Immunological responses involving IgG antibodies or specific T cells can also cause adverse hypersensitivity reactions. Although these effector arms of the immune response normally participate in protective immunity to infection, they occasionally react with noninfectious antigens to produce acute or chronic hypersensitivity reactions. We will describe common examples of such reactions in this part of the chapter.
12-15. Innocuous antigens can cause type II hypersensitivity reactions in susceptible individuals by binding to the surfaces of circulating blood cells
Antibody-mediated destruction of red blood cells (hemolytic anemia) or platelets (thrombocytopenia) is an uncommon side-effect associated with the intake of certain drugs such as the antibiotic penicillin, the anti-cardiac arrhythmia drug quinidine, or the antihypertensive agent methyldopa. These are examples of type II hypersensitivity reactions in which the drug binds to the cell surface and serves as a target for anti-drug IgG antibodies that cause destruction of the cell (see Fig. 12.2). The anti-drug antibodies are made in only a minority of individuals and it is not clear why these individuals make them. The cell-bound antibody triggers clearance of the cell from the circulation, predominantly by tissue macrophages in the spleen, which bear Fcγ receptors.
12-16. Systemic disease caused by immune complex formation can follow the administration of large quantities of poorly catabolized antigens
Type III hypersensitivity reactions can arise with soluble antigens. The pathology is caused by the deposition of antigen:antibody aggregates or immune complexes at certain tissue sites. Immune complexes are generated in all antibody responses but their pathogenic potential is determined, in part, by their size and the amount, affinity, and isotype of the responding antibody. Larger aggregates fix complement and are readily cleared from the circulation by the mononuclear phagocytic system. The small complexes that form at antigen excess, however, tend to deposit in blood vessel walls. There they can ligate Fc receptors on leukocytes, leading to leukocyte activation and tissue injury.
A local type III hypersensitivity reaction can be triggered in the skin of sensitized individuals who possess IgG antibodies against the sensitizing antigen. When antigen is injected into the skin, circulating IgG antibody that has diffused into the tissues forms immune complexes locally. The immune complexes bind Fc receptors on mast cells and other leukocytes, which creates a local inflammatory response with increased vascular permeability. The enhanced vascular permeability allows fluid and cells, especially polymorphonuclear leukocytes, to enter the site from the local vessels. This reaction is called an Arthus reaction (Fig. 12.19). The immune complexes also activate complement, releasing C5a, which contributes to the inflammatory reaction by ligating C5a receptors on leukocytes (see Sections 2-12 and 6-16). This causes their activation and chemotactic attraction to the site of inflammation. The Arthus reaction is absent in mice lacking the α or γ chain of the FcγRIII receptor (CD16) on mast cells, but remains largely unperturbed in complementdeficient mice, showing the primary importance of FcγRIII in triggering inflammatory responses via immune complexes.
The deposition of immune complexes in local tissues causes a local inflammatory response known as an Arthus reaction (type III hypersensitivity reaction). In individuals who have already made IgG antibody against an antigen, the same antigen injected (more...)
A systemic type III hypersensitivity reaction, known as serum sickness (Drug-Induced Serum Sickness, in Case Studies in Immunology, see Preface for details), can result from the injection of large quantities of a poorly catabolized foreign antigen. This illness was so named because it frequently followed the administration of therapeutic horse antiserum. In the preantibiotic era, antiserum made by immunizing horses was often used to treat pneumococcal pneumonia; the specific anti-pneumococcal antibodies in the horse serum would help the patient to clear the infection. In much the same way, antivenin (serum from horses immunized with snake venoms) is still used today as a source of neutralizing antibodies to treat people suffering from the bites of poisonous snakes.
Serum sickness occurs 7–10 days after the injection of the horse serum, an interval that corresponds to the time required to mount a primary immune response that switches from IgM to IgGantibody against the foreign antigens in horse serum. The clinical features of serum sickness are chills, fever, rash, arthritis, and sometimes glomerulonephritis. Urticaria is a prominent feature of the rash, implying a role for histamine derived from mast-cell degranulation. In this case the mast-cell degranulation is triggered by the ligation of cellsurface FcγRIII by IgG-containing immune complexes.
The course of serum sickness is illustrated in Fig. 12.20. The onset of disease coincides with the development of antibodies against the abundant soluble proteins in the foreign serum; these antibodies form immune complexes with their antigens throughout the body. These immune complexes fix complement and can bind to and activate leukocytes bearing Fc and complement receptors; these in turn cause widespread tissue injury. The formation of immune complexes causes clearance of the foreign antigen and so serum sickness is usually a self-limiting disease. Serum sickness after a second dose of antigen follows the kinetics of a secondary antibody response and the onset of disease occurs typically within a day or two. Serum sickness is nowadays seen after the use of anti-lymphocyte globulin, employed as an immunosuppressive agent in transplant recipients, and also, rarely, after the administration of streptokinase, a bacterial enzyme that is used as a thrombolytic agent to treat patients with a myocardial infarction or heart attack.
Serum sickness is a classic example of a transient immune complex-mediated syndrome. An injection of a foreign protein or proteins leads to an antibody response. These antibodies form immune complexes with the circulating foreign proteins. The complexes (more...)
A similar type of immunopathological response is seen in two other situations in which antigen persists. The first is when an adaptive antibody response fails to clear an infectious agent, for example in subacute bacterial endocarditis or chronic viral hepatitis. In this situation, the multiplying bacteria or viruses are continuously generating new antigen in the presence of a persistent antibody response that fails to eliminate the organism. Immune complex disease ensues, with injury to small blood vessels in many tissues and organs, including the skin, kidneys, and nerves. Immune complexes also form in autoimmune diseases such as systemic lupus erythematosus where, because the antigen persists, the deposition of immune complexes continues, and serious disease can result (see Section 13-7).
Some inhaled allergens provoke IgG rather than IgEantibody responses, perhaps because they are present at relatively high levels in inhaled air. When a person is reexposed to high doses of such inhaled antigens, immune complexes form in the alveolar wall of the lung. This leads to the accumulation of fluid, protein, and cells in the alveolar wall, slowing blood-gas interchange and compromising lung function. This type of reaction occurs in certain occupations such as farming, where there is repeated exposure to hay dust or mold spores. The disease that results is therefore called farmer's lung. If exposure to antigen is sustained, the alveolar membranes can become permanently damaged.
12-17. Delayed-type hypersensitivity reactions are mediated by TH1 cells and CD8 cytotoxic T cells
Unlike the immediate hypersensitivity reactions described so far, which are mediated by antibodies, delayed-type hypersensitivity or type IV hypersensitivity reactions are mediated by antigen-specific effector T cells. These function in essentially the same way as during a response to an infectious pathogen, as described in Chapter 8. The causes and consequences of some syndromes in which type IV hypersensitivity responses predominate are listed in Fig. 12.21. These responses can be transferred between experimental animals by purified T cells or cloned T-cell lines.
Type IV hypersensitivity responses. These reactions are mediated by T cells and all take some time to develop. They can be grouped into three syndromes, according to the route by which antigen passes into the body. In delayed-type hypersensitivity the (more...)
The prototypic delayed-type hypersensitivity reaction is an artifact of modern medicine—the tuberculin test (see Appendix I, Section A-38). This is used to determine whether an individual has previously been infected with Mycobacterium tuberculosis. Small amounts of tuberculin—a complex mixture of peptides and carbohydrates derived from M. tuberculosis—are injected intradermally. In individuals who have previously been exposed to the bacterium, either by infection with the pathogen or by immunization with BCG, an attenuated form of M. tuberculosis, a local T cell-mediated inflammatory reaction evolves over 24–72 hours. The response is mediated by TH1 cells, which enter the site of antigen injection, recognize complexes of peptide:MHC class II molecules on antigen-presenting cells, and release inflammatory cytokines, such as IFN-γ and TNF-β. The cytokines stimulate the expression of adhesion molecules on endothelium and increase local blood vessel permeability, allowing plasma and accessory cells to enter the site; this causes a visible swelling (Fig. 12.22). Each of these phases takes several hours and so the fully developed response appears only 24–48 hours after challenge. The cytokines produced by the activated TH1 cells and their actions are shown in Fig. 12.23.
The stages of a delayed-type hypersensitivity reaction. The first phase involves uptake, processing, and presentation of the antigen by local antigen-presenting cells. In the second phase, TH1 cells that were primed by a previous exposure to the antigen (more...)
The delayed-type (type IV) hypersensitivity response is directed by chemokines and cytokines released by TH1 cells stimulated by antigen. Antigen in the local tissues is processed by antigen-presenting cells and presented on MHC class II molecules. Antigen-specific T (more...)
Very similar reactions are observed in several cutaneous hypersensitivity responses. These can be elicited by either CD4 or CD8 T cells, depending on the pathway by which the antigen is processed. Typical antigens that cause cutaneous hypersensitivity responses are highly reactive small molecules that can easily penetrate intact skin, especially if they cause itching that leads to scratching. These chemicals then react with self proteins, creating protein-hapten complexes that can be processed to hapten-peptide complexes, which can bind to MHC molecules that are recognized by T cells as foreign antigens. There are two phases to a cutaneous hypersensitivity response—sensitization and elicitation. During the sensitization phase, cutaneous Langerhans' cells take up and process antigen, and migrate to regional lymph nodes, where they activate T cells (see Fig. 8.15), with the consequent production of memory T cells, which end up in the dermis. In the elicitation phase, further exposure to the sensitizing chemical leads to antigen presentation to memory T cells in the dermis, with release of T-cell cytokines such as IFN-γ and IL-17. This stimulates the keratinocytes of the epidermis to release cytokines such as IL-1, IL-6, TNF-α and GM-CSF, and CXC chemokines including IL-8, interferon-inducible protein (IP)-9, IP-10, and MIG (monokine induced by IFN-γ). These cytokines and chemokines enhance the inflammatory response by inducing the migration of monocytes into the lesion and their maturation into macrophages, and by attracting more T cells (Fig. 12.24).
Elicitation of a delayed-type hypersensitivity response to a contact-sensitizing agent. The contact-sensitizing agent is a small highly reactive molecule that can easily penetrate intact skin. It binds covalently as a hapten to a variety of endogenous (more...)
The rash produced by contact with poison ivy (Fig. 12.25) is caused by a T-cell response to a chemical in the poison ivy leaf called pentadecacatechol. This compound is lipid-soluble and can therefore cross the cell membrane and modify intracellular proteins. These modified proteins generate modified peptides within the cytosol, which are translocated into the endoplasmic reticulum and are delivered to the cell surface by MHC class I molecules. These are recognized by CD8 T cells, which can cause damage either by killing the eliciting cell or by secreting cytokines such as IFN-γ. The well-studied chemical picryl chloride produces a CD4 T-cell hypersensitivity reaction. It modifies extracellular self proteins, which are then processed by the exogenous pathway (see Section 5-5) into modified self peptides that bind to self MHC class II molecules and are recognized by TH1 cells. When sensitized TH1 cells recognize these complexes they can produce extensive inflammation by activating macrophages (see Fig. 12.24). As the chemicals in these examples are delivered by contact with the skin, the rash that follows is called a contact hypersensitivity reaction (Contact Sensitivity to Poison Ivy, in Case Studies in Immunology, see Preface for details).
Blistering skin lesions on hand of patient with poison ivy contact dermatitis. Photograph courtesy of R. Geha.
Some insect proteins also elicit delayed-type hypersensitivity response. However, the early phases of the host reaction to an insect bite are often IgE-mediated or the result of the direct effects of insect venoms. Important delayed-type hypersensitivity responses to divalent cations such as nickel have also been observed. These divalent cations can alter the conformation or the peptide binding of MHC class II molecules, and thus provoke a T-cell response. Finally, although this section has focused on the role of T cells in inducing delayed-type hypersensitivity reactions, there is evidence that antibody and complement may also play a part. Mice deficient in B cells, antibody, or complement show impaired contact hypersensitivity reactions. These requirements for B cells, antibody, and complement may reflect their role in the early steps of the elicitation of these reactions.
Hypersensitivity diseases reflect normal immune mechanisms directed against innocuous antigens. They can be mediated by IgG antibodies bound to modified cell surfaces, or by complexes of antibodies bound to poorly catabolized antigens, as occurs in serum sickness. Hypersensitivity reactions mediated by T cells can be activated by modified self proteins, or by injected proteins such as those in the mycobacterial extract tuberculin. These T cell-mediated responses require the induced synthesis of effector molecules and develop more slowly, which is why they are termed delayed-type hyper-sensitivity.
In spite of the widespread use of amalgam as a posterior restorative material, reported cases of hypersensitivity to amalgam are relatively infrequent. By far the most common type is the delayed oral lichenoid reaction (OLR).6,8 Essentially this involves a cell- mediated, type IV hypersensitivity response to a constituent of the amalgam restoration and as such is the oral equivalent of skin allergic contact dermatitis. Most often the allergen is mercury but occasionally the response is to one of the other components of amalgam alloy such as copper, tin or zinc. The lesions of OLR are similar to those of LP. However, they can be distinguished from the lesions of LP by their close relationship with amalgam restorations, and their tendency to be localised and asymmetrically distributed.9 In contrast the lesions of classical LP tend to be more widespread, bilateral and symmetrical in distribution. As with LP, OLRs may have reticular, plaque-like, atrophic and erosive components (Table 1). A positive patch test to mercury or another component of amalgam may help to confirm the diagnosis. Final confirmation, however, may have to await resolution of the lesion following removal of the offending amalgam restoration.
OLRs may also occur in response to a number of other allergens including drugs such as antimalarials, antihypertensive agents, oral hypoglycaemics and non steroidal anti-inflammatory agents,10 as well as palladium-based crowns11 and composite resins.12 It is, therefore, not surprising that reports of mercury sensitivity on patch-testing in OLR have varied between 12–62%.3 Patients in whom the OLR is intimately associated with an amalgam restoration are much more likely to have a positive mercury sensitivity response3 than those with more extensive or non contacting lesions.13 Similarly, the reported benefit of replacing amalgam restorations in patients with LP or OLRs has been variable14 but has been most effective in those patients with OLR lesions in direct contact with amalgam fillings. Bratel et al.15 demonstrated improvement in 95% of such patients treated. The time course of resolution is also variable. Ibbotson et al.14 found that the lesions resolved in 16 out of 17 patients with patch test positive OLR to amalgam within 12 months of replacement, with a mean time of 2 months. Providing that there is no amalgam in physical contact with the mucosa there is no need to replace all the amalgam restorations in the mouth and, as in case 1, amalgam can even be left in situ as a core for a crown.
It is not known why particular individuals should develop lichenoid reactions to amalgam. Many patients have had amalgam restorations in their mouths without problems for many years before becoming sensitised. Once sensitised they then consistently develop lesions wherever amalgam is in prolonged contact with the mucosa.
To become sensitised, patients require prior exposure to the antigen. In the case of mercury this commonly occurs through exposure to dental amalgam but may also involve other sources of mercury notably disinfectants, cosmetics, dyes, foods and vaccine preservatives.7 Sensitising allergens are usually highly reactive molecules of relatively low molecular weight (<1 kDa) which bind covalently to skin or tissue proteins prior to becoming immunogenic. In the case of mercury it is believed that sensitised mercury specific T cells react with, and damage, basal keratinocytes within the epithelium that present mercury-peptide complexes, or mercury alone, bound to MHC molecules on their surface.7 Skin patch testing is a useful way of identifying the allergen responsible for such cell mediated, delayed, type IV hypersensitivity responses and typically results in a positive response after 72–96 hours. Although it is a relatively straightforward procedure, it needs to be performed at a specialist dermatology, allergy or oral medicine centre for correct interpretation.
More acute and generalised hypersensitivity reactions to the components of amalgam are far less common. In a recent review, Veron et al.16 found that of 41 cases of amalgam allergy with systemic features reported between 1905–1986, 37 were due to mercury hypersensitivity. Of the remainder, 2 were due to copper and 2 due to silver. In contrast to the more delayed amalgam-associated OLRs, inhalation or absorption of mercury vapour leads to the development of a cutaneous, erythematous, urticarial rash affecting the face and limbs, usually on their flexural aspect.17 Often, as in case 2, the rash is on the same side of the body as the dental intervention. Acute oral manifestations of exposure to mercury or amalgam are much less common than cutaneous lesions18,19 and may present as a burning sensation associated with the formation of vesicles which rupture to leave areas of erosion.19 Frykholm20 found that only one in seven patients with mercury allergy developed oral signs and symptoms whereas all had a cutaneous reaction. Reports have often been imprecise on the time course of the reaction but they tend to occur relatively rapidly in comparison to OLRs, usually within hours21,22,23,24,25,26 of placement or removal of an amalgam filling. As with patient 2 these reactions are usually self-limiting and completely resolve over a period of a few days,25 presumably as the release of free mercury from the amalgam restoration diminishes.
The precise nature of these acute allergic responses is not clear. However, the rapid onset, urticarial rash and more widespread nature of the response, suggest that a type I hypersensitivity response may be involved. Type I immediate hypersensitivity reactions tend to occur rapidly and are caused by antigen binding to, and cross linking, allergen specific IgE or IgG4 on the surface of mast cells. This causes the mast cells to degranulate releasing histamine and other acute inflammatory mediators. If localised, this results in an urticarial rash and other local changes. However, if more severe, the effects may be more widespread with oedema, tachycardia and respiratory difficulty.
Skin patch testing is not designed for investigating type I hypersensitivity responses, although reactions within the first 24 hours of applying skin patch test allergens may occur in sensitised individuals. Normally, type I hypersensitivity responses are detected using the in-vitro radioallergosorbent test (RAST) for antigen-specific IgE antibodies or by skin prick testing. However, the small size of the mercury molecule, the lack of its conjugate and its toxicity, precludes performing these assays and to our knowledge no center currently offers a RAST for mercury. This makes it difficult to confirm that acute reponses to mercury in amalgam are the result of a type I hypersensitivity response. Indeed, it is possible that some cases, particularly those that are very localised or have a more prolonged time course, represent a mixed or more acute type IV response. In general, however, the clinical features help to clearly distinguish between acute responses to mercury and chronic lichenoid reactions (Table 1).
Since lesions associated with acute hypersensitivity reactions to mercury are generally self-limiting and resolve after a few days, existing sound amalgam restorations may be left in situ. However, alternative materials should be used for new restorations. When removal of amalgam restorations is necessary, it should be performed using rubber dam, water-spray, and high-volume suction to minimise exposure to any mercury that may be liberated.17,27 Antihistamine cover may also be beneficial during amalgam removal in those where the response is thought to be due to type I hypersensitivity.