Edgar L. Milford M.D.1
Charles B. Carpenter M.D.2
1Associate Professor of Medicine, Harvard Medical School, Physician, Brigham and Women's Hospital
2Professor of Medicine, Harvard Medical School, Senior Physician, Brigham and Women's Hospital
The authors have no commercial relationships with manufacturers of products or providers of services discussed in this subsection.
Differences in genetic makeup from individual to individual have long been recognized to have physiologic consequences in both health and disease. The recent ability to do high-throughput sequencing of genes has revealed that many genes have variants that are present in a significant proportion of the population. Inherited variants of specific genes, either alone or in combination with other genes, may confer a differential risk of disease or of rejection of transplanted tissue.
The fundamental basis of genetic polymorphism in a population is variation of the nucleotide sequence of DNA at homologous locations in the genome. These differences in sequence can result from mutations involving a single nucleotide or from deletions or insertions of variable numbers of contiguous nucleotides. Each of these variants presumably occurred in a single ancestor in the distant past. Most new mutations are extinguished through random genetic drift and never become established in the population at any significant frequency. When the gene frequency of a mutation becomes established at more than 1% to 2%, it is often given the more dignified appellation of allele.
Allelic variants can occur anywhere in the genome. Some are found within coding regions of genes, and others are located in introns or gene regulatory regions. However, still others are found in areas that are not closely linked to any known expressed gene.
EQUILIBRIUM, DISEQUILIBRIUM, GENOTYPES, AND HAPLOTYPES
There can be multiple polymorphic nucleotide positions in or near an expressed gene on the same chromosome. In such cases, it is desirable to know whether specific variants at each of the polymorphic positions are independent of the variants at the other positions. If examination of a population shows that the variants at the different positions occur independently of one another, the system is said to be in Hardy-Weinberg equilibrium.1 If certain variants at one of the positions are statistically associated with specific variants at another of the linked positions, the system is said to exhibit linkage disequilibrium.1
Hardy-Weinberg equilibrium can be reestablished over many generations through recombination events. The closer the polymorphic loci are to each other on the chromosome, the less likelihood there is of a recombination and the more likely it is for the specific alleles at the two linked loci to be inherited en bloc as a haplotype. For example, if there are two polymorphic positions within a gene, each of which has two alleles, a given individual will have up to four definable alleles. These alleles are inherited as two parental haplotypes, each of which carries one allele from each of the two loci. Most methods used to type individuals cannot organize the genotype into haplotypes without additional information. The common assays simply define the genotype at each of the two polymorphic positions. Extensive population studies permit sophisticated maximum-likelihood estimates of haplotype frequencies within the population.2 These studies, combined with confirmatory cloning and sequencing studies of individual DNA strands, often reveal that some theoretically possible haplotypes never occur, whereas others can be assumed when a specific allele is present (because of linkage disequilibrium) [see Figure 1]. The ability to deduce haplotypes provides a much higher degree of specificity to the analysis of genetic polymorphism, because the haplotype more accurately defines a larger inherited region of DNA.
Figure 1. Variants of the Interleukin-10 Gene
Single-nucleotide variants occur at positions -1082, -819, and -592 in the promoter region of the interleukin-10 (IL-10) gene. Although eight variants are theoretically possible, only three of these potential IL-10 variants (in purple) are actually observed in large population studies. This is a consequence of strong linkage disequilibrium between the variants at those three positions.
TYPES OF GENETIC POLYMORPHISM
Single-nucleotide polymorphisms (SNPs) are allelic variants that have been generated as the result of conversion of one nucleotide to another at a homologous position. When present within a coding region (exon) of a gene, the expressed product may or may not have a single amino acid difference, depending on the resulting codon change. In some cases, the change can lead to either a nonsense codon or a stop codon, which halts the transcription process and results in the production of a truncated peptide. SNPs that are located in regulatory regions of an expressed gene can alter the transcription efficiency of that gene but not the protein sequence [see Figure 2].
Figure 2. TGF-β1 Gene Polymorphisms
Single-nucleotide polymorphisms have been identified in the gene for transforming growth factor-β1 (TGF-β1). Each polymorphism involves two alleles in the leader sequence of the gene. These biallelic nucleotide substitutions produce codon changes that result in alternative amino acids. Leu10 is in linkage disequilibrium with Arg25, and Pro10 is in linkage disequilibrium with Pro25. The Leu10Arg25 variant is associated with high TGF-β1 production, whereas the Pro10Pro25 variant is associated with lower production. This may be the consequence of different efficiency of posttranslational modification for the two variants, which differ only in the leader amino acid sequence.
Deletion or insertion mutants have also been found in functional genes, sometimes at frequencies that merit their inclusion as alleles. Again, the consequence of a deletion depends on the precise location of the deletion; whether it produces a nonsense frameshift; and whether it alters the function of the expressed product. Angiotensin-converting enzyme (ACE) represents a gene that has a deletion variant in which a 278-base-pair segment of intron 16 is excised. This deletion variant is associated with increased ACE levels.
Another class of allelic variance in association with a particular gene is short tandem repeat (STR) polymorphism. Short sequences of two to four base pairs at a given location can be duplicated back-to-back a specific number of times and inherited as a genetic variant. Because such variation would usually result in a nonsense codon, these STRs are almost always located in noncoding regions. The interferon gamma (IFN-γ) gene has such an STR within intron 1, in which the (CA) dinucleotide motif is repeated a variable number of times. The allele with (CA)12—that is, with 12 repeats of the CA motif—is associated with high IFN-γ production [see Figure 3].
Figure 3. IFN-γ Gene Variants and Expression
Illustration of a short tandem repeat (STR) polymorphism within intron 1 of the interferon gamma (IFN-γ) gene. STR polymorphisms in this intron differ according to the number of repetitions of the cytosine-arginine (CA) motif. The allelic variant with 12 tandem repeats [(CA12)] is associated with higher IFN-γ production.
METHODS OF DETECTION OF GENETIC POLYMORPHISM
DNA-based genotyping methods are rapid, accurate, and economical. SNPs can easily be detected, with a high degree of specificity and sensitivity. The assays depend on amplification of the polymorphic locus in question to produce sensitivity in the setting of a background of sample genomic DNA. Specificity is ensured by using tailored oligonucleotides that are complementary to the DNA sequence of the allele one wants to detect.
One strategy for typing is to use polymerase chain reaction to amplify a segment of DNA that includes the polymorphic position and a moderate amount of flanking DNA on both the 3′ and 5′ sides of the polymorphic position. This is done with primers that are complementary to conserved sequences in either side of the desired segment to be amplified. This yields an amplicon of known size that contains inherited alleles and is present in an amount that can be tested for the presence of specific alleles without significant interference from genomic DNA. The amplicon can then be probed, using a set of fluoresceinated or radiolabeled oligonucleotides, each of which is complementary to the DNA sequence of one of the possible alleles. This method is often referred to as site-specific oligonucleotide probe (SSOP) testing.
Another strategy for SNP typing, which does not require two steps, is called site-specific priming (SSP). This method takes advantage of the fact that the 3′ terminal base of a primer is where DNA synthesis commences during each cycle of PCR. For synthesis to proceed, the 3′ base must be closely bonded to its complementary base on the template DNA. Therefore, the terminal 3′ base of the primer can be used to render the PCR reaction itself exquisitely sensitive to the identity of the base that is on the template. For detection of SNPs, one can craft a set of PCR primers that are complementary to the alleles to be detected, with the terminal 3′ base of one of the primers located at the polymorphic position. The second PCR primer is usually complementary to a conserved segment of DNA and positioned to yield a product of a convenient size. If an allele is present, use of the appropriate set of primers will produce an amplicon. The amplicon can be separated from genomic DNA by simple agarose gel electrophoresis and identified by ethidium bromide staining under ultraviolet light, and the expected size can be confirmed.
Both SSOP and SSP can be modified to detect deletion or insertion variants. With SSP, using primers that flank the deletions or insertions, amplicons of characteristic sizes are produced. SSOP can confirm the presence or absence of the deletions or insertions through the use of probes that include the junctions of the deleted or inserted regions.
RELEVANCE OF GENETIC POLYMORPHISM IN HUMANS
Historically, polymorphisms in several genetic systems have been recognized as a barrier to transfusion and transplantation. The ABO blood group antigens were among the earliest genetically determined glycoproteins that exhibited mendelian inheritance and had biologic relevance in humans.3 Mismatch for the ABO antigens is a risk factor not only for transfusion reactions but also for solid-organ transplantation because of the prominent expression of these antigens on the vascular endothelium.
The major histocompatibility complex (MHC)—so called because of its prominent role in rejection of allogeneic tissue—is a primary barrier to transplantation of solid organs, tissue, and hematopoietic stem cells. This closely linked cluster of highly polymorphic genes, grouped on the short arm of chromosome 6, encodes cell surface molecules (human leukocyte antigens [HLA]). The normal role of the MHC is presentation of endogenous and exogenous peptide antigen fragments to T cells, thereby initiating an immune response against the molecule (or pathogenic organism) from which the peptide was derived.4 The extreme variability of molecular structure in the MHC antigens permits a wide range of different peptides to be presented by autologous human antigen-presenting cells, although some persons may have a specific repertoire of MHC antigens that do not present certain antigens effectively. The focused immunogenicity of MHC molecules and the variability of these molecules from person to person render them prominent targets for the immune response in the context of solid-organ and bone marrow transplantation. In cases in which live allogeneic cells are the target of the immune response, the apparent target is the nonself MHC molecule itself. Freedom from rejection and, in the case of bone marrow transplantation, graft versus host disease (GVHD) is improved with HLA matching of donor and recipient.
Innate and Adaptive Responses
It has become abundantly clear that the selective (adaptive) immunologic response, which is important in organ transplantation, tissue transplantation, and defense against certain microorganisms, is closely associated with the innate cellular and humoral pathways of nonspecific tissue injury, inflammation, hypoxia, and healing. Macrophages, for example, play a central role in the response to hypoxia, trauma, bacterial invasion, and inflammation caused by exogenous toxins, but they are also important in the processing and presentation of antigen to the specific immune system. Natural killer (NK) cells, which constitute approximately 10% of human mononuclear cells, are thought to be important mediators of innate immunity. Their cytolytic activity is regulated by inhibitory receptors, called killer immunoglobulin-like receptors (KIRs).5 Class I MHC molecules are ligands for the KIRs—in particular, genetically determined epitopes on HLA-B and HLA-C molecules that have limited polymorphism.5 In bone marrow transplantation, recipients who present the appropriate class I ligands to donor NK cells will downregulate the NK response. This is thought to decrease both GVHD and graft versus tumor activity.
Humans also have innate humoral immunity against a number of glycoprotein antigens. This so-called natural antibody is thought to have protective effects against a wide range of bacterial products. At the same time, the humoral immune system is able to mount a robust adaptive response to an astonishingly broad spectrum of specific antigens, if challenged to do so. The genes responsible for the adaptive immune response are highly polymorphic, but they are found only in specific subsets of T cells with antigen receptor genes that are rearranged during thymic development and in B cells that undergo somatic mutation in response to antigenic challenge. Specific germline variant alleles of the T cell receptor for antigen (before somatic mutation) are also associated with differential susceptibility to a number of immunologically mediated conditions, including renal allograft rejection and several rheumatic diseases, such as rheumatoid arthritis.6
Other Polymorphic Genes Involved in Organ and Tissue Injury
Variants of genes can influence organ and tissue physiology, directly induce diseases, or render the person more susceptible or resistant to a pathologic state. Variants that directly induce a profound disease state are usually rare in the population, because the disease may cause death before the person can reproduce. Variants or mutations that cause severe early disease are not discussed in this subsection. Polymorphic variants of loci that have a more subtle effect on disease susceptibility are more likely to become established in the population at frequencies of 1% or more (i.e., to become alleles). Several patterns can be appreciated with these alleles. Variant alleles may exhibit a gene-dose effect, with heterozygotes having an intermediate influence, between that of the normal genotype (the so-called wild type) and the homozygous variant genotype. In other cases, a variant allele appears to have a dominant influence; presumably, these variants are able to achieve significant frequency in the population because the condition they produce does not substantially decrease reproduction. The disease phenotype that is a measurable physiologic consequence of a particular genotype may be a downstream effect that depends on multiple influences, including the genotype in question, interaction with other genes, and environmental exposure.
Loci that encode cytokines, chemokine receptors, costimulatory molecules, and components of physiologically important pathways such as the angiotensin system are all concrete examples in which genetic polymorphism influences pathophysiology. These examples can be used to highlight some ways in which determination of individual genotype can assist in assessing risk of disease.
Cytokines and chemokines are secreted proteins and glycoproteins that act as important signaling devices in both the innate and the adaptive responses. They serve variously as chemoattractants and as inducers or suppressors of leukocyte, endothelial cell, platelet, fibroblast, and myocyte function. They have a particularly notable effect on cells that bear the appropriate receptors. Cytokines and chemokines often represent a common pathway that links the classical immune pathway and other pathways of tissue injury and repair, such as those involved in ischemia, trauma, and toxic damage.7
Costimulatory molecules such as CTLA-4 are expressed on the cell membranes of T cells and serve as ligands for complementary molecules on antigen-presenting cells [see 6:IX Immunologic Tolerance and Autoimmunity]. The engagement of costimulatory molecules with their ligands can augment or suppress the magnitude of the immune response induced by the recognition of antigen via the T cell receptor.8,9Soluble CTLA-4 has been used to block antigen-dependent T cell activation by competitive blockade of normal cell membrane-bound interaction.
Functional Consequences of Specific Genetic Variants
Variants in the genes that govern the production of cytokines such as interleukin-10 (IL-10), tumor necrosis factor-α (TNF-α), and transforming growth factor-β (TGF-β) can help determine whether a person has high or low levels of these cytokines.10 The cytokine network is thought to play an important role both in rejection of allografts and in tolerance,11 and a number of clinical effects of these polymorphisms in cytokine genes have now been described [see Tables 1 and 2].
Table 1 Cytokine Genetic Polymorphisms and Their Pathophysiologic Effects9,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59
Table 2 Renin-Angiotensin System and Chemokine Polymorphism and Pathophysiology26,60,61,62,63,64,65,66,67,68,69,70,71,72
TGF-β has two well-studied dimorphic positions within the leader sequence of the gene [see Figure 2]. These polymorphisms are in linkage disequilibrium; only two variants of the TGF-β gene have been described, rather than the four theoretically possible combinatorial variants. TGF-β is considered to be a major mediator of fibrosis in kidney and lung allografts.12,13 Specific variants of the TGF gene that result in high production of TGF-β (so-called high-producer genotypes) are associated with poor outcome in lung transplants: 98% of patients with chronic rejection are homozygous for the high-producer TGF-β genotype represented by Leu at position 10 and Arg at position 25. Moreover, fibrosis develops in the lung grafts of 93% of those with homozygous high-producer TGF-β genotype but only in 7% of those with heterozygous (high/low) producer genotype.12 TGF-β also mediates the gingival hypertrophy induced by the immunosuppressive agent cyclosporine. Increased gingival hypertrophy has been reported in patients with the low-producer TGF-β genotype, represented by Pro at both position 10 and position 25. Because the two variants differ only in the leader amino acid sequence, the different production levels may be the consequence of different efficiency of posttranslational modification.
ANGIOTENSIN SYSTEM POLYMORPHISMS
The renin-angiotensin system is a metabolic-hormonal pathway that plays a critical role in blood pressure homeostasis and salt and water balance. In the renin-angiotensin pathway, the prohormone angiotensinogen (AGT) is converted to angiotensin I by renin. Angiotensin-converting enzyme then catalyzes the conversion of angiotension I to angiotensin II [see Figure 4]. Angiotensin II is one of the most potent vasoconstrictive human hormones. In addition, angiotensin II has indirect inflammatory and fibrotic effects, which are distinct from its physiologic vasoconstrictive role. These indirect effects appear to be mediated by cytokines. Angiotensin II promotes the secretion of a number of inflammatory cytokines, including TGF-β, platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), IL-6, IL-12, TNF-α, and IFN-γ.14,15,16,17 There are two receptors for angiotensin II, type 1 (AT1) and type 2 (AT2). AT1 receptors mediate the major vasoconstrictive activity of angiotensin II but also appear to be involved in angiotensin II-dependent augmentation of immune activation and stimulation of TGF-β production. The AT2 receptors are implicated in remodeling; may promote angiotensin II-dependent apoptosis; and have some functions that oppose the AT1 receptor, including vasodilation and increased production of nitric oxide.18,19
Figure 4. Renin-angiotensin System
The renin-angiotensin system is illustrated, along with proven variants of genes responsible for its components. The variant genes indicated are thought to result in a quantitative increase in function in the system. The final hormone, angiotensin II (AII), has a variety of vasoactive, inflammatory, or anti-inflammatory effects, which appear to be dependent on the receptor that is engaged. (ACE—angiotensin-converting enzyme; AGT—angiotensinogen; AT1—angiotensin type 1 receptor; AT2—angiotensin type 2 receptor).
Several of the genes encoding members of the renin-angiotensin pathway exhibit polymorphisms that influence function. Genomic variants of the genes encoding AGT, ACE, AT1, and AT2 have been described.20 There is evidence that the AGT(A/A) and ACE(D/D) variants result in increased an-giotensin II activity; in turn, the angiotensin II can interact differentially with receptors of different genotypes and influence ultimate pathophysiology. A deletion variant of the ACE gene (D14091-14378) and a single-nucleotide polymorphic variant of the AGT gene (G → A, -6) are correlated with increased peripheral ACE activity and AGT levels, respectively.21,22 Both genotypes confer increased susceptibility to hypertension, and ACE(D14091-14378) also worsens ischemic heart disease and progression of intrinsic renal insufficiency [see Table 2].23 An analysis of ACE polymorphism in diabetes revealed that the ACE(D) allele is highly associated with diabetic nephropathy.24
Of the different classes of white blood cells, T cells contain the highest level of ACE, approximately 28-fold more than monocytes. Indeed, immunologically competent T cells appear to be the major cell type expressing ACE in blood.25 The ACE expression can vary up to 100-fold during the differentiation of T cells. Monocytes express angiotensin II, the final product of the angiotensin synthetic pathway. Monocyte angiotensin II appears to mediate recruitment of inflammatory cells during renal damage through the synthesis of monocyte chemoattractant protein-1.26
A variant of the AT2 gene (A → G, 1332) has been associated with congenital anomalies of the kidney and urinary tract. These developmental abnormalities are preceded by delayed apoptosis of undifferentiated mesenchymal cells surrounding the urinary tract during key ontogenic events.27 In kidney transplant recipients, specific variants of the ACE and AGT genes are correlated with poor clinical outcomes. Renal transplant patients who have either the ACE(D14091-14378) or the AGT homozygous (G → A, -6)/(G → A, -6) variant have poorer renal transplant function at 3 years, as well as more rapid progression of transplant failure, defined as an increase of serum creatinine levels over time. Diastolic blood pressure in these patients was also significantly higher as a function of the AT1(A → C, 1166) C gene dose. The pathophysiologic reasons for the association between specific angiotensin system gene polymorphisms and renal transplant outcomes are not well understood. Further work is needed to reveal the degree to which this association is a function of hypertensive organ damage or modulation of the immunologic response mediated by angiotensin.
CHEMOKINE RECEPTOR POLYMORPHISMS
Chemokines are molecules with a variety of functions, some of which influence the recruitment of inflammatory cells to sites of injury. Three of the genes encoding chemokine receptors are located on one chromosome. CCR2 and CCR5 are located within 20,000 base pairs of each other; CX3CR1 is located 10 million base pairs away from CCR5 [see Figure 5].
Figure 5. Polymorphisms in Chemokine Genes
Locations of chemokine receptor genes and variant positions associated with those genes are shown. The CCR2, CCR5, and CX3CR1 genes are located on the same chromosome. CX3CR1 has two variable positions within the coding region. The location of the deletion variant of CCR5 and the location of the G-to-A variant of CCR5 in the 5′ promoter region of the gene are illustrated. The CCR2 gene has a G-to-A variant located within the coding region.
The leukocyte chemokine receptor CCR5 is expressed on monocytes, as well as on helper T cells involved in augmentation of the immune response (TH1 subset)28 [see 6:X Allergic Response]. CCR5 is a coreceptor for entry of HIV-1 into macrophages. CCR5 binds the inflammatory chemokines RANTES (regulated on activation, normal T cell expressed and repeated), macrophage inflammatory protein (MIP)-1α, and MIP-1β, whereas CCR2 and CX3CR1 are receptors for the chemokines monocyte chemoattractant protein (MCP)-1 and fractalkine, respectively. Antagonists of CCR5, such as met-RANTES, prolong renal allograft survival in MHC-incompatible mice. Furthermore, prolonged heart transplant survival is achieved if the recipient is a homozygous CCR2 or CCR5 knockout. In humans, a 32-base-pair deletion of the CCR5 gene (CCR5 Δ32) renders the gene nonfunctional. There is also a polymorphic single-nucleotide variant of CCR5, CCR5-9029(G → A), which is located in the promoter region of the gene. The G variant is associated with defective transcription. Gene variants of some of these chemokine receptors have been associated with different rates of HIV disease progression.29 The CCR5 Δ32 variant is associated with lower incidence and severity of asthma and rheumatoid arthritis.30 Patients with homozygosity for CX3CR1-V2491I(G → A), CX3CR1-T280M(C → T), and CCR5-9029(G → A) have higher HIV progression rates. In contrast, patients with CCR2V64I(G → A) and CCR5(Δ32) exhibit slower progression of HIV, presumably because of reduced binding of the virus to target cells.
In renal transplant patients, the A/A homozygous genotype of the CCR59029(A → G) polymorphic locus is associated with significantly lower incidence of acute rejection episodes in the first posttransplant year. Although this could be explained by a protective effect of A/A homozygosity, it might instead be from a dominant detrimental effect of the G variant, given that both A/G heterozygotes and G/G homozygotes have been found to have similarly high rejection frequencies, which were twice that of patients with A/A genotype.14
Practical Applications of Genotyping Polymorphisms
The genetic polymorphisms discussed in this subsection represent but examples of the many inherited variants of physiologically important genes that influence susceptibility to disease. These variants can act alone, in conjunction with variants at other loci, or through interaction with environmental factors to increase or decrease disease incidence or severity. The cytokine genes, chemokine genes, and genes of the renin-angiotensin system are important modulators of the immune response and, in the case of the renin-angiotensin axis, of hypertension and vascular disease. Knowledge of a patient's genotype may assist physicians in assessing prior risk of a pathophysiologic outcome and in tailoring therapy. Clinical trials may, in some cases, be better interpreted by knowledge of participant genotypes, because certain genotypes may impart differential incidence of disease and responsiveness to pharmacologic agents.
Figures 1 through 5 Seward Hung.
Editors: Dale, David C.; Federman, Daniel D.