Table of contents :

• mechanical factors
• chemical factors
• biological factors
• phagocytic cells
• pattern recognition receptors (PRRs) for pathogen-associated molecular patterns (PAMPs)
• secreted PRRs
• cell-surface PRRs
• Toll-like receptors (TLRs)
• CD1 family members
• intracellular PRRs
• lectin complement activation pathway
• alternative complement activation pathway
• the genome's immune system

• mechanical factors
• stopping
• mucus (high viscosity impairs diffusion rate)
• nasal exudate
• tracheobronchial mucus
• gastrointestinal mucus
• cervicovaginal mucus
Mucins are the main component of the mucus protecting the internal epithelial layers of our body. They are characterized by dense O-glycosylation in tandem repeat domains that are rich in serine, threonine and proline. Mucins carry a large repertoire of O-oligosaccharides, which are believed to provide attachment sites for microorganisms, as well as being implicated in events such as leukocyte migration from the bloodstream. In diseases such as cystic fibrosis and cancer, altered mucin O-glycosylation has been observed. Despite a large body of research on mucin oligosaccharides, defining mucin glycoforms is an enormously challenging task but is nevertheless an essential prerequisite for addressing fundamental questions of mucin function in helath and diseases. Online LC-ES-MS/MS has been used to sequence sulfated mucin oligosaccharides purified from porcine large intestine. Information on core sequences and terminal blood-group antigenic determinants was obtained for 28 different oligosaccharides at a sensitivity appropriate for biological studies. A combination of Q-TOF and Edman sequencing of partially deglycosylated glycopeptides, prepared from the tandem repeat region of MUC1 derived from a breast cancer cell line, has provided evidence that all 5 potential glycosylation sites in the tandem repeat are O-glycosylation targets.
• mucin 1 (MUC1), transmembrane
• mucin 2 (MUC2), intestinal/tracheal
• mucin 3A (MUC3A), intestinal
• mucin 3B (MUC3B)
• mucin 4 (MUC4), tracheobronchial
• mucin 5, subtypes A (MUC5A) and C (MUC5C)
• mucin 5, subtype B (MUC5B), tracheobronchial / mucin 10 (MUC10) (endocervical epithelium, peaks before midcycle)
• mucin 6 (MUC6), gastric
• mucin 7 (MUC7), salivary
• mucin 8 (MUC8), tracheobronchial
• mucin 9 / oviductal glycoprotein 1, 120kDa / oviductin
• mucin 11 (MUC11)
• mucin 12 (MUC12)
• mucin 13 (MUC13), epithelial transmembrane
• mucin 15 (MUC15)
• mucin 16 (MUC16)
• mucin 17 (MUC17)
• mucin and cadherin-like (MUCDHL)
• small breast epithelial mucin
• CD164 antigen, sialomucin
• Helicobacter pylori is rarely found in deeper portions of the gastric mucosa, where O-glycans are expressed in gastric mucin that have terminal 1,4-linked N-acetylglucosamine : these O-glycans have antimicrobial activity against H. pylori, inhibiting its biosynthesis of cholesteryl-D-glucopyranoside, a major cell wall component^ref
• epiglottis closure reflex [decreased by ethanol intoxication or central anaesthesics and analgesics]
• blinking reflex
• peritoneum (it sticks over a perforation in bowels preventing spread of lumenal microorganisms)
• lining epithelia
• keratinized epithelia : keratin is a protein present in all cuticular structures of the body, such as hair, epidermis, and the organic matrix of the enamel of the teeth (and horns in other species).
• mucosal epithelia
• uterine epithelium during pregnancy synthetise M-CSF / CSF-1 which stimulate trophoblast to synthesize the neutrophil chemoattractants CXCL1 and CXCL2 to organize the maternal immune response to bacterial infection at the utero–placental interface
• turn-overs of lining epithelia
• skin desquamation
• mucosal desquamation
• flows
• tear flushing [decreased in xerophthalmia]
• respiratory tract flows (expiration, kinociliated epithelium, coughing, sneezing and swallowing) [decreased in smokers]
• urinary tract flows [decreased in prostatic hypertrophy and urethral stenosis]
• menstruations [decreased after menopause]
• gastrointestinal tract flows (swallowing => peristalsis => defecation : stools eliminate ~ 10¹² Bacteria / die) [decreased during constipation and intestinal occlusions]
Fibronectin binding proteins (FnBP) on the surface of Staphylococcus aureus have previously been shown to mediate adherence of the organism to resting endothelial cells in static adhesion assays but physiologic levels of shear stress prevent FnBP-mediated adhesion to resting endothelial cells : a tip-pulling velocity of 1 mm/s after an Fn-adhesin interaction had been established, determined a de-adhesion force of 85 pN.
• chemical factors
• pH
• acidic pH
• pH 5.5 on skin and in sweat
• pH 1.2÷3 in gastric juice [decreased in individuals with gastroresecation or achlorhydria]
• pH 4.5 in vagina (thanks to Lactobacillus acidophilus that ferments estrogen-induced glycogen storages) [decreased in puberty, xerovaginia, at the beginning of menstruations, at the beginning of pregnancy and after menopause]
• pH 5.5÷6 in pus
• pH 4.5-7 in urine [decreased in urinary pH alterations]
• basic pH
• pH 8 in pancreatic juice
• unsaturated fatty acids (in sebum)
• bile
• Zn²⁺ (in prostatic juice)
• hyperosmolarity of kidney medulla
• reactive oxygen species (ROS) / reactive oxygen intermediates (ROIs) : the reactivity of individual oxygen-derived molecules differs greatly.
• as a relatively strong reductant, superoxide can function either as a reductant or an oxidant, depending on the oxidation-reduction potential of the reacting molecule. Although it is a precursor to more reactive species, superoxide reacts with a limited repertoire of chemical targets, such as the iron-sulphur centres that function as electron carriers in respiratory chains of bacteria and mitochondria.
• hydrogen peroxide (H₂O₂) is a more potent oxidant and is more reactive, although its targets are still rather limited, and include methionine and certain highly reactive cysteine residues such as those found in the active sites of some enzymes. Oxidation of such cysteines inactivates enzymes such as protein tyrosine phosphatases^ref. Peroxidases use H₂O₂ to produce highly reactive oxidants either at the active site or as discrete diffusable oxidants such as hypochlorous acid (HOCl). Myeloperoxidase (MPO) is an abundant heme protein^{ref1, ref2} released during the oxidative burst by activated neutrophils and monocytes. A major function of MPO is to hold a central role in microbial killing, and recent findings revealed an association between MPO levels and the risk of coronary artery disease^ref. MPO is also present in tissue macrophages such as those in vascular lesions^{ref1, ref2, ref3}. The MPO-hydrogen peroxide system plays a specific role in monocyte/macrophage-mediated oxidation of LDL by 3 major pathways. First, MPO catalyzes oxidation of L-tyrosine, generating the tyrosyl radical that may initiate dityrosine cross-linking of proteins or initiate LDL lipid peroxidation^ref. Second, MPO may use nitrite, the major end product of nitric oxide radical metabolism, as a substrate to nitrate (lipo)protein tyrosine residues and to initiate lipid peroxidation^{ref1, ref2}. MPO may also generate a nitrating intermediate through secondary reaction of hypochlorous acid/hypochlorite (HOCl/OCl^–) with nitrite, presumably forming nitryl chloride as a reactive intermediate^ref. Third, because of its high concentrations in biological matrices, chloride is the preferred substrate for MPO, and HOCl/OCl^–, a potent chlorinating oxidant, is formed. Under acidic conditions chlorine gas is formed, leading to generation of chlorotyrosine^ref. Alternatively, MPO-generated HOCl oxidizes free a-amino acids to aldehydes^ref, leading to advanced glycation products present in human lesion material^ref. However, the most common reaction of HOCl is with free amino groups of (apolipo)proteins, leading to formation of chloramines. MPO-generated HOCl carries out a wider variety of oxidative reactions, including chlorination of tyrosines and the oxidative modification (and often inactivation) of enzymes. Interestingly, myeloperoxidase shows preferential oxidation of adjacent tryptophan and glycine residues in MMP7, indicating site-specific modification and some degree of target specificity^ref. Hydroxyl radicals are highly reactive with various biomolecules, initiating free-radical chain reactions that produce marked oxidative damage. Similarly, both singlet oxygen and ozone have high reactivity, for example, with double bonds of unsaturated fatty acids.
Many of the ROS described mainly occur in inflamed areas where there is infiltration of neutrophils. Without MPO or antibody, many of the more highly reactive species are not formed. So, rather than producing widespread molecular damage via cellular oxidative stress (OS), as probably occurs in an inflamed site, ROS produced in non-immune tissues will oxidize a more limited spectrum of target molecules, and these more specific oxidations can be used for specific biological functions. They are generated in high levels by professional phagocytes using a superoxide-generating multicomponent NADPH phagocyte oxidase (Phox), whose catalytic subunit is gp91^phox. ERK is one of the ROS-responsive serine/threonine kinases : tyrosine kinases and small G proteins (Ga₁ and Ga₀^ref) are involved in the activation of ERK by ROS^{ref1, ref2}. Hydrogen peroxide generated extracellularly by receptor-ligand interaction facilitates cell signaling^ref

(reproduced with permission from Nature Reviews Immunology (Vol 4, No. 3, pp 181-189 (2004)) copyright Macmillan Magazines Ltd)

(reproduced with permission from Nature Reviews Immunology (Vol 4, No. 3, pp 181-189 (2004)) copyright Macmillan Magazines Ltd)



Laboratory examinations : a variety of methods are used to assess oxidation status in vivo. However, classical methods such as thiobarbituric acid reactive substances (TBARs) and the conjugated diene assay suffer from lack of specificity and limited application to clinical specimens^ref. Measurement of peroxidation products of free polyunsaturated fatty acids (PUFAs) such as malondialdehyde, 4-hydroxy-2-nonenal (HNE), or lipid hydroperoxides is limited by the relative instability of the analytes and their ready formation ex vivo^ref. Measurement of isoprostanes is widely used as a viable alternative for monitoring oxidant stress in vivo^{ref1, ref2, ref3, ref4}. Isoprostanes, products of free radical-induced oxidation of arachidonic acid (AA), are isomers of enzymatically formed prostaglandins. They are relatively stable chemically and can be measured in biological tissue and fluids with good sensitivity and specificity^{ref1, ref2, ref3, ref4, ref5}. However, their half-life in blood is limited by their rapid metabolism and excretion^{ref1, ref2, ref3, ref4, ref5, ref6, ref7, ref8}. Thus, isoprostanes provide a snapshot of oxidant stress in vivo over a relatively brief period of time but their utility in providing an integrated assessment of oxidant stress over a longer interval may be limited. Besides isoprostanes, free radical peroxidation of AA produces another class of compounds—isolevuglandins (isoLGs)—through common isoprostanoid endoperoxide precursors^{ref1, ref2}. Isoprostanes include structural and stereoisomers of prostaglandins that may be generated by COX-promoted oxygenation of AA, as well as via free radical-mediated mechanisms. Similarly, isolevuglandins comprise structural isomers (e.g., iso[4]LGE₂) as well as stereoisomers (e.g., isoLGE₂) of COX-generated levuglandins (e.g., LGE₂) and may also be formed via free radical-mediated pathways^{ref1, ref2}. IsoLGs differ from isoprostanes by containing a characteristic aldehydic group in a 1,4-dicarbonyl array, making them extremely reactive toward primary amino groups in proteins.

IsoLGs initially form Schiff base adducts, then pyrrole adducts, with the e-amino group of lysyl residues^{ref1, ref2}. However, the pyrrole adducts are unstable in the presence of oxygen and are further transformed to lactam and hydroxylactam adducts, which accumulate as stable end products^ref. Formation of isoLG protein adducts from free AA in vitro has been confirmed immunologically and by a variety of mass spectrometry methods^{ref1, ref2}. Elevated levels of isoLG protein adducts in plasma from patients with atherosclerosis compared with healthy age-matched subjects have been shown using polyclonal antibodies raised against synthetic isoLG-pyrrole-derived adducts^ref. Collectively, these data suggest that isoLGs and their protein adducts may serve as markers of oxidant stress. However, mechanisms for generation of isoLGs in vivo have not been established. Recent studies reveal that myeloperoxidase (MPO) serves as an enzymatic catalyst for initiation of lipid peroxidation and lipoprotein oxidation in vivo^ref. MPO is an abundant heme protein secreted by phagocytes in response to stimulation. MPO^ref and its distinct products [HOCl-damaged proteins^ref and 3-chlorotyrosine^ref] are enriched in human atherosclerotic aortic intima and LDL recovered from atheroma. MPO uses H₂O₂ together with low molecular weight cosubstrates like chloride^ref, tyrosine^ref, and nitrite (NO₂^-)^{ref1, ref2} to generate a variety of reactive oxidants and diffusible radical species^{ref1, ref2}. Recent studies using MPO knockout mice reveal that NO₂^-, the autoxidation product of nitric oxide, serves as a preferred substrate for MPO to generate nitrogen dioxide, a species capable of aromatic nitration and initiation of lipid peroxidation in vivo^{ref1, ref2}. MPO knockout mice are more susceptible to Candida infection, making the Candida sepsis model a useful tool for studying the role of MPO in inflammation^{ref1, ref2, ref3}. There are significant increases in isoLG protein adducts in plasma proteins of mice after Candida sepsis using enzyme-linked immunosorbent assays (ELISA) with antibodies specific for isoLG protein adducts. Plasma levels of F₂ isoprostanes failed to significantly increase in wild-type or MPO knockout mice in this model. Comparison of plasma levels of iso[4]LGE₂ protein adducts in wild-type vs. MPO knockout mice revealed a significant reduction in levels within plasma recovered from MPO knockout mice. The MPO-H₂O₂-NO₂- system forms isoLGs and isoLG protein adducts from target phospholipid vesicles and lipoproteins, respectively. The present findings thus confirm that MPO serves as one pathway for generation of isoLGs in vivo. They also suggest that monitoring longer lived protein/lipid adducts in plasma may be a more sensitive means than F₂ isoprostanes of detecting enhanced oxidant stress in vivo^ref.

• reactive nitrogen intermediates (RNI) released by macrophages expressing inducible nitric oxide synthase (iNOS) / NOS2 :
• nitrite
• S-nitrosoglutathione (GSNO)
..., which are bactericidal in vitro at a pH characteristic of the phagosome of activated macrophages
[peptide methionine sulfoxide reductase (MsrA) from Escherichia coli and Mycobacterium tuberculosis catalyzes the reduction of methionine sulfoxide (Met-O) in proteins to methionine (Met)]
• biological factors
• hemostasis
• bacterial interference or symbiosis [decreased by bacteriocins and during broad-spectrum antimicrobial chemotherapies] on both inner and outer surface areas of the body.
• [bactericidin : a substance that leads to the death of bacteria; such substances include both antibody and certain nonantibody components in the serum]
• enzymes
• lactoperoxidase (in salivaand colostrum)
• xanthine dehydrogenase / oxidase (in neutrophils and colostrum)
• lysozyme (in tears, saliva, sweat, colostrum, nasal mucus, blood plasma, both primary / azurophilic / aspecific and secondary / specific granules of neutrophils, secretory granules of Paneth cells in intestinal Lieberkuhn cryptae, serous cells in submucosal glands of airways) breaks the (b1->4) glycosidic bond between MurNAc and GlcNAc in the peptidoglycan of bacterial cell walls. It is also present in egg albumen [hen egg lysozyme (HEL) is used commercially and also as food preservative (E1105)] and is a late gene of lytic bacteriophages. Alone, lysozyme has been found to be bactericidal against some Gram-positive Bacteria, but in concert with lactoferrin (present in exocrine secretions that are commonly exposed to normal flora: milk, tears, nasal exudate, saliva, bronchial mucus, gastrointestinal fluids, cervicovaginal mucus and seminal fluid), it is also bactericidal against some Gram-negative Bacteria.
• group IIA, IID and X secretory phospholipase A₂ (in secretory granules of Paneth cells and secondary / specific granules of neutrophils)
• acidic mammalian chitinase (AMCase) / eosinophil chemotactic cytokine / ECF-L in response to infection with parasites that have chitin coats (induced in lung epithelial cells and macrophages by IL-13 produced by T_h2)^ref
• phagocyte-derived chitotriosidase (CHIT1), the plasma levels of which are markedly elevated in some pathological conditions. Both polymorphonuclear neutrophils (PMNs) and macrophages are a source of chitotriosidase. The enzyme is located in specific granules of human PMNs and secreted following stimulation with GM-CSF. In addition, GM-CSF induces expression of chitotriosidase in macrophages that constitutively secrete the enzyme and partly accumulate it in their lysosomes. Studies with recombinant human chitotriosidase revealed that the enzyme targets chitin-containing fungi. These findings are consistent with earlier observations concerning anti-fungal activity of homologous plant chitinases and beneficial effects of GM-CSF administration in individuals suffering from invasive fungal infections^ref
• structural proteins
• siderophilins
• "natural" antibodies
• IgGs acquired from mother (average t_1/2 = 20 dd.)
• in utero : placental transcytosis of IgG₁, IgG₃ and IgG₄ through syncitiumtrophoblast and yolk sac
• in newborn : IgGs undergo transcytosis in mammary gland => colostrum => transcytosis across the epithelium of the neonatal intestine
The neonatal FcR (FcRn) plays a central role in both transfers.
• heterophilic IgMs produced by B-1 B cells or by marginal zone (MZ) B cells following an adaptive immune response against T-dependent (TD) or T-independent (TI) Ags, respectively, coming from symbiontic bacterial flora. IgM are directed against :
• some blood group antigens : AB0, P1
• Gal-a-1,3-Gal epitopes : the major xenoantigen that causes hyperacute rejection in pig-to-human xenotransplantation.
• other molecules
• polyamines (in prostatic and pancreatic juices, leukocytes and colostrum) inhibit Gram +ve Bacteria and Mycobacteriaceae.
• spermine
• spermidine

• basophils
• tissue mast cells
• platelets
• epithelial cells
• cell types having a role also in adaptive immune system
• NK cells
• innate lymphocytes
• phagocytic cells / mononuclear phagocyte and immunoregulatory system (M-PIRE) : in Eukarya they can be thought of as arising from the need of a Protista to discriminate between food and other Protista. Such phagotrophic Protista may welll have occupied the coelomic cavity of early multicellular organisms extablishing a mutualistic symbiosis : probably the ontogeny of modern macrophages recapitulates their phylogeny. Phagocytic cells expressing pattern recognition receptors link innate immune system to the variable Ag-recognition receptors of the adaptive immune system : TcR. More exactly they act as Ag-presenting cells (APCs) (MHC I⁺, MHC II⁺). Anyway the study of Drosophila melanogaster immunology clearly demonstrated the relative efficacy of a nonclonal system of host defense : one of the most obvious advantages is the absence of autoimmune diseases, which are therefore clearly shown to depend on adaptive immune system.

Phagocytosis is an active process first discovered by Elie Metchnikoff in 1883 (Nobel Prize for Medicine in 1908 together with Paul Ehrlich), who observed that Asteroidea (starfishes) larvae have amoeboid cells that congregate at sites of infection, engulf carmine (dye) particles.
• Metchnikoff's (Mechnikov's) cellular immunity theory : the theory, proposed in the 1880s, that phagocytosis by macrophages and PMN leukocytes is the main mechanism of host defense against bacterial infection and that inflammation is the result of the enzymatic digestion process occurring with phagocytosis.
pH lowering allows contemporaneous activation of lysosomal acidic hydrolases. On the other hand occurs an oxidative, metabolic or respiratory burst : a sequence of 4 metabolic events that occur during oxidative killing of ingested microorganisms by granulocytes and mononuclear phagocytes, consisting of ...
• (1) an increase in oxygen consumption
• (2) formation of superoxide anion
• (3) formation of hydrogen peroxide
• (4) activation of the hexose monophosphate shunt.
Molecular oxygen is converted to superoxide by NADPH oxidase and NADH oxidase. Superoxide is converted to hydrogen peroxide by superoxide dismutase. Hydrogen peroxide is utilized in myeloperoxidase-dependent bacterial killing, and both superoxide and hydrogen peroxide are spontaneously converted to other toxic metabolites, e.g., singlet oxygen and hydroxyl radicals. The hexose monophosphate shunt regenerates NADPH. The burst produces radicals that damage phagocytated biopolymers by creating reactive oxygen intermediates (ROI) and reactive nitrogen intermediates (RNI). DCs migrating from peripheral tissues can transfer captured Ags to other DCs for their presentation in secondary lymphoid tissue, either by phagocytosis or by release of exosomes. 2 rates of phagosome maturation exist : a constitutive rate and a more rapid inducible rate, triggered by TLR signalling by the specific phagosome cargo. The inducible rate of maturation is phagosome autonomous and the p38^MAPK that is activated downstream of TLRs is required for accelerated phagosome maturation triggered by bacteria. The TLR–MyD88–p38 signalling pathway might be disrupted by some intracellular bacteria that avoid phagolysosomal fusion and the subsequent recognition by the innate and adaptive immune systems^ref
• constitutive or professional APCs (pAPC)
• bone-marrow-derived or hematopoietic APCs
• precursor of dendritic cell 1 (pDC1) / bone marrow-derived macrophages (some but not all macrophages according to Metchnikoff's classification) : the most potent type of APC, responsible for initiating immune responses.
• B-cells : this is not phagocytosis but rather RME. Resting B cells are poor at presenting Ag or are tolerogenic to T cells^Ref : this may relate to the phenomena of low- and high-zone tolerance, neonatal tolerance, and the beneficial effect of blood transfusions on allograft survival. They turn into effective APCs with the increased expression of costimulatory molecules after the activation^Ref. When they use BcR, they are mainly active in secondary immune responses and at low Ag charges because of maturated BcR affinity. However, Ags specifically bound to surface Ig (sIg) on B cells are presented to T cells 10³-to 10⁴-fold more efficiently than those entering the cells in an sIg-independent manner. B cells are efficient APCs for both naive T cells and Ag-primed T cells and are likely to induce T_h2-dominant responses. Anyway when they use TLR7, TLR9 or TLR10 they induce T_h1-dominant responses.
• neutrophils (a.k.a. microphages in Metchnikoff's classification)
• lymphoid or stromal or plasmacytoid dendritic cells
• precursor of dendritic cell 2 (pDC2) / plasmacytoid DC (PDC / P-DC) / IFN-producing cell (IPC) (equivalent to murine plasmacytoid DC precursor) (CD2⁺3^-4^hi7⁺8a^-10^-11b^-11c^-/lo13^-14^-15^-16^-w17^-19^-20^-21^-23^-25^-26^-27^-28^-31⁺33^-/+34^-

36⁺38⁺43⁺40⁺44⁺45RA^hi45RO^-56^{+ (1%) / - (99%)}57^-62L / L-selectin^hiw65^-68⁺69^-89^-94^-116 / GM-CSFR⁺122^-123 / IL-3Ra^{hi / bright}152^-161^-244^- preTcRa⁺, CXCR3⁺, CXCR4⁺, CCR5⁺, LIR-5 / ILT-3⁺, BDCA-2⁺, BDCA-4^+ref TLR1⁺2^-3^-4^-5^-6⁺7⁺⁺8^-9⁺⁺10^+/-, cMHC II⁺ but sMHC II^-, no phagocytic activity). They produce mRNA for germ-line IgK. Development requires IRF8. The different steps of lymphoid and myeloid DC differentiation are not fully characterized and the ontogenic origin of pDCs is still under debate
• arguments in favor of a lymphoid origin :
• pDCs express lymphoid-specific surface molecules and transcripts^ref, including : pTa^{ref2, ref3}, Spi-B^ref, and TCL1^ref5
• common blockage of T- and B-cell and pDC, but not myeloid, differentiation with inhibitors of DNA binding (Id-2) and Id-3^ref
• arguments in favor of a myeloid origin :
• expression of the myeloid marker CD33^ref by tumoral and normal pDCs^ref
• few cases of CD4⁺CD56⁺ lineage-negative malignancies demonstrated a transformation in chronic myelomonocytic leukemia and acute myeloid leukemia^{ref1, ref2} whereas the evolution of CD4⁺CD56⁺ malignancies in ALL was never reported to our knowledge.
In their interesting study, Herling et al attested that the myelomonocytic blasts and the CD4⁺CD56⁺ blasts arose from a common leukemic clone because they both expressed the TCL1 proto-oncogene, a marker never encountered in de novo myeloid leukemias^ref. A common precursor for pDCs and myeloid DCs is also strongly suggested by the study by Mothy and colleagues who reported that both in vitro expanded myeloid DCs and pDCs exhibit the same chromosomal abnormality as that detected in the original myeloid leukemic clone from which they arise.44 Moreover, myelodysplastic features in bone marrow cells of CD4⁺CD56⁺ leukemia patients^ref or a history of myelodysplastic syndrome before the diagnosis of CD4⁺CD56⁺ leukemia^{ref1, ref2, ref3} suggest a common origin of pDCs and the myeloid lineage. The fact that normal CD123^high pDCs could be generated from a CD34⁺ myeloid progenitor expressing M-CSF receptor^ref is also an argument in favor of a myeloid lineage. CD36 and CD68, 2 well-known myeloid markers, are expressed by pDCs^{ref1, ref2, ref3, ref4}. In the literature, among myeloid markers, CD33 is usually considered as nonexpressed on normal as well as tumoral pDCs. However, some recent reports described the expression of CD33 on normal and tumoral pDCs. Thus, in the GEIL study^{ref1, ref2}, CD33 was previously observed at relapse on the pDC leukemia of 2 patients even though it was absent at diagnosis. In addition, 5 other studies reported a low CD33 expression in a least one case in their series of CD4⁺CD56⁺ lineage-negative malignancies^{ref1, ref2, ref3, ref4, ref5}. Up-regulation of CD33 expression on leukemic pDCs in culture after maturation into potent APCs^ref also supports the idea that CD33 is expressed by pDCs during their differentiation process. One has also to consider the plasticity of DC differentiation. Canque and Gluckman^ref and Gluckman et al (Gluckman JC, Canque B, Rosenzwajg M. Dendritic cells: a complex simplicity. Transplantation. 2002;73(1 suppl): S3-S6) have proposed that DCs may differentiate from many precursors (either immature or already committed to lymphoid or myeloid lineages). A recent publication suggests that pDCs may represent a population of lymphoid cells undergoing an in vivo cell fate conversion from a lymphoid to a myeloid cell type^ref. Finally, the recent model of lineage commitment in hematopoiesis proposed by Katsura^ref with a common myeloid-lymphoid progenitor that gives rise to B and T cells as well as the myeloid lineage allows us to postulate that pDCs may derive from a multipotent common myeloid-lymphoid progenitor. This may explain why CD4⁺CD56⁺ leukemic cells express some features of lymphoid and myeloid as well as dendritic lineage. In addition, Miller et al identified a human bone marrow multipotent progenitor that is able to generate mature myeloid, natural killer (NK), B, and dendritic cells^ref. Overall, this may reconcile our findings on these CD4⁺CD56⁺ malignancies and may account for all the contradictory results published in the literature about the ontogenic origin of pDCs. Besides, one has to consider the function delivered by CD33 present on leukemic lymphoid DCs. The biologic role of CD33 still remains unclear. As a Siglec family member, CD33 has lectin activity for a2-6 and a2-3 sialylated oligosaccharides expressed on different cells^ref. However, nothing is known about its in vivo role in cell-to-cell or cell-to-matrix interactions^{ref1, ref2}. To date, mainly inhibitory signals following CD33 triggering have been reported using specific anti-CD33 mAb^{ref1, ref2, ref3, ref4, ref5, ref6, ref7}. This inhibitory signal is related to immunoreceptor tyrosine-based inhibitory motifs (ITIMs) present on CD33 cytoplasmic tail^{ref1, ref2, ref3}. Interestingly, CD33 stimulation inhibits myeloid DC differentiation/maturation^ref and induces apoptosis^ref. Another series did not observe such effects on leukemic lymphoid DCs^ref. This can be related to our experimental conditions because it was necessary to culture tumoral pDCs with survival factors (eg, IL-3 or GM-CSF) to prevent spontaneous cell death^ref. Such cytokines can deliver survival signals including tyrosine phosphorylation that may interfere with CD33 signaling, as described^ref. Furthermore, CD33 function was assessed in tumoral pDCs. As reported for a myeloid leukemia line^ref, activation of multiple signaling pathways related to oncogenic transformation may modify the consequence of CD33 ligation and in our hands may have led to proinflammatory cytokine production. Whether this also occurs in normal circulating pDCs is currently under investigation. However, we provide evidence that CD33 is also expressed by pDCs and may modulate the function of their tumoral counterpart. The use of Mylotarg (an anti-CD33 calicheamicin immunoconjugate)^ref can be a new therapeutic option to be tested.
See also activation of pDC2 to DC2.
• CD3^-4⁺11c^-40^lo80^lo86^lo153 / TNFSF8 / 30L^hiCD134L / OX40L / gp34^hiMHC class II^lo cells are located mainly in B-cell follicles and at the interface between T cells and B cells (on the contrary of DCs which are located mainly in T-cell zones). They potentiate the survival of T_h2 lymphocytes but not T_h1 lymphocytes in vitro and in vivo in the later phase of antibody response.
• follicular DC (FDC) (CD13^-14⁺19⁺21⁺35 / CR1⁺) in B areas of primary lymphoid follicles or in germinal centres of secondary lymphoid follicles of secondary lymphoid organs. They carry Ag on their surface as immune complex bound to CR3 and act as APCs for B lymphocytes. They produce IFN-a, CXCL13 / BLC, and FDC secreted peptide (FDC-SP). FcgRIIb1 expression is upregulated during the germinal center (GC) response.
• facultative APCs : temporarily induced by IFN-g to express MHC class II molecule. They are effective only if they have phagocitary activity and express costimulatory molecules.
• eosinophils
• autophagocytosis : autophagosome-like compartments have been observed containing bacteria such as Rickettsia conorii in endothelial cells^ref and Listeria monocytogenes in macrophages^ref, hinting autophagy could act as an innate host defensive pathway. Non-immune cells can eliminate invading bacteria effectively by enveloping them with autophagic machinery (autophagosome-specific membrane marker microtubule-associated protein light chain 3 (MAP-LC3) and autophagy inhibitors 3-methyladenine and wortmannin). The number of cells with group A Streptococcus (GAS)-containing LC3-positive autophagosome-like vacuoles, the area of these vacuoles, and the ratio of GAS trapped in them to total intracellular GAS increase over time after infection, with the vacuoles eventually trapping about 80% of intracellular GAS. LC3-II, a form of LC3 that correlates directly with the number of autophagosomes, increases after infection : intracellular GAS die 4 hours after infection, while in Atg5-deficient cells, GAS remained viable. Autophagosome-like vacuoles 4 hours after infection contain degraded cytosol and partially degraded GAS, features characteristic of autophagosomes fused with lysosomes. LC3 and LAMP-1, a lysosomal membrane protein, also colocalized 2 to 3 hours after infection, suggesting fusion with lysosomes after formation of the vacuoles, similar to what occurs in the standard autophagic pathway. When lysosomal protease inhibitors are used to mimic inhibition of lysosome function, wildtype cells have significantly more viable intracellular GAS than before, suggesting lysosomes need to fuse with autophagosomes to decrease GAS viability. This newly discovered autophagic response is specifically induced by GAS invasion. Moreover, the size of the GAS-specific autophagosomes is over 10 times larger than that of standard autophagosomes. The distinctions between classical autophagy and GAS-induced autophagy suggest that the latter is a specific rescue program for counterattack against intracellular pathogens^ref
Strategies of innate immune recognition :
• recognition of missing markers of normal self : specialized markers of normal self are dedicated gene products of metabolic pathways that are unique to the host and absent from microorganisms. Thery are expressed constitutively on normal healthy cells of the host and, by engaging inhibitory receptors, prevent phagocytosis of these cells by macrophages, killing by NK cells, and perhapes phagocytosis and presentation of antigens derived from normal healthy cells by dendritic cells. Conversely the absence of these markers markers on microbial cells and their down-regulation on infected, transformed, and senescent host cells render these targets susceptible to NK-mediated cytotoxicity or phagocytosis by macrophages.
• down-regulation of MHC class I expression (ubiquitously expressed by all self cells) on surface of viral infected or transformed self cells : NK cells inhibitory receptors are no longer activated
• lack of several gene products (CD46 / MCP, CD55 / DAF, ...) (ubiquitously expressed by all self cells) on surfaces of microbial cells : formation of an active C3 convertase in the alternative complement pathway is no longer inhibited
• lack of sialic acid (ubiquitously expressed by all self cells) on surfaces of microbial nonself (which lacks biosynthetic pathways for sialic acid) and abnormal self (which lost sialic acid expression because of infection, transformation, or senescence).
• sialic acid binding Ig-like lectins (SIGLEC) on macrophages, dendritic cells, and neutrophils no longer inhibit phagocytosis of microbial nonself (which lacks biosynthetic pathways for sialic acid) and abnormal self (which lost sialic acid expression because of viral infection, transformation, or senescence).
• SIGLEC-1 / sialoadhesin
• SIGLEC-2 / CD22 on B cells is not activated and BcR signaling is no longer blocked
• SIGLEC-3 / CD33
• SIGLEC-5
• SIGLEC-6
• SIGLEC-7
• SIGLEC-8
• SIGLEC-9
• SIGLEC-10
• SIGLEC-11
• SIGLEC-like 1 (SIGLECL1)
• H factor can no longer inhibit the formation of the C3 convertase in the alternative complement pathway
• down-regulation of CD47 expression by senescent erythrocytes : signal inhibitory regulatory protein a (SIRPa) / PTPNS1 can no longer inhibit phagocytosis, leading to erythrocatheresis
Although the missing self strategy is an efficient way to distinguish normal self from nonself or abnormal self, it is not resistant to fraud. Indeed, in multiple examples of "stolen identity", pathogens acquire by horizontal transfer the genes encoding self-markers and thus protect themselves from detection and destruction by the host.
• recognition of markers of abnormal self that are induced upon infection (in particular, viral infection) and cellular transformation. Markers of abnormal self tag the affected cells for elimination by the immune system.
• up-regulation of ligands for NK cells and CD8⁺ CTLsactivatory receptors on surface of viral infected or transformed self cells. This strategy is more effective than cell-autonomous apoptosis because :
• recognition of these ligands by NK cells and CD8⁺ CTLs does not lead just to apoptosis of the target cell, but also production of cytokines such as IFN-g
• the involvement of the extrinsic pathway may help to counteract viral strategies of blocking the intrinsic pathway and to bypass possible blocks in the cell autonomous pathway caused by mutations in tumor cells
• markers of apoptosis induce phagocytosis by macrophages
• the tumour-associated disialoganglioside GD₃ is recognized by Va14-Ja18 TcR NKT lymphocytes
• recognition of microbial nonself (discrimination between infectious nonself and noninfectious self) : pattern recognition receptors (PRR) recognize different sets of homologous molecules (homotopes / pathogen-associated molecular patterns (PAMP) / microbial-associated molecular patterns (MAMP)) with housekeeping functions in pathogens : PAMPs are essential for pathogen survival and, on the contrary of virulence factors, are highly conserved among microorganisms of a given class, so that escape mutants are not generated. Sometimes the conserved pattern is just a part of a molecule : e.g. in LPS only the lipid A is conserved and recognized by its PRR, while the O antigen is variable and is not recognized by the innate immune system. As PAMPs are invariant the number of germ-line encoded PRRs required to detect them is limited. Anyway PRRs cannot distinguish between pathogenic and commensal microorganisms expressing the same PAMPs : however, only pathogens evolved the means to gain access to the compartments within the host where the host's PRRs can detect them and can induce immune responses (e.g. TLR5 is expressed on the basolateral membrane of enterocyte). One of the puzzling features of PRRs is that each one can bind to many different kinds of molecules : this can be explained according to some theories
• PRRs have not evolved to bind to pathogens at all. Perhaps PRRs originally evolved as receptors for injury-related signals, and the microbes subsequently evolved convergent mechanisms to use these receptors to enhance their own survival
• the same alarm signals may be used by many different organisms. Because life evolved in water, any hydrophobic portion (Hyppo) of a given molecule is usually buried in the depths of that molecule, or hidden in the lipid membrane of the cell, and could act as an alarm signal if exposed. For example, the hydrophobic part of LPS is crucial for its immunostimulating properties, yet LPS is normally an integral bacterial membrane molecule and its Hyppos are hidden in the membrane. However, when released by damaged or dead bacteria, the newly exposed Hyppos could act as a bacterial alarma signal (or perhaps a type of quorum sensor, perhaps signaling the surviving bacteria to sporulate or otherwise change their behavior. Plant and animal cells also have an abundant supply of hidden Hyppos in their membrane and cytoplasm. During protein synthesis, Hsps and other chaperones bind to the Hyppos of nascent proteins to prevent their aggregation. Should a cell be disrupted, the Hyppos of both the nascent proteins and their chaperones would be exposed^ref.
Kinds of PRRs :
• secreted PRRs / ribosomally synthesized antimicrobial peptides (AMP) (RAMP) (< 100 amino acids) : they are cationic and hydrophobic at physiologic pH, attributes that assist peptide binding and insertion into microbial membranes. Some peptides then aggregate to form pores, and death of the microbe occurs once a threshold number of pores have been formed.
• receptor decoys (e.g. mucines)
• mindin / spondin 2 (SPON2) : a highly conserved secreted ECM protein highly expressed in lymphoid tissues, that can bind directly to both Gram-negative and Gram-positve bacteria, and can also act as an opsonin for macrophage phagocytosis of some pathogens^ref
• lipocalins transport low-molecular-weight chemical signals. They exhibit a structurally simple one-domain fold with a conformationally well conserved beta-barrel as their central motif. This type of supersecondary structure is made of a cylindrically closed b-sheet of 8 antiparallel strands. At the open end of the barrel the b-strands are connected by 4 loops in a pairwise manner so that a pocket for the ligand is formed.  In a rational protein design study a metal-binding site was functionally grafted on the solvent-exposed surface of the b-barrel, whereby the rigid backbone conformation permitted the spatially defined arrangement of 3 His side chains. In a combinatorial protein design approach, the natural ligand pocket of a lipocalin was reshaped. In this manner variants of the BBP were engineered which exhibit high affinity and remarkable specificity for haptens like fluorescein and digoxigenin. The so-called 'anticalins', i.e. artificial lipocalins recognizing prescribed ligands, could provide an interesting alternative to recombinant antibody fragments. Consequently, the use of lipocalins as a scaffold opens new applications for members of this functionally diverse protein family in biotechnology and medicine.
• lipocalin 1 (LCN1) / tear prealbumin (TP) / Von Ebners gland protein precursor (VEG protein) binds to lipocalin-1 interacting membrane receptor (LIMR)^ref
• lipocalin 1-like 1 (LCN1L1)
• lipocalin 2 (LCN2) / siderocalin / neutrophil gelatinase-associated lipocalin (NGAL) expression is upregulated upon TLR ligation^ref : it binds iron-laden enterochelin^ref and induces the formation of kidney epithelia^ref
• lipocalin 5 or 6 (LCN5 or LCN6) / epididymal-specific lipocalin : predominant expression in the epididymis and location on sperm surface are consistent with a role for LCN6 in male fertility^ref
• lipocalin 7 (LCN7)
• lipocalin 8 (LCN8)
• lipocalin 9 (LCN9)
• lipocalin 10 (LCN10)
• lipocalin 12 (LCN12)
• glycodelin
• C8g
• retinol-binding protein (RBP)
• bilin-binding protein (BBP)
The lipocalins were once regarded as a eukaryotic protein family, but new members have been recently discovered in bacteria. The first bacterial lipocalin (Blc) was identified in Escherichia coli as an outer membrane lipoprotein expressed under conditions of environmental stress. Blc is distinguished from most lipocalins by the absence of intramolecular disulfide bonds, but the presence of a membrane anchor is shared with 2 of its closest homologues, apolipoprotein D and lazarillo (a neuronal lipocalin in grasshoppers^ref). Several common features of the membrane-anchored lipocalins suggest that each may play an important role in membrane biogenesis and repair. Additionally, Blc proteins are implicated in the dissemination of antibiotic resistance genes and in the activation of immunity. Recent genome sequencing efforts reveal the existence of at least 20 bacterial lipocalins. The lipocalins appear to have originated in Gram-negative bacteria and were probably transferred horizontally to eukaryotes from the endosymbiotic alpha-proteobacterial ancestor of the mitochondrion. The genome sequences also reveal that some bacterial lipocalins exhibit disulfide bonds and alternative modes of subcellular localization, which include targeting to the periplasmic space, the cytoplasmic membrane, and the cytosol. The relationships between bacterial lipocalin structure and function further illuminate the common biochemistry of bacterial and eukaryotic cells^ref. Lipocalins separate into 14 monophyletic clades, some of which are grouped in well supported superclades. The lipocalin tree was rooted with the bacterial lipocalin genes under the assumption that they have evolved from a single common ancestor with the metazoan lipocalins, and not by horizontal transfer. The topology of the rooted tree and the species distribution of lipocalins suggest that the newly arising lipocalins show a higher rate of amino acid sequence divergence, a higher rate of gene duplication, and their internal pocket has evolved towards binding smaller hydrophobic ligands with more efficiency^ref.
• C-type lectins
• collectins : upon binding to their ligand, collectins associate with the common collectin receptor complex made up of CD91 / LRP / a₂-macroglobulin R and C1qR / collectins receptor / calreticulin, on plasma membrane of macrophages, thereby enhancing the phagocytosis of pathogens and apoptotic cells.
• conglutinin / collectin subfamily member 8 (in plasma) : it binds specifically the carbohydrate moiety of iC3b.
• collectin 43 (CL-43) / collectin subfamily member 9 (in plasma, produced by hepatocytes). Specificity : Man >>> ManNAc >>> L-Fuc > GlcNAc > Glc > Mal >>> Gal.
• collectin 46 (CL-46). Specificity : GlcNAc >>> ManNAc > aMeMan > Mal, GlcN > Man, Glc, L-Fuc >> Gal > aMeGlu > Fuc
• surfactant, pulmonary-associated proteins (SP) in lamellar bodies of type II pneumocytes and Clara cells (and then in lung surfactant (LS)) and in enterocytes. They consist of 2 regions : a globular head that recognizes pathogens and a collagenous tail. They have a dual function : a non-inflammatory environment is maintained through interaction with signal inhibitory regulatory protein a (SIRPa) / PTPNS1 (an ITIM-containing receptor) and inhibition of pro-inflammatory cytokine secretion by alveolar macrophages, but after recognition of microbial pathogens, the orientation of SP-A and SP-D is reversed such that the collagenous tail interacts with the calreticulin-CD91 complex to initiate receptor-mediated phagocytosis and induce a pro-inflammatory response^ref.
• surfactant, pulmonary-associated  protein A
• surfactant, pulmonary-associated  protein A1 (SP-A1)
• surfactant, pulmonary-associated  protein A2 (SP-A2)
SP-A is the most abundant surfactant-associated protein. Human monomeric SP-A subunits are 35-kDa polypeptides that contain one N-linked oligosaccharide attachment site, a hydroxyproline-rich collagen-like domain, and carbohydrate recognition domains (CRDs). Subunits of SP-A assemble into trimers and then further associate into 18-mers composed of 6 trimers through interchain disulfide bond formation at the N terminus and noncovalent interactions between the collagen-like sequences. The resulting "bouquet of tulips" configuration, containing CRDs arranged on stalk-like scaffolding of collagen tails, provides a high valency of binding sites for microbe and host cell molecules. It requires a functional TLR4 complex to induce leukocyte activation. Specificity : ManNAc >>> L-Fuc, Mal > Glc > Man. SP-A binds to dipalmitoyl phosphatidylcholine (DPPC) and galactosylceramide (GalCer) and interacts with alveolar type II cells. It up-regulates surface expression of MMR from intracellular pools. Human SP-A isolated from patients with pulmonary alveolar proteinosis (PAP) forms even larger aggregates by self-association of multiple bouquets. It binds DNA with lower affinity than SP-D at physiological salt conditions^ref
• surfactant protein D (SP-D) reduces alveolar macrophage apoptosis and binds to ...
• Mal > d-Man > l-Man >>> Gal > GlcNAc >>> L-Fuc
• Klebsiella pneumoniae smooth forms of LPS expressed by O-serotypes with mannose-rich repeating units in their O-polysaccharides.
• several, but not all, strains of Escherichia coli
• Enterobacter aerogenes
• linear plasmid DNA from bacteria or mice directly, as well as synthetic oligonucleotides, DNA and RNA polymer-related compounds with higher affinity than SP-A and mannose-binding lectin at physiological salt conditions, and even at high salt concentrations.Whereas this strong interaction was mediated largely by the collagen-like region of SP-D, electron microscopy indicated that the globular regions of the molecule also associated with DNA. Furthermore, because SP-D colocalizes with the DNA of apoptotic cells, SP-D functions as an opsonin, binding the DNA of apoptotic cells to enhance their clearance^ref.
It inhibits bacterial synthetic functions by increasing membrane permeability through a C-terminal domain-dependent mechanism.
It is present in surfactant at about 1/10th the concentration of SP-A.
• C1q is described in the classical pathway for complement cascade activation
• mannose-binding protein / lectin (MBP / MBL) is described in the lectin pathway for complement cascade activation
• ficolins are described in the lectin pathway for complement cascade activation
• galectinsare b-galactoside binding lectins that are found in the cytoplasm of various cells, including phagocytic leukocytes, endothelial cells and thymic epithelial cells^{ref1, ref2, ref3, ref4, ref5, ref6}. Although galectins are not sorted into the classical secretory pathway, it has been well established that galectins are released from cells. As discussed below, galectin release is often regulated in a differentiation or activation-dependent manner^ref. Moreover, the cytoplasmic location (thus free from the majority of galectin&rsquo;s oligosaccharide ligands) serves as a unique reservoir of galectin, which can release the lectins when cells are damaged by infections. As recent evidence indicates that galectins are potential immunomodulators in various immune responses, in this review I would like to try to shed light on and discuss the roles of galectins as Danger signal molecules, which could be evocative of an immune response to infection. Thus, some of galectins&rsquo; extracellular biological functions related to immunity will be reviewed but I will not be able to cover the potential intracellular roles of galectins in the immune system1 or extracellular roles of galectins in non-immune systems. Readers would be recommended to look at the recent reviews on galectins on those functions^{ref1, ref2, ref3, ref4, ref5, ref6}. Galectins belong to a ?-galactoside binding protein family defined by the conserved peptide sequence elements involved in the carbohydrate binding activity of those lectins^ref. Up to 14 galectins (galectin-1&ndash;14) have been found in mammals

so far as well as many in other phyla including birds, amphibians, fish, nematodes, Drosophila, sponges and fungi ^{ref1, ref2, ref3, ref4, ref5, ref6, ref7, ref8, ref9, ref10, ref11}. While all galectins share a core sequence (carbohydrate recognition domain CRD) consisting of &ndash;130 amino acids, galectins exhibit interesting structural differences in the presentation of the CRD (Fig. 1). While some galectins contain one CRD (prototype), and exist as monomers (galectin-5, 7, 10) or dimers (galectin-1, 2, 11, 13, 14), other galectins, such as galectin-4, 6, 8, 9, 12 contain 2 CRD connected by a short linker region (tandem repeat)^ref. In contrast, galectin-3 uniquely occurs as a chimeric protein with one CRD and an additional non-CRD domain, which is involved in the oligomerization of galectin-3^{ref1, ref2, ref3}. Regardless of the presentation of the CRD in the galectin molecule, galectins can cross-link their ligands (Fig. 2) because the majority of galectins are intrinsically divalent (with the exception of galectin-3, 5, 7, and 10). Galectin-3, which exists as a monomer, also mediates cross-linking of its ligands through its oligomerization. Upon binding to its glycoconjugate ligands at the cell surface, the conformation of galectin-3 appears to be altered in a concentration and time-dependent manner, and galectin-3 begins to oligomerize through the self-assembly of the N-terminal regulatory domain of the galectin-3 molecules^{ref1, ref2}. It has been suggested that C-terminal domain of galectin-3 is also implicated in the oligomerization^ref. The oligomerization results in the formation of a galectin-3 molecule which possesses multivalent CRDs. This dynamism between monomer and oligomer to regulate &lsquo;lectin activity&rsquo; (activity to cross-link ligands) is a very unique feature of galectin-3, as the majority of galectins do not alter their multivalency. Once they are released extracellularly, galectins could cross-link their surface glycoprotein ligands (Fig. 2). As discussed later, the cross-linking of galectin ligands induces various activities, including cell adhesion (Fig. 2A), signal transduction through receptor clustering (Fig. 2B) and the formation of multivalent galectin-glycoprotein lattices on cell surface, which could negatively regulate the movement of surface receptors (Fig. 2C)^{ref1, ref2, ref3, ref4}. Many of those reported functions of galectins are closely related to immunity, implying the roles played by galectins during infection. However, in order to crosslink surface ligands to exert their activities, galectins, the majority of which are found in the cytoplasm, have to be released extracellularly (20). C. Extracellualr Release of Galectins; Passive Release and Active Secretion Since it is known that galectins are found in the extracellular space and at the cell surface, galectin molecules have to be exported from those cells^ref. The majority of the secretory proteins contain either signal sequences or potential transmembrane domains, which are essential in order for them to be sorted to the secretory pathway for secretion. However, most members of the galectin family 2 do not contain any of those sequences. Thus, galectin molecules are synthesized on free ribosomes and first accumulate in the cytosol^ref (20,35,36). There could be at least two ways for cytosolic molecules to be released; release caused by cell necrosis (passive release) and active secretion from live cells through &lsquo;leaderless&rsquo; (alternative) secretory pathway (see below for details), as it is proposed in the case of some cytoplasmic cytokines or cytokine-like molecules proteins, such as interleukin (IL)-1 (37&ndash;39), fibroblast growth factors (FGF) (40&ndash;42), thioredoxin^ref (43) and possibly some of the S100 protein family (44). In chronic or severe acute infections, various cells could be damaged by exudated leukocytes or infectious agents, resulting in the release of cellular contents. Thus, in some cases of severe infections, cell necrosis could result in passive release of intracellular components such as galectins. Interestingly, release of some cytoplasmic molecules, such as thioredoxin, IL-1, heat shock proteins (HSPs) and calreticulin are recognized by immunity as &lsquo;danger signals&rsquo; and those molecules can trigger various immune responses (8,10,45). As galectins also play roles as immunomodulators (see details below), it is likely that some of the galectin molecules can be released through necrosis at the infected sites by this passive mechanism and those extracellular galectin molecules could also serve as &lsquo;danger signals&rsquo;. Historically speaking, it had been casually postulated by the majority of the immunology community, that cytosolic cytokines, such as IL-1 and fibroblast growth factors are released by this passive mechanism, which is triggered by cell necrosis, but not by active secretion. However, various lines of investigation over the secretion mechanism of IL-1 and FGF have now established the solid bases that cells are able to actively secrete some cytosolic cytokines through the &lsquo;leaderless&rsquo; secretory pathway, without compromising membrane integrity^ref. Active secretion of galectins (galectin-1, 3 and 9) from live cells has also been well demonstrated by various reports ^ref,46). It appears that galectin-3 utilizes, at least in part, the same &lsquo;leaderless&rsquo; secretory pathways that is used by FGF and IL-1^ref. The secretion of galectins is regulated in a differentiation-dependent manner (36, 47&ndash;49). For example, we, as well as others, have demonstrated that inflammatory macrophages that are exudated to peritoneal cavities by thioglycollate secrete 40% of their total galectin-3, while peritoneal resident macrophages, monocytes and alveolar macrophages are unable to secrete galectin-3 actively and retain it intracellularly (48,50,51). From the point of view that galectin-3 acts as a potential molecule to send &lsquo;danger signals&rsquo;, the differentiation-specific secretion of galectin-3 from macrophages is intriguing. When infectious pathogens or some stimulants are introduced in the body, the first population of leukocytes to be recruited at the affected sites are not inflammatory macrophages but neutrophils, which are able to control the majority of mild infections such as E. coli infections. Interestingly, it appears that  neutrophils3 cannot secrete galectin-3 through the leaderless secretory pathway leaving very little extracellular galectin-3 available (51). In contrast to neutrophils, &ldquo;activated&rdquo; inflammatory macrophages, which can secrete galectin-3 (48,52,53), generally emigrate to the affected sites when the infections are persistent and have not been controlled by the exudated neutrophils (23,54&ndash;56). Thus,
the emigration of inflammatory macrophages itself is one of the situations where our primary defense system is to send the &lsquo;danger signals&rsquo; to the general immunity for further protection. Since galectin-3 is actively secreted from inflammatory macrophages but not from resident peritoneal macrophages, such strict control of galectin&rsquo;s secretion could also be a clear indication that galectins are molecules of &lsquo;danger signals&rsquo; for the innate immunity.
D. Potential Galectins&rsquo; Ligands Involved in Immune Responses Once galectins are secreted or released extracellularly, galectins bind to their ligands, which are expressed on the cell surface (Fig. 2). While galectins were categorized as ?-galactoside binding lectin, the majority of galectins, with the exception of galectin-10 (57), show a preference for lactose/Nacetyllactosamine (Gal ?1-3(4)Glc(NAc)) structures which are ubiquitously found in mammalian cells (16&ndash;19). Studies on the protein structures of galectins have also revealed that the carbohydrate binding pockets of the CRD differ among galectins; for example, galectin-3 appears to have an extended binding pocket compared to galectin-1 (58&ndash;60). Such a structural difference could be related to the different carbohydrate binding specificity observed among galectins; extension at the non-reducing terminal lactosamine units with Gal?1-3, GalNAc?1-3, Fuc?1-2 or Neu5Ac?2-3 substituents enhances affinity to some galectins such as galectin-3 but has a lesser effect or reduces binding of others, such as galectin-1 (61&ndash;63). Hence, some of the surface glycoconjugate ligands could be recognized by various galectins, while others could be relatively specific to a particular galectin (14). Various surface ligands of galectins that are implicated in the regulation of immune responses include CD2, 3 (64), 7 (65), 11b (66), 43, 45 (65), lysosomal associated membrane glycoproteins (LAMP) 1, 2 (67,68), Thy-1 (69) and laminin (70) in the case of galectin-1. Galectin-3 can bind to CD11b (71), CD32(Fc?RII) (72), CD49a/CD29 (73), CD66 (74,75), CD98 (71), LAMP (71,74,76), laminin (62,77&ndash;79), and fetal tissue-derived but not plasma fibronectin (62).
E. Immunomodulative Activity of Galectins
Galectins either autocrinely bind back to the cells or paracrinely bind to the adjacent cells once released or secreted. The number of reports on galectins- inducing various activities related to immune responses has overwhelmingly increased in the last decade and it has begun to be difficult to decipher the biological significance of all those activities (16&ndash;19). Some clues might rest in the fact that galectins are soluble cytosolic proteins, the extracellular release of which is controlled in various ways, as discussed above. Thus when and where galectins are released or secreted is important in order for them to exert their activities. Another clue would be the fact that soluble galectins could cross-link the surface ligands in at least three different ways (Fig. 2A, B and C), although those ways are not mutually exclusive. Although important, it remains to be established how the cross-linking of ligands on cell surface by
galectins preferentially induces one way over the others. In each case, the way galectins mediate the cross-linking appears to determine how the cells respond to extracellular galectin.
F. Cell Adhesion
One of the initial innate immune responses to infection is leukocyte recruitment to the sites of infection&mdash;this process requires strict coordination of cell-cell adhesion. The recruited leukocytes are required to engulf the pathogenic entity by phagocytosis, and release various bactericidal factors, to control the infection, and cytokines, which regulate the subsequent response towards the infection. Leukocyte recruitment to the infected/inflammatory sites involves their movement from the bloodstream to the infected sites through an extravasation process, called emigration or diapedesis. Cell adhesion molecules play a critical role in this recruitment process by mediating the interaction of leukocytes with the vascular endothelial cell barrier separating them from the infected tissue (80&ndash;82). As shown in Fig. 3, tight adhesion of leukocytes to the vascular endothelium is required prior to actual diapedesis in order to withstand the shear stress imparted by blood flow (80&ndash;82). Recent reports have demonstrated that galectins promote leukocyte adhesion to extracellular matrices, dendritic cells and the vascular endothelium (16&ndash;19,51,83). Galectin-1 mediates the adhesion of splenocytes to the extracellular matrix protein laminin (84), T lymphoblastoid cells to thymic epithelial cells (85) and neutrophils to the endothelium4 . Galectin-3 can also promote the binding of L-selectin-triggered lymphocytes to dendritic
cells (83) and the adhesion of neutrophils to laminin (86), and to the endothelium (51). Such galectin promotion of cell adhesion is induced in at least two ways: (i) direct cross-linking of the surface ligands expressed on the different cells (Fig. 2A) and (ii) the activation of classical cell adhesion molecules integrins through galectins&rsquo; ligand clustering (Fig. 2B) (18). While data indicate that both mechanisms (Fig 2A and B) participate to promote leukocyte adhesion, some reports have demonstrated that a significant proportion of galectin promotion of cell adhesion is mediated directly through cross-linking of the cells (51,86). As a matter of fact, galectin promotion of cell adhesion could be observed either at 4 °C, at which temperature cells could not transduce inside-out signal efficiently to activate integrins, or in the absence of cations, which are essential for the integrin activity (51,86). The series of reports employing in vitro cell adhesion assay have demonstrated that galectin can promote the leukocyte adhesion process. The pioneer work of Cohneheim in 1877 using intravital microscopy revealed a sequence in leukocyte emigration events (Fig. 3) (87). First leukocytes begin rolling by adhering to the vascular endothelium inside the blood vessel adjacent to the infection/inflammation site. They tether and form tight adhesion with the endothelium, and begin crawling until they reach the intracellular junction of the endothelium, where they insert their pseudopods in between the tightly apposed endothelial cells and &ldquo;crawl&rdquo; through&mdash; an action termed transendothelial migration. Transinterstitial migration follows as the leukocyte continues migrating through the connective tissues (interstitium). Finally, in transepithelial migration, the leukocyte completely exits by passing through the epithelium into the surrounding tissue where it can participate in the inflammatory response against infection. > 100 years after the initial observations of leukocyte extravasation, we know that two cell adhesion molecules, selectins and integrins mediate leukocyte recruitment to the affected site in the case of general infected tissues (skin and peritoneal cavity) and acute lung inflammation (80&ndash;82,88&ndash;91). If those adhesion molecules that are stably integrated to the plasma membrane always play critical roles in leukocyte emigration, how can we implicate the unique soluble adhesion molecules, galectins in the paradigm of leukocyte emigration? If galectins can act as adhesion molecules, when and how is such mediation initiated during inflammation? The series of studies using mice lacking classical neutrophil adhesion molecules, ?2-integrins5 or selectins, and the observation of patients with leukocyte adhesion deficiency-1, who lack the expression of ?2-integrins, have now demonstrated that in some infections and chronic inflammations, those classical
adhesion molecules do not necessarily play critical roles in neutrophil emigration (92&ndash;97). For example, neutrophils are recruited to the lungs infected with Streptococcal pneumoniae6 in a ?2-integrins and selectins independent manner(91&ndash;98). Since tight adhesion of neutrophils to the endothelium is essential prior to the emigration from bloodstream, other adhesion proteins are expected to be involved in the process. Galectin-3 could act as a unique adhesion molecule in the recruitment of neutrophils in lungs infected by S. pneumoniae (51). Upon this infection, galectin-3 is released and our data suggest that the galectin-3 participates as an adhesionmolecule in the initial step of ?2-integrin-independent neutrophil emigration. The regulation of galectin-3-dependent adhesion is apparently unique compared to the regulation of ?2-integrin activity, which is regulated by the neutrophils insideout signal induced by chemoattractants released and presented on the endothelium at the affected sites (80,82). Without alveolar infection with S. pneumoniae, galectin-3 is absent on the surface of neutrophils or endothelial cells, as they cannot secrete galectin-3. Once the S. pneumoniae infection is initiated in the lungs, preexisting lung macrophages and possibly other alveolar parenchymal cells secrete galectin-3, which binds immediately to neutrophils and endothelial cells (both express galectin-3 ligands) resulting in neutrophil adhesion. ?2-Integrin-independent neutrophil emigration is also observed in some peritonitis models and interestingly, the presence of inflammatory macrophages in the peritoneal cavity appears to be prerequisite for the induction of this ?2-Integrin-independent pathway (99,100). As discussed above, such inflammatory peritoneal macrophages, but not resident peritoneal macrophages, actively secrete galectin-3 (48). Thus, it is possible that galectin-3 secreted from inflammatory macrophages plays a role as an adhesion molecule