Robbins Basic Pathology / Основи на Патологията на Робинс: 1. The Cell as a Unit of Health and Disease

Pathology literally translates to the study of suffering (Greek pathos = suffering, logos = study); as applied to modern medicine, it is the study of disease. Virchow was certainly correct in asserting that disease originates at the cellular level, but we now realize that cellular disturbances arise from alterations in molecules (genes, proteins, and others) that influence the survival and behavior of cells. Thus, the foundation of modern pathology is understanding the cel- lular and molecular abnormalities that give rise to diseases. It is helpful to consider these abnormalities in the context of normal cellular structure and function, which is the theme of this introductory chapter.

#1

It is unrealistic (and even undesirable) to condense the vast and fascinating field of cell biology into a single chapter. Consequently, rather than attempting a comprehensive review, the goal here is to survey basic principles and highlight recent advances that are relevant to the mechanisms of disease that are emphasized throughout the rest of the book.

#2

THE GENOME

#3

The sequencing of the human genome at the beginning of the 21st century represented a landmark achievement of biomedical science. Since then, the rapidly dropping cost of sequencing and the computational capacity to analyze vast amounts of data promise to revolutionize our under- standing of health and disease. At the same time, the emerging information has also revealed a breathtaking level of complexity far beyond the linear sequencing of the genome. The potential for these new powerful tools to expand our understanding of pathogenesis and drive therapeutic innovation excites and inspires scientists and the lay public alike.

#4

Noncoding DNA

#5

The human genome contains about 3.2 billion DNA base pairs. Yet, within the genome there are only roughly 20,000 protein-encoding genes, comprising just 1.5% of the genome. The proteins encoded by these genes are the fun- damental constituents of cells, functioning as enzymes, structural elements, and signaling molecules. Although 20,000 underestimates the actual number of proteins encoded (many genes produce multiple RNA transcripts that encode distinct protein isoforms), it is nevertheless startling that worms composed of fewer than 1000 cells— and with genomes 30-fold smaller—are also assembled from roughly 20,000 protein-encoding genes. Perhaps even more unsettling is that many of these proteins are recogniz- able homologs of molecules expressed in humans. What then separates humans from worms?

#6

The answer is not completely known, but evidence sup- ports the assertion that the difference lies in the 98.5% of the human genome that does not encode proteins. The function of such long stretches of DNA (which has been called the “dark matter” of the genome) was mysterious for many years. However, it is now clear that more than 85% of the human genome is ultimately transcribed, with almost 80% being devoted to the regulation of gene expression. It follows that whereas proteins provide the building blocks and machinery required for assembling cells, tissues, and organisms, it is the noncoding regions of the genome that provide the critical “architectural planning.”

#7

The major classes of functional non–protein-coding DNA sequences found in the human genome include (Fig. 1.1):

#8

Fig. 1.1  The organization of nuclear DNA. At the light microscopic level, the nuclear genetic material is organized into dispersed, transcriptionally active  euchromatin or densely packed, transcriptionally inactive heterochromatin; chromatin can also be mechanically connected with the nuclear membrane, and  nuclear membrane perturbation can thus influence transcription. Chromosomes (as shown) can only be visualized by light microscopy during cell division.  During mitosis, they are organized into paired chromatids connected at centromeres; the centromeres act as the locus for the formation of a kinetochore  protein complex that regulates chromosome segregation at metaphase. The telomeres are repetitive nucleotide sequences that cap the termini of chromatids  and permit repeated chromosomal replication without loss of DNA at the chromosome ends. The chromatids are organized into short “P” (“petite”) and  long “Q” (“next letter in the alphabet”) arms. The characteristic banding pattern of chromatids has been attributed to relative GC content (less GC content  in bands relative to interbands), with genes tending to localize to interband regions. Individual chromatin fibers are composed of a string of nucleosomes— DNA wound around octameric histone cores—with the nucleosomes connected via DNA linkers. Promoters are noncoding regions of DNA that initiate  gene transcription; they are on the same strand and upstream of their associated gene. Enhancers are regulatory elements that can modulate gene expression  across distances of 100 kB or more by looping back onto promoters and recruiting additional factors that are needed to drive the expression of pre-mRNA  species. The intronic sequences are subsequently spliced out of the pre-mRNA to produce the definitive message that includes exons that are translated into  protein and 3′- and 5′-untranslated regions (UTR) that may have regulatory functions. In addition to the enhancer, promoter, and UTR sequences, noncoding  elements are found throughout the genome; these include short repeats, regulatory factor binding regions, noncoding regulatory RNAs, and transposons. 

#9

• Promoter and enhancer regions that bind protein tran- scription factors

#10

• Binding sites for proteins that organize and maintain higher order chromatin structures

#11

• Noncoding regulatory RNAs. Of the 80% of the genome dedicated to regulatory functions, the vast majority is transcribed into RNAs—micro-RNAs and long noncod- ing RNAs (described later)—that are never translated into protein, but can regulate gene expression

#12

• Mobile genetic elements (e.g., transposons). Remarkably, more than one-third of the human genome is composed of such “jumping genes.” These segments can cruise around the genome, and are implicated in gene regula- tion and chromatin organization.

#13

• Special structural regions of DNA, including telomeres (chromosome ends) and centromeres (chromosome “tethers”)

#14

Importantly, many genetic variations (polymorphisms) associated with diseases are located in non–protein- coding regions of the genome. Thus, variation in gene regulation may prove to be more important in disease cau- sation than structural changes in specific proteins. Another surprise that emerged from genome sequencing is that any two humans are typically >99.5% DNA-identical (and are 99% sequence-identical with chimpanzees)! Thus, individ- ual variation, including differential susceptibility to dis- eases and environmental exposures, is encoded in <0.5% of our DNA (importantly, this still represents about 15 million base pairs).

#15

The two most common forms of DNA variation in the human genome are single-nucleotide polymorphisms (SNPs) and copy number variations (CNVs).

#16

• SNPs are variants at single nucleotide positions and are almost always biallelic (only two choices exist at a given site within the population, such as A or T). More than 6 million human SNPs have been identified, with many showing wide variation in frequency in differ- ent populations. The following features are worthy of note:

#17

• SNPs occur across the genome—within exons, introns, intergenic regions, and coding regions.

#18

• Roughly 1% of SNPs occur in coding regions, which is about what would be expected by chance, because coding regions comprise about 1.5% of the genome.

#19

• SNPs located in noncoding regions can occur in regu- latory elements in the genome, thereby altering gene expression; in such instances the SNP may have a direct influence on disease susceptibility.

#20

SNPs can also be “neutral” variants with no effect on gene function or carrier phenotype.

#21

•
Even “neutral” SNPs may be useful markers if they happen to be coinherited with a disease-associated gene as a result of physical proximity. In other words, the SNP and the causative genetic factor are in linkage disequilibrium.

#22

•
The effect of most SNPs on disease susceptibility is weak, and it remains to be seen if the identification of such variants, alone or in combination, can be used to develop effective strategies for disease prediction or prevention.

#23

•
CNVs are a form of genetic variation consisting of dif- ferent numbers of large contiguous stretches of DNA; these can range from 1000 base pairs to millions of base pairs. In some instances these loci are, like SNPs, bial- lelic and simply duplicated or deleted in a subset of the population. In other instances there are complex rear- rangements of genomic material, with multiple alleles in the human population. CNVs are responsible for several million base pairs of sequence difference between any wo individuals. Approximately 50% of CNVs involve
gene-coding sequences; thus, CNVs may underlie a large portion of human phenotypic diversity.

#24

It is important to note that alterations in DNA sequence cannot by themselves explain the diversity of phenotypes in human populations; moreover, classical genetic inheri- tance cannot explain differing phenotypes in monozygotic twins. The answers to these conundrums probably lie in epigenetics—heritable changes in gene expression that are not caused by alterations in DNA sequence (see later).

#25

Histone Organization

#26

Even though virtually all cells in the body have the same genetic composition, differentiated cells have distinct structures and functions arising through lineage-specific programs of gene expression. Such cell type–specific differ- ences in DNA transcription and translation are regulated by epigenetic modifications that consist of several changes that profoundly influence gene expression, including:

#27

• Chromatin organization (Fig. 1.2).

#28

Fig. 1.2  Chromatin organization. (A) Nucleosomes are comprised of octamers of histone proteins (two each of histone subunits H2A, H2B, H3, and H4)  encircled by 1.8 loops of 147 base pairs of DNA; histone H1 sits on the 20 to 80 nucleotide linker DNA between nucleosomes and helps stabilize the overall  chromatin architecture. The histone subunits are positively charged, thus allowing the compaction of the negatively charged DNA. (B) The relative state of  DNA unwinding (and thus access for transcription factors) is regulated by histone modification, for example, by acetylation, methylation, and/or phosphoryla- tion (so-called “marks”); marks are dynamically written and erased. Certain marks such as histone acetylation “open up” the chromatin structure, whereas  others, such as methylation of particular histone residues, tend to condense the DNA and lead to gene silencing. DNA itself can also be also be methylated,  a modification that is associated with transcriptional inactivation. 

#29

Genomic DNA is packed into nucleosomes, which are composed of 147 base pair DNA segments wrapped around a central core

#30

of proteins called histones. Nucleosomes resemble beads joined by short DNA linkers; the entire structure is generically called chromatin. Importantly, the winding and compaction of chromatin in any given cell varies in different genomic regions. Thus, nuclear chromatin exists in two basic forms (visualizable by standard his- tology): (1) histochemically dense and transcriptionally inactive heterochromatin and (2) histochemically dis- persed and transcriptionally active euchromatin. Because only euchromatin permits gene expression and thereby dictates cellular identity and activity, there are a host of mechanisms that tightly regulate the state of chromatin (described below).

#31

•
DNA methylation. High levels of DNA methylation in gene regulatory elements typically result in chroma- tin condensation and transcriptional silencing. Like histone modifications (see later), DNA methylation is tightly regulated by methyltransferases, demethylating enzymes, and methylated-DNA-binding proteins.

#32

•
Histone modifying factors. Nucleosomes are highly dynamic structures regulated by an array of nuclear proteins and post-translational modifications:

#33

•
Chromatin remodeling complexes can reposition nucleo- somes on DNA, exposing (or obscuring) gene regula- tory elements such as promoters.

#34

•
“Chromatin writer” complexes carry out more than 70 different covalent histone modifications generi- cally denoted as marks. These include methylation, acetylation, and phosphorylation of specific histone amino acid residues: Histone methylation of lysines and arginines is accomplished by specific writer enzymes; methylation of histone lysine residues can lead to transcriptional activation or repression, depending on which histone residue is “marked.” Histone acetylation of lysine residues (occurring through histone acetyl transferases) tends to open up chromatin and increase transcription; histone deacetylases (HDAC) reverse this process, leading to chromatin condensation. Histone phosphorylation of serine residues can variably open or condense chromatin, to increase or decrease transcription, respectively.

#35

•
Histone marks are reversible through the activity of “chromatin erasers.” Other proteins function as “chro- matin readers,” binding histones that bear particular marks and thereby regulating gene expression.

#36

The mechanisms involved in the cell-specific epigenetic regulation of genomic organization and gene expression are undeniably complex. Despite the intricacies, learning to manipulate these processes will likely bear significant therapeutic benefits because many diseases are associated with inherited or acquired epigenetic alterations, and dysregulation of the “epigenome” has a central role in the genesis of benign and malignant neoplasms (Chapter 6). Moreover—unlike genetic changes—epigenetic alterations (e.g., histone acetylation and DNA methylation) are readily reversible and are therefore amenable to intervention; indeed, HDAC inhibitors and DNA methylation inhibitors are already being used in the treatment of various forms of cancer.

#37

Micro-RNA and Long Noncoding RNA

#38

Another mechanism of gene regulation depends on the functions of noncoding RNAs. As the name implies, these are encoded by genes that are transcribed but not trans- lated. Although many distinct families of noncoding RNAs exist, only two examples are discussed here: small RNA molecules called microRNAs and long noncoding RNAs >200 nucleotides in length.

#39

• Micro-RNAs (miRNAs) are relatively short RNAs (22 nucleotides on average) that function primarily to modulate the translation of target mRNAs into their corresponding proteins. Posttranscriptional silencing of gene expression by miRNA is a fundamental and evolutionarily conserved mechanism of gene regula- tion present in all eukaryotes (plants and animals). Even bacteria have a primitive version of the same general machinery that they use to protect themselves against foreign DNA (e.g., from phages and viruses).

#40

• The human genome contains almost 6000 miRNA genes, only 3.5-fold less than the number of protein-coding genes. Moreover, individual miRNAs appear to regu- late multiple protein-coding genes, allowing each miRNA to coregulate entire programs of gene expres- sion. Transcription of miRNA genes produces a primary transcript (pri-miRNA) that is processed into progres- sively smaller segments, including trimming by the enzyme Dicer. This generates mature single-stranded miRNAs of 21 to 30 nucleotides that associate with a multiprotein aggregate called RNA-induced silencing complex (RISC; Fig. 1.3). Subsequent base pairing between the miRNA strand and its target mRNA directs the RISC to either induce mRNA cleavage or to repress its translation. In this way, the target mRNA is posttran- scriptionally silenced.

#41

Taking advantage of the same pathway, small interfering RNAs (siRNAs) are short RNA sequences that can be intro- duced into cells. These serve as substrates for Dicer and interact with the RISC complex in a manner analogous to endogenous miRNAs. Synthetic siRNAs that can target specific mRNA species are therefore powerful laboratory tools to study gene function (so-called knockdown technol- ogy); they also are promising as therapeutic agents to silence pathogenic genes, e.g., oncogenes involved in neo- plastic transformation.

#42

•

#43

Long noncoding RNA (lncRNA). The human genome also contains a very large number of lncRNAs—at least 30,000, with the total number potentially exceeding coding mRNAs by 10- to 20-fold. lncRNAs modulate gene expression in many ways (Fig. 1.4); for example, they can bind to regions of chromatin, restricting RNA polymerase access to coding genes within the region. The best-known example of a repressive function involves XIST, which is transcribed from the X chromo- some and plays an essential role in physiologic X chro- mosome inactivation. XIST itself escapes X inactivation, but forms a repressive “cloak” on the X chromosome from which it is transcribed, resulting in gene silencing. Conversely, it has been appreciated that many enhanc- ers are sites of lncRNA synthesis, with the lncRNAs expanding transcription from gene promoters through a variety of mechanisms (Fig. 1.4). Ongoing studies are exploring the role of lncRNAs in diseases like athero- sclerosis and cancer.

#44

Fig. 1.3  Generation of microRNAs (miRNA) and their mode of action in  regulating gene function. miRNA genes are transcribed to produce a primary  miRNA (pri-miRNA), which is processed within the nucleus to form pre- miRNA  composed  of  a  single  RNA  strand  with  secondary  hairpin  loop  structures  that  form  stretches  of  double-stranded  RNA. After  this  pre- miRNA is exported out of the nucleus via specific transporter proteins, the  cytoplasmic enzyme Dicer trims the pre-miRNA to generate mature double- stranded miRNAs of 21 to 30 nucleotides. The miRNA subsequently unwinds,  and the resulting single strands are incorporated into the multiprotein RISC.  Base  pairing  between  the  single-stranded  miRNA  and  its  target  mRNA  directs RISC to either cleave the mRNA target or to repress its translation.  In either case, the target mRNA gene is silenced posttranscriptionally. 

#45

Fig. 1.4  Roles  of  long  noncoding  RNAs  (lncRNAs).  (A)  Long  noncoding 

#46

RNAs (lncRNAs) can facilitate transcription factor binding and thus promote  gene activation. (B) Conversely, lncRNAs can preemptively bind transcription  factors and thus prevent gene transcription. (C) Histone and DNA modifica- tion by acetylases or methylases (or deacetylases and demethylases) may be  directed by the binding of lncRNAs. (D) In other instances lncRNAs may act  as scaffolding to stabilize secondary or tertiary structures and/or multisub- unit complexes that influence general chromatin architecture or gene activity.  (Adapted from Wang KC, Chang HY: Molecular mechanisms of long noncoding RNAs, Mol Cell 43:904, 2011.)

#47

Gene Editing

#48

Exciting new developments that permit exquisitely specific

#49

genome editing stand to usher in an era of molecular revo- lution. These advances come from a wholly unexpected source: the discovery of clustered regularly interspaced short palindromic repeats (CRISPRs) and Cas (or CRISPR- associated genes). These are linked genetic elements that endow prokaryotes with a form of acquired immunity to phages and plasmids. Bacteria use this system to sample the DNA of infecting agents, incorporating it into the host genome as CRISPRs. CRISPRs are transcribed and pro- cessed into an RNA sequence that binds and directs the nuclease Cas9 to a sequences (e.g., a phage), leading to its cleavage and the destruction of the phage. Gene editing repurposes this process by using artificial guide RNAs (gRNAs) that bind Cas9 and are complementary to a DNA sequence of interest. Once directed to the target sequence

#50