, the recent discovery that CFSs are cell-type specific indicates that DNA sequence by itself cannot account for CFS instability. Nevertheless, the large gene FHIT that includes FRA3B, the most highly expressed CFS in human lymphocytes, is commonly deleted in a variety of tumors suggesting a tumor suppressor role for its product. Here, we report that the epicenter of fragility of Fra14A2/Fhit, the mouse ortholog of human FRA3B/FHIT that like its human counterpart is the most highly expressed CFS in mouse lymphocytes, is largely attached to the nuclear matrix compartment in naive B lymphocytes but not in primary hepatocytes or cortical neurons that do not express such a CFS. Our results suggest a structural explanation for the difficult-to-replicate nature of such a region and so for its common fragility in lymphocytes, that is independent of the possible tumor suppressor role of the gene harboring such CFS.

Lessons we can learn from neurons to make cancer cells quiescent.

Cancer is a major concern for contemporary societies. However, the incidence of cancer is unevenly distributed among tissues and cell types. In particular, the evidence indicates that neurons are absolutely resistant to cancer and this is commonly explained on the basis of the known postmitotic state of neurons. The dominant paradigm on cancer understands this problem as a disease caused by mutations in cellular genes that result in unrestrained cell proliferation and eventually in tissue invasion and metastasis. However, the evidence also shows that mutations and gross chromosomal anomalies are common in functional neurons that nevertheless do not become neoplastic. This fact suggests that in the real nonexperimental setting mutations per se are not enough for inducing carcinogenesis but also that the postmitotic state of neurons is not genetically controlled or determined, otherwise there should be reports of spontaneously transformed neurons. Here we discuss the evidence that the postmitotic state of neurons has a structural basis on the high stability of their nuclear higher order structure that performs like an absolute tumor suppressor. We also discuss evidence that it is possible to induce a similar structural postmitotic state in nonneural cell types as a practical strategy for stopping or reducing the progression of cancer.

Landscaping the epigenetic landscape of cancer.

Waddington's epigenetic landscape was introduced in biology for understanding the complex process of metazoan development in an accessible fashion. The epigenetic landscape concept implies the coupling of cell differentiation and tissue/organ morphogenesis under a simple visual metaphor or analogy with significant heuristic value. Yet in recent times the epigenetic landscape has been reduced to an illustration device just for cell differentiation thus diminishing its explanatory power and heuristic value. On the other hand, the current mainstream in cancer research is concentrated on the search for proximate causes but not on achieving a deeper understanding of the phenomenon. Nevertheless an emerging alternative perspective that understands cancer as a problem related to tissue/organ morphology and structural organization is getting wider attention. Within such a perspective here we present and discuss a historically restored, non-reductionist, version of the epigenetic landscape that when applied to the problem of cancer improves our understanding of it as a common biological phenomenon resulting from the uncoupling of morphogenesis and cell differentiation as a consequence of the progressive erosion of the epigenetic landscape. The following discussion aims at finding a general framework, not dependent on proximate causes, for understanding the phenomenon of cancer and suggests new research strategies on this problem but away from the current emphasis on the putative genetic causes of cancer.

DNA Length Modulates the Affinity of Fragments of Genomic DNA for the Nuclear Matrix In Vitro.

Classical observations have shown that during the interphase the chromosomal DNA of metazoans is organized in supercoiled loops attached to a compartment known as the nuclear matrix (NM). Fragments of chromosomal DNA able to bind the isolated NM in vitro are known as matrix associated/attachment/addressed regions or MARs. No specific consensus sequence or motif has been found that may constitute a universal, defining feature of MARs. On the other hand, high-salt resistant DNA-NM interactions in situ define true DNA loop anchorage regions or LARs, that might correspond to a subset of the potential MARs but are not necessarily identical to MARs characterized in vitro, since there are several examples of MARs able to bind the NM in vitro but which are not actually bound to the NM in situ. In the present work we assayed the capacity of two LARs, as well as of shorter fragments within such LARs, for binding to the NM in vitro. Paradoxically the isolated (≈2 kb) LARs cannot bind to the NM in vitro while their shorter (≈300 pb) sub-fragments and other non-related but equally short DNA fragments, bind to the NM in a high-salt resistant fashion. Our results suggest that the ability of a given DNA fragment for binding to the NM in vitro primarily depends on the length of the fragment, suggesting that binding to the NM is modulated by the local topology of the DNA fragment in suspension that it is known to depend on the DNA length. J. Cell. Biochem. 118: 4487-4497, 2017. © 2017 Wiley Periodicals, Inc.

Why Cortical Neurons Cannot Divide, and Why Do They Usually Die in the Attempt?

Cortical neurons are prime examples of terminally differentiated, postmitotic cells. However, under experimental or pathological conditions, they can re-enter the cell cycle and replicate DNA but are unable to divide, dying by apoptosis or becoming either polyploid or aneuploid. Any cellular state that depends on the action of genes and their products can be reverted or bypassed by spontaneous or induced mutations, yet there are currently no reports of dividing cortical neurons. Thus, it seems unlikely that the remarkably stable postmitotic condition of cortical neurons depends on specific gene functions. This Review summarizes evidence that the postmitotic state of cortical neurons depends on the high stability of its underlying nuclear structure that results from an entropy-driven process aimed at dissipating the intrinsic structural stress present in chromosomal DNA in such a way that the structural stability of the neuronal nucleus becomes an insurmountable energy barrier for karyokinesis and mitosis. From this perspective, the integral properties of the nuclear higher order structure in neurons provide an explanation not only for why cortical neurons cannot divide but also for why they usually die if they happen to replicate their DNA. © 2016 Wiley Periodicals, Inc.

The Set of Structural DNA-Nuclear Matrix Interactions in Neurons Is Cell-Type Specific and Rather Independent of Functional Constraints.

In metazoans, nuclear DNA is organized during the interphase in negatively supercoiled loops anchored to a compartment or substructure known as the nuclear matrix. The interactions between DNA and the nuclear matrix (NM) are of higher affinity than those between DNA and chromatin proteins since the last ones do not resist the procedures for extracting the NM. The structural interactions DNA-NM constitute a set of topological relationships that define a nuclear higher order structure (NHOS) although there are further higher order levels of organization within the nucleus. So far, the evidence derived from studies with primary hepatocytes and naïve B lymphocytes indicates that the NHOS is cell-type specific at the local and at the large-scale level, and so it has been suggested that such NHOS is primary determined by structural and thermodynamic constraints. We carried out a comparative characterization of the NHOS of postmitotic cortical neurons with that of hepatocytes and naïve B lymphocytes. Our results indicate that the NHOS of neurons is completely different at the large scale and at the local level from that one observed in hepatocytes or in naïve B lymphocytes, confirming on the one hand that the set of structural DNA-NM interactions is cell-type specific and supporting, on the other hand the notion that structural constraints that impinge on chromosomal DNA and the NM are more important for determining this NHOS than functional constraints related to replication and/or transcription. J. Cell. Biochem. 118: 2151-2160, 2017. © 2016 Wiley Periodicals, Inc.

The nuclear higher-order structure defined by the set of topological relationships between DNA and the nuclear matrix is species-specific in hepatocytes.

During the interphase the nuclear DNA of metazoan cells is organized in supercoiled loops anchored to constituents of a nuclear substructure or compartment known as the nuclear matrix. The stable interactions between DNA and the nuclear matrix (NM) correspond to a set of topological relationships that define a nuclear higher-order structure (NHOS). Current evidence suggests that the NHOS is cell-type-specific. Biophysical evidence and theoretical models suggest that thermodynamic and structural constraints drive the actualization of DNA-NM interactions. However, if the topological relationships between DNA and the NM were the subject of any biological constraint with functional significance then they must be adaptive and thus be positively selected by natural selection and they should be reasonably conserved, at least within closely related species. We carried out a coarse-grained, comparative evaluation of the DNA-NM topological relationships in primary hepatocytes from two closely related mammals: rat and mouse, by determining the relative position to the NM of a limited set of target sequences corresponding to highly-conserved genomic regions that also represent a sample of distinct chromosome territories within the interphase nucleus. Our results indicate that the pattern of topological relationships between DNA and the NM is not conserved between the hepatocytes of the two closely related species, suggesting that the NHOS, like the karyotype, is species-specific.

The interphase mammalian chromosome as a structural system based on tensegrity.

Each mammalian chromosome is constituted by a DNA fiber of macroscopic length that needs to be fitted in a microscopic nucleus. The DNA fiber is subjected at physiological temperature to random thermal bending and looping that must be constrained so as achieve structural stability thus avoiding spontaneous rupturing of the fiber. Standard textbooks assume that chromatin proteins are primarily responsible for the packaging of DNA and so of its protection against spontaneous breakage. Yet the dynamic nature of the interactions between chromatin proteins and DNA is unlikely to provide the necessary long-term structural stability for the chromosomal DNA. On the other hand, longstanding evidence indicates that stable interactions between DNA and constituents of a nuclear compartment commonly known as the nuclear matrix organize the chromosomal DNA as a series of topologically constrained, supercoiled loops during interphase. This results in a primary level of DNA condensation and packaging within the nucleus, as well as in protection against spontaneous DNA breakage, independently of chromatin proteins which nevertheless increase and dynamically modulate the degree of DNA packaging and its role in the regulation of DNA function. Thus current evidence, presented hereunder, supports a model for the organization of the interphase chromosome as resilient system that satisfies the principles of structural tensegrity.

The higher-order structure in the cells nucleus as the structural basis of the post-mitotic state.

In metazoan cells during the interphase nuclear DNA is organized in supercoiled, topologically constrained loops anchored to a proteinaceous compartment or substructure commonly known as the nuclear matrix (NM). The DNA-NM interactions result from a thermodynamically-driven process leading to the necessary dissipation of structural stress along chromosomal DNA, otherwise the chromosomes would break into pieces. Such DNA-NM interactions define a nuclear higher-order structure that is independent of chromatin proteins. On the other hand, a metazoan cell no longer able to undergo mitosis is defined as post-mitotic and this condition indicates a terminally differentiated cell that may survive in such a state for indefinite time. The non-reversible nature of the post-mitotic state suggests a non-genetic basis for it since no spontaneous or induced mutations can revert it. Yet in individual cells the loss of proliferative potential has both a developmental and a stochastic component. Here we discuss evidence suggesting that the stability of the nuclear higher-order structure is the factor that links the stochastic and developmental components leading to the post-mitotic state.

The post-mitotic state in neurons correlates with a stable nuclear higher-order structure.

Neurons become terminally differentiated (TD) post-mitotic cells very early during development yet they may remain alive and functional for decades. TD neurons preserve the molecular machinery necessary for DNA synthesis that may be reactivated by different stimuli but they never complete a successful mitosis. The non-reversible nature of the post-mitotic state in neurons suggests a non-genetic basis for it since no set of mutations has been able to revert it. Comparative studies of the nuclear higher-order structure in neurons and cells with proliferating potential suggest that the non-reversible nature of the post-mitotic state in neurons has a structural basis in the stability of the nuclear higher-order structure.

Reorganization of the DNA-nuclear matrix interactions in a 210 kb genomic region centered on c-myc after DNA replication in vivo.

In the interphase nucleus of metazoan cells DNA is organized in supercoiled loops anchored to a nuclear matrix (NM). DNA loops are operationally classified in structural and facultative. Varied evidence indicates that DNA replication occurs in replication foci organized upon the NM and that structural DNA loops may correspond to the replicons in vivo. In normal rat liver the hepatocytes are arrested in G0 but synchronously re-enter the cell cycle after partial-hepatectomy leading to liver regeneration. Using this model we have previously determined that the DNA loops corresponding to a gene-rich genomic region move in a sequential fashion towards the NM during replication and then return to their original configuration in newly quiescent cells, once liver regeneration has been achieved. In the present work we determined the organization into structural DNA loops of a gene-poor region centered on c-myc and tracked-down its movement at the peak of S phase and after the return to cellular quiescence during and after liver regeneration. The results confirmed that looped DNA moves towards the NM during replication but in this case the configuration of the gene-poor region into DNA loops becomes reorganized and after replication only the loop containing c-myc resembles the original in the control G0 hepatocytes. Our results suggest that the local chromatin configuration around potentially active genes constraints the formation of specific structural DNA loops after DNA replication, while in non-coding regions the structural DNA loops are only loosely determined after DNA replication by structural constraints that modulate the DNA-NM interactions.

The organization of a large transcriptional unit (Fyn) into structural DNA loops is cell-type specific and independent of transcription.

In the interphase nucleus of metazoan cells the DNA is organized in supercoiled loops anchored to a nuclear matrix (NM). The DNA is anchored by non-coding sequences known as MARs, in situ operationally classified in structural-constitutive and transient-functional. We have previously shown that the organization of the multi-gene rat-albumin family locus into structural DNA loops is remarkably different between primary hepatocytes, where such genes are expressed, and naïve B lymphocytes, where such genes are not expressed. These results together with previous observations from other authors suggested that the local organization into structural DNA loops might determine the potential for a gene to be expressed or not. Thus in the present work we determined the organization of the Fyn locus, a single large transcriptional unit, into structural DNA loops in both primary rat hepatocytes and B lymphocytes. Our results indicate that the organization of the Fyn locus in structural DNA loops is cell type-specific and yet the gene is expressed in both cell types, supporting the notion that in vivo the organization of DNA into structural loops is primarily determined by factors independent of transcription but also that transcription adapts to work upon radically different structural DNA loop organizations.

Continued stabilization of the nuclear higher-order structure of post-mitotic neurons in vivo.

BACKGROUND: Cellular terminal differentiation (TD) correlates with a permanent exit from the cell cycle and so TD cells become stably post-mitotic. However, TD cells express the molecular machinery necessary for cell proliferation that can be reactivated by experimental manipulation, yet it has not been reported the stable proliferation of any type of reactivated TD cells. Neurons become post-mitotic after leaving the ventricular zone. When neurons are forced to reenter the cell cycle they invariably undergo cell death. Wider evidence indicates that the post-mitotic state cannot solely depend on gene products acting in trans, otherwise mutations in the corresponding genes may lead to reentry and completion of the cell cycle in TD cells, but this has not been observed. In the interphase, nuclear DNA of metazoan cells is organized in supercoiled loops anchored to a nuclear nuclear matrix (NM). The DNA-NM interactions define a higher-order structure in the cell nucleus (NHOS). We have previously compared the NHOS of aged rat hepatocytes with that of early post-mitotic rat neurons and our results indicated that a very stable NHOS is a common feature of both senescent and post-mitotic cells in vivo. PRINCIPAL FINDINGS: In the present work we compared the NHOS in rat neurons from different post-natal ages. Our results show that the trend towards further stabilization of the NHOS in neurons continues throughout post-natal life. This phenomenon occurs in absence of overt changes in the post-mitotic state and transcriptional activity of neurons, suggesting that it is independent of functional constraints. CONCLUSIONS: Apparently the continued stabilization of the NHOS as a function of time is basically determined by thermodynamic and structural constraints. We discuss how the resulting highly stable NHOS of neurons may be the structural, non-genetic basis of their permanent and irreversible post-mitotic state.

DNA moves sequentially towards the nuclear matrix during DNA replication in vivo.

BACKGROUND: In the interphase nucleus of metazoan cells DNA is organized in supercoiled loops anchored to a nuclear matrix (NM). There is varied evidence indicating that DNA replication occurs in replication factories organized upon the NM and that DNA loops may correspond to the actual replicons in vivo. In normal rat liver the hepatocytes are arrested in G0 but they synchronously re-enter the cell cycle after partial-hepatectomy leading to liver regeneration in vivo. We have previously determined in quiescent rat hepatocytes that a 162 kbp genomic region containing members of the albumin gene family is organized into five structural DNA loops. RESULTS: In the present work we tracked down the movement relative to the NM of DNA sequences located at different points within such five structural DNA loops during the S phase and after the return to cellular quiescence during liver regeneration. Our results indicate that looped DNA moves sequentially towards the NM during replication and then returns to its original position in newly quiescent cells, once the liver regeneration has been achieved. CONCLUSIONS: Looped DNA moves in a sequential fashion, as if reeled in, towards the NM during DNA replication in vivo thus supporting the notion that the DNA template is pulled progressively towards the replication factories on the NM so as to be replicated. These results provide further evidence that the structural DNA loops correspond to the actual replicons in vivo.

Cell-type-specific organization of nuclear DNA into structural looped domains.

In the interphase nucleus of metazoan cells the DNA is organized in supercoiled loops anchored to a proteinaceous substructure known as the nuclear matrix (NM). The DNA is anchored to the NM by means of non-coding sequences of variable length known as matrix attachment regions or MARs operationally classified in structural-constitutive, resistant to high-salt extraction and transient-functional, non-resistant to high-salt extraction. The former are also known as true loop attachment regions or LARs that determine structural DNA loops. The DNA-NM interactions define a higher order structure within the cell nucleus (NHOS). We studied in a comparative fashion the NHOS in two primary cell types from the rat: hepatocytes and naive B lymphocytes, by analyzing the topological relationships between the NM and a set of eight short gene sequences located in six separate chromosomes and as such representing a coarse-grained, large-scale sample of the actual organization of nuclear DNA into structural loop domains. Our results indicate that such an organization is cell-type specific since most of the gene sequences studied showed significant differences in their relative position to the NM according to cell type. Such cell-type specific differences in the NHOS have no obvious correlation with the tissue-specific transcriptional activity of the corresponding genes, supporting the notion that permanent, structural DNA loops are different from transient, functional DNA loops that may be associated with transcription.

Aged and post-mitotic cells share a very stable higher-order structure in the cell nucleus in vivo.

In the mammalian liver the quiescent primary hepatocytes preserve a proliferating potential in vivo, yet natural aging correlates with loss of proliferating potential and progression towards terminal differentiation of the hepatocytes. Thus aged, terminally-differentiated hepatocytes may survive in a de facto post-mitotic state, similarly to early post-mitotic cells, like neurons, suggesting that there might be a common factor linking both cellular states. In the interphase of metazoan cells the nuclear DNA is organized in supercoiled loops anchored to a proteinaceous substructure known as the nuclear matrix (NM). The DNA-NM interactions define a higher-order structure in the cell nucleus (NHOS). Natural aging of the rat liver correlates with a progressive strengthening of the NM framework and the stabilization of the DNA-NM interactions in the hepatocytes indicating that the NHOS becomes highly stable with age. We compared the NHOS of post-mitotic rat neurons with that of aged rat hepatocytes. Our results indicate that a very stable NHOS is a common feature of both aged and post-mitotic cells in vivo.

NeuN/Fox-3 is an intrinsic component of the neuronal nuclear matrix.

NeuN is an antigen detected in the nucleus of neurons in a wide range of vertebrates and so it is widely used as a tool for detecting neuronal cells. NeuN has been recently identified as Fox-3, a new member of the Fox-1 gene family of splicing factors. The predominant localization of NeuN/Fox-3 to neuronal nuclei and its role in splicing pose the question of the nuclear compartmentalization of such a protein. Here we provide evidence that NeuN/Fox-3 is an intrinsic component of the neuronal nuclear matrix and a reliable marker of nuclear speckles in neurons.

Determination of the in vivo structural DNA loop organization in the genomic region of the rat albumin locus by means of a topological approach.

Nuclear DNA of metazoans is organized in supercoiled loops anchored to a proteinaceous substructure known as the nuclear matrix (NM). DNA is anchored to the NM by non-coding sequences known as matrix attachment regions (MARs). There are no consensus sequences for identification of MARs and not all potential MARs are actually bound to the NM constituting loop attachment regions (LARs). Fundamental processes of nuclear physiology occur at macromolecular complexes organized on the NM; thus, the topological organization of DNA loops must be important. Here, we describe a general method for determining the structural DNA loop organization in any large genomic region with a known sequence. The method exploits the topological properties of loop DNA attached to the NM and elementary topological principles such as that points in a deformable string (DNA) can be positionally mapped relative to a position-reference invariant (NM), and from such mapping, the configuration of the string in third dimension can be deduced. Therefore, it is possible to determine the specific DNA loop configuration without previous characterization of the LARs involved. We determined in hepatocytes and B-lymphocytes of the rat the DNA loop organization of a genomic region that contains four members of the albumin gene family.