Humoral Mechanisms Provide An Additional Defensive Strategy
Microbicidal factors in secretions
Turning now to those defense systems that are mediated entirely by soluble pattern recognition molecules (Figure 1.2), we recollect that many microbes activate the complement system and may be lysed by the insertion of the membrane attack complex. The spread of infection may be limited by enzymes released through tissue injury that activate the clotting system. Of the soluble bactericidal substances elaborated by the body, perhaps the most abundant and widespread is the enzyme lysozyme, a muramidase that splits the exposed peptidoglycan wall of susceptible bacteria (see Figure 11.5).
Like the α‐defensins of the neutrophil granules, the human β‐defensins are peptides derived by proteolytic cleavage from larger precursors; they have β‐sheet structures, 29–40 amino acids, and three intramolecular disulfide bonds, although they differ from the α‐defensins in the placement of their six cysteines. The main human β‐defensin, hDB‐1, is produced abundantly in the kidney, the female reproductive tract, the oral gingiva, and especially the lung airways. As the word has it that we are all infected every day by tens of thousands of airborne bacteria, this must be an important defense mechanism. This being so, inhibition of hDB‐1 and of a second pulmonary defensin, hDB‐2, by high ionic strength could account for the susceptibility of cystic fibrosis patients to infection as they have an ion channel mutation that results in an elevated chloride concentration in airway surface fluids. Another airway antimicrobial active against Gram‐negative and Gram‐positive bacteria is LL‐37, a 37‐residue α‐helical peptide released by proteolysis of a cathelicidin (cathepsin L‐inhibitor) precursor.
This theme surfaces again in the stomach where a peptide split from lactoferrin by pepsin could provide the gastric and intestinal secretions with some antimicrobial policing. A rather longer two‐domain peptide with 107 residues, termed secretory leukocyte protease inhibitor (SLPI), is found in many human secretions. The C‐terminal domain is anti‐protease but the N‐terminal domain is distinctly unpleasant to metabolically active fungal cells and to various skin‐associated microorganisms, which makes its production by human keratinocytes particularly appropriate. In passing, it is worth pointing out that many d‐amino acid analogs of peptide antibiotics form left‐handed helices that retain the ability to induce membrane ion channels and hence their antimicrobial powers and, given their resistance to catabolism within the body, should be attractive candidates for a new breed of synthetic antibiotics.
Lastly, we may mention the two lung surfactant proteins SP‐A and SP‐D that, in conjunction with various lipids, lower the surface tension of the epithelial lining cells of the lung to keep the airways patent. They belong to a totally different structural group of molecules termed collectins (Figure 1.35) that contribute to innate immunity through binding of their lectin‐like domains to carbohydrates on microbes, and their collagenous stem to cognate receptors on phagocytic cells thereby facilitating the ingestion and killing of the infectious agents.
Acute phase proteins increase in response to infection
A number of plasma proteins collectively termed acute phase proteins show a dramatic increase in concentration in response to early “alarm” mediators such as macrophage‐derived interleukin‐1 (IL‐1) released as a result of infection or tissue injury. These include C‐reactive protein (CRP), mannose‐ binding lectin (MBL), and serum amyloid P component (Table 1.2). Expression levels of the latter proteins can increase by as much as 1000‐fold in response to proinflammatory cytokines such as IL‐1 and IL‐6. Other acute phase proteins showing a more modest rise in concentration include α1‐ antichymotrypsin, fibrinogen, ceruloplasmin, C9, and factor B.
The acute phase proteins are a relatively diverse group of proteins belonging to several different families (including, but not limited to, the pentraxin, collectin, and ficolin families) that have a number of functional effects in common. All of these proteins act as soluble pattern recognition molecules and are capable of binding directly to infectious agents to function as opsonins (i.e., “made ready for the table”), thereby enhancing uptake of microorganisms by macrophages and neutrophils. Many of these proteins also have the ability to activate complement and the assembly of a membrane attack complex. The ability to agglutinate microorganisms, thereby impeding their spread within the infected tissue, is another common theme. Some of these molecules can also form heterocomplexes, extending the range of PAMPs that can be detected.
These soluble pattern recognition molecules are frequently synthesized by activated macrophages upon stimulation of their pattern recognition receptors, or are stored within neutrophil granules available for immediate release via degranulation in response to infection. The liver is another major source of many acute phase proteins that are released into the circulation as a result of the systemic effects of the major proinflammatory cytokines IL‐1 and IL‐6. Let us look at some examples further.
Pentraxins, so‐called because these agents are made up of five identical subunits, constitute a superfamily of conserved proteins typified by a cyclic multimeric structure and a C‐terminal 200‐amino‐acid‐long pentraxin domain. CRP, serum amyloid P component (SAP), and pentraxin 3 are members of this family (Figure 1.36). Human CRP is composed of five identical polypeptide units noncovalently arranged as a cyclic pen- tamer around a calcium (Ca)‐binding cavity, was the first pentraxin to be described, and is the prototypic acute phase response protein. Pentraxins have been around in the animal kingdom for some time, as a closely related homolog, limulin, is present in the hemolymph of the horseshoe crab, not exactly a close relative of Homo sapiens. A major property of CRP is its ability to bind in a Ca‐dependent fashion, as a pattern recognition molecule, to a number of microorganisms that contain phosphorylcholine in their membranes, the complex having the useful property of activating complement (by the classical and not the alternative pathway with which we are at present familiar). This results in the deposition of C3b on the surface of the microbe that thus becomes opsonized for adherence to phagocytes.
SAP can complex with chondroitin sulfate, a cell matrix glycosaminoglycan, and subsequently bind lysosomal enzymes such as cathepsin B released within a focus of inflammation. The degraded SAP becomes a component of the amyloid fibrillar deposits that accompany chronic infections – it might even be a key initiator of amyloid deposition. SAP also binds several bacterial species via LPS and, similar to CRP, can also activate the classical complement pathway. CRP and SAP represent the main acute phase reactants in human and mouse, respectively.
Nine members of the collectin family have been described in vertebrates to date, the most intensively studied of which is mannose‐binding lectin (MBL). MBL can react not only with mannose but several other sugars, so enabling it to bind with an exceptionally wide variety of Gram‐negative and Gram‐ positive bacteria, yeasts, viruses, and parasites; its subsequent ability to trigger the classical C3 convertase through two novel associated serine proteases (MASP‐1 and MASP‐2) is the basis of what is known as the lectin pathway of complement activation. (Please relax, we unravel the secrets of the classical and lectin pathways in the next chapter.)
MBL is a multiple of trimeric complexes, each unit of which contains a collagen‐like region joined to a globular lectin‐binding domain (Figure 1.37). This structure places it in the family of collectins (collagen + lectin) that have the ability to recognize “foreign” carbohydrate patterns differing from “self ” surface polysaccharides, normally terminal galactose and sialic acid groups, whereas the collagen region can bind to and activate phagocytic cells through complementary receptors on their surface. The collectins, especially MBL and the alveolar surfactant molecules SP‐A and SP‐D mentioned earlier (Figure 1.35), have many attributes that qualify them for a first‐line role in innate immunity as soluble PRRs. These include the ability to differentiate self from nonself, to bind to a variety of microbes, to generate secondary effector mechanisms, and to be widely distributed throughout the body including mucosal secretions. They are of course the soluble counterparts to the cell surface C‐type lectin PRRs described earlier.
Interest in the collectin conglutinin has intensified with the demonstration, first, that it is found in humans and not just in cows, and second, that it can bind to N‐acetylglucosamine; being polyvalent, this implies an ability to coat bacteria with C3b by cross‐linking the available sugar residue in the complement fragment with the bacterial proteoglycan. Although it is not clear whether conglutinin is a member of the acute phase protein family, we mention it here because it embellishes the general idea that the evolution of lectin‐like molecules that bind to microbial rather than self polysaccharides, and which can then hitch themselves to the complement system or to phagocytic cells, has proved to be such a useful form of protection for the host.
These proteins are structurally and functionally related to col- lectins (Figure 1.38), and can also recognize carbohydrate‐ based PAMPs on microorganisms to activate the lectin pathway of complement activation. Ficolins typically recognize N‐acetylglucosamine residues in complex‐type carbohydrates in addition to other ligands. Three ficolins have been identified in humans, ficolin‐1, ‐2, and ‐3 (also known as M‐, L‐, and H‐ficolin, respectively), and a role as opsonins for the enhancement of phagocytosis has also been demonstrated for these proteins. Ficolins can also interact with CRP to widen the range of bacteria recognized by the latter and also to enhance complement‐mediated killing. The range of bacterial structures recognized by ficolins and MBL are complementary and recognize different but overlapping bacterial species.
Interferons inhibit viral replication
Recall from our earlier discussion of pattern recognition receptors (PRRs) that engagement of many of these receptors by PAMPs results in the production of cytokines and chemokines that act to amplify immune responses by binding to cells in the vicinity. An important class of cytokines induced by viral as well as bacterial infection is the type I interferons (IFNα and IFNβ). These are a family of broad‐spectrum antiviral agents present in birds, reptiles, and fish as well as the higher animals, and first recognized by the phenomenon of viral interference in which an animal infected with one virus resists superinfection by a second unrelated virus. Different molecular forms of interferon have been identified, the genes for all of which have been isolated. There are at least 14 different α‐interferons (IFNα) produced by leukocytes, while fibroblasts, and probably all cell types, synthesize IFNβ. We will keep a third type (IFNγ), which is not directly induced by viruses, up our sleeves for the moment.
Cells synthesize interferon when infected by a virus and secrete it into the extracellular fluid, where it binds to specific receptors on uninfected neighboring cells. As we saw earlier, engagement of several members of the TLR family, as well as the RIG‐like helicase receptors and the cytoplasmic DNA sensors, with their cognate PAMPs results in the induction of members of the interferon‐regulated factor (IRF) family of transcription factors (Figure 1.22 and Figure 1.23). In combination with NFkB, another transcription factor activated by engagement of several of the PRRs, IRFs induce expression of type I interferons that are secreted and bind to cells in the vicinity. Long double‐stranded RNA molecules, which are produced during the life cycle of most viruses, are particularly good inducers of interferons. The bound interferon now exerts its antiviral effect in the following way. At least two genes are thought to be derepressed in the interferon‐binding cell, allowing the synthesis of two new enzymes. The first, a protein kinase called protein kinase R (PKR), catalyzes the phosphorylation of a ribosomal protein and an initiation factor (eIF‐2) necessary for protein synthesis. The net effect of this is to dramatically reduce protein translation as a means of reducing the efficiency of virus production. Another gene product induced by interferons, oligoadenylate synthetase, catalyzes the formation of a short polymer of adenylic acid which activates a latent endoribonuclease; this in turn degrades both viral and host mRNA. This is another clever adaptation that is designed to reduce the production of viral products. Another consequence of the downturn in protein synthesis is a reduction in the expression of major histocompatibility complex (MHC) proteins, making cells susceptible to the effects of natural killer cells.
The net result is to establish a cordon of uninfectable cells around the site of virus infection, so restraining its spread. The effectiveness of interferon in vivo may be inferred from experiments in which mice injected with an antiserum to murine interferons could be killed by several hundred times less virus than was needed to kill the controls. However, it must be presumed that interferon plays a significant role in the recovery from, as distinct from the prevention of, viral infections.
As a group, the interferons may prove to have a wider bio- logical role than the control of viral infection. It will be clear, for example, that the induced enzymes described above would act to inhibit host cell division just as effectively as viral replication.