Brodell, Lindsey A. BA*; Rosenthal, Kenneth S. PhD†
The skin serves as a wall-like barrier that separates the inside of our body from the microbial enemies of the environment and provides a primary defense against infection.1,2 The layers of the skin, like the outer wall and secondary inner walls surrounding a medieval city, not only provide protection from external enemies, but also provide niches where normal flora bacteria and fungi can live and conduct business. When portions of the wall are disrupted or broken, enemies have access to the critical "inner sanctum" where they can cause major damage. Therefore, the walls are constantly monitored by sentinels who call for rapid reinforcement from a garrison of innate and immune host defenders housed within the city on the first sign of attack. Oftentimes, the spaces between the layers of the wall or the openings for hair follicles become a battlefield between squatters, invaders, and host defenders. For the skin, the peripheral damage on that battlefield may be seen as inflammatory rashes, cutaneous infections, and autoimmune skin problems, which keep dermatologists busy in their practices. This review will discuss the structure and host protections of the skin, its normal flora inhabitants, and the disease challenges caused when normal flora oversteps its bounds. A subsequent article will go into more detail on the immunologic battle against microbial attack and the pathological consequences.
STRUCTURE OF THE SKIN
The skin is made up of 2 primary layers that interact anatomically and functionally. The epidermis forms a thin overlying protective coat that is easily regenerated after injury and serves to keep moisture inside the body while resisting external chemical corrosion. The dermis is the thick fibrous portion that gives skin its strength and contains blood vessels, nerves, and adnexal structures, such as hair follicles, sweat glands, and sebaceous glands. The skin rests on a subcutaneous tissue that provides support and padding.
Keratinocytes are the building blocks of the epidermis and comprise the bulk of this layer.2 These cells have the unique ability to regenerate via mitosis and repair any defect as long as the underlying dermis is not damaged. They also minimize transepidermal water loss (TEWL)1 while serving as a barrier to chemical or microbiological attack. The basal layer (stratum basale) of the epidermis is composed of proliferative cuboidal cells that flatten as they move toward the surface and undergo differentiation in the stratum malpighii (stratum spinosum). In the granular layer (stratum granulosum), keratohyaline granules are produced that signal a loss of keratinocyte nucleii and production of a fully keratinized stratum corneum.1 Filaggrin is produced in the granular layer of the epidermis and is cross-linked to form the cornified cell envelope. When produced in appropriate quantities, filaggrin allows the skin to create a tight protective barrier.3,4 Eventually, individual cell migration ceases, and dead squames are firmly interconnected, forming a constantly regenerating confluent outer barrier. As the outer stratum corneum weathers, the dead squames flake off and appear as a scaling on the surface of the skin. The spaces between these flakes serve as a comfortable home for bacterial and fungal squatters known as the normal flora. The antimicrobial barrier function of the skin is primarily localized to the stratum corneum, which limits the invasive growth of bacteria because of its low water content, acidic pH, resident microflora, and surface-deposited lipids.5 It also minimizes water loss and prevents environmental microbes and allergens from entering the body.
Unlike a medieval city, the skin is constantly refortifying its walls as each layer is added to the inner portion of the epidermis, pushing the previous layer toward the surface. Whereas the keratinocytes are firmly connected to each other by a cement substance and desmosomes that allow the epidermis to maintain its integrity,3 other cells are present in smaller numbers in the epidermis that remain anchored in the basal layer (melanocytes) or stratum malpighii (Langerhans cells and Merkle cells) as keratinocytes migrate to the surface. Perhaps, more amazingly, when signaled, cellular migrants, such as T lymphocytes, are able to traffic through the epidermis within the intercellular cement substance and directly interact with resident Langerhans cells and macrophages to support protective responses.
The dermis is the scaffolding composed of collagen and elastic fibers in an extracellular matrix that supports the outer epidermal wall. It is the protective part of a leather jacket (bovine skin), whereas the epidermis is the brown part that is easily scratched or worn away. Every abrasion that draws blood has eroded the epidermis and entered the dermis. Perhaps, most importantly, the dermis provides the pathways that allow the body to transport defenders (see Skin Immunologic Function [Repelling Barbarian Invaders]) to the outer wall through blood vessels and removes damaged skin and dead invaders through the lymphatics. It also houses the nerve endings, which permit the brain to make choices that prevent many accidents that might breach the skin's protective barrier. The dermis is attached to the overlying epidermis at the periodic acid-Schiff-positive basal membrane junction zone. Hemidesmosomes anchor basal cells onto a basal lamina and, at the same time, send anchoring filaments and microfibrils into the dermis, firmly attaching these 2 layers.6
The human papilloma virus (HPV) takes advantage of the development of the keratinocyte and the layering of the epidermis during its life cycle. After entering through microcracks in the skin, HPV accesses and infects the keratinocytes and becomes a squatter within these cells. Perfectly happy to keep a low profile in the keratinocytes, HPV stimulates the growth of these cells and at the same time promotes the replication of the viral DNA genome. The growing cells expand the epidermis and produce a visible wart. These infected cells make very little viral protein, allowing these infections to slip under the surveillance of the immune response. As the keratinocytes mature, move to the surface, and differentiate, specific transcription factors express different types of keratin. These same transcription factors are used by the HPV to produce messenger RNAs for viral proteins. As the keratinocytes mature and move toward the surface, viral proteins are produced, virion particles are assembled, and virus is released with the exfoliation of the surface layer of dead skin cells.
SKIN IMMUNOLOGIC FUNCTION (REPELLING BARBARIAN INVADERS)
As the outermost barrier of the body, the skin is constantly being challenged by microbial invaders. The body will tolerate some of these microbes as a normal flora (see The Role of Normal Flora), but others must be controlled or eliminated. In addition, the skin must signal to the rest of the body upon attack and, if necessary, call for immune reinforcements to control the incursion. The barrier function of the stratum corneum can only work effectively when it is supported by several innate and adaptive immune protection mechanisms to dissuade or kill invading microbes.
Innate Immunity (First and Fast)
Cells of the innate host response sense the presence of microbes by directly interacting with pathogen-associated molecular patterns (PAMPs) on the microbe (such as peptidoglycan in the cell wall of bacteria, lipopeptides of gram-positive bacteria, lipopolysaccharide of gram-negative bacteria, or surface glycans of fungi). Like "keys" fitting into cell-surface locks, these PAMPs interact with toll-like receptors (TLRs) to trigger protective responses. Similarly, fungal mannans can interact with mannose-binding receptors on these cells and other cells to activate responses.7,8
Keratinocytes, which comprise greater than 90% of all epidermal cells,9 provide the structure of the wall and also the capacity for an initial attack on invading microbes. Keratinocytes can detect the presence of bacteria and fungi by sensing the PAMPs with their TLRs,7,8 triggering innate protective responses. Activation of the TLRs stimulates the production of nitrous oxide (NO) and antimicrobial peptides and enzymes. Nitrous oxide is both a potent antimicrobial and an activator of other antimicrobial responses. The antimicrobial peptides produced by keratinocytes include defensins and cathelicidin.10 β-Defensins are cationic antimicrobial peptides and, like the polymyxin antibiotics, can disrupt membranes and kill gram-negative bacteria.11 The efficacy of this weapon is demonstrated by the rarity of gram-negative bacterial skin infections. Production of cathelicidin provides protection against Group A Streptococcus and other infections.12 Production of enzymes, such as RNAse713 and antileukoprotease,14 also provides antimicrobial activity against bacteria and fungi.
Although the molecular missiles produced by the keratinocytes may be sufficient to control a normal microbial infestation, reinforcements may be needed to control a larger infestation to protect a break in the outer wall or to stop a microbial invasion. The same TLR sensors that promote production of antimicrobial peptides by keratinocytes also promote production of chemoattractant molecules called chemokines. Chemokines attract neutrophils, macrophages, and T cells to the site of infection. These molecules include interleukin (IL) 8 (CXCL8) and Rantes.8 Keratinocytes can also produce cytokines, such as granulocyte-macrophage colony-stimulating factor and the acute-phase cytokines, tumor necrosis factor-α (TNF-α), IL-1, and IL-6, to facilitate the activation of neutrophils, Langerhans cells, dendritic cells (DCs), and other cells in the skin.
Dendritic and Langerhans Cells
Dendritic cells are the sentries of the immune response. Myeloid DCs and a special type of DC, the Langerhans cell, are present in the skin. The Langerhans cells are specialized skin cells and can be distinguished from other DCs by the presence of langerin, which is a C-type lectin and Birbeck granules. As sentries, these cells and related precursor DCs are constantly surveying the environment by endocytosing or phagocytosing proteins and particles, chewing up the proteins into peptides and decorating their cell surface with these peptides attached to major histocompatibility complex (MHC) class II molecules.7 An encounter with a microbe or microbial PAMP through an interaction with the appropriate TLRs will activate and promote the maturation and conversion of these cells into DC1 cells (see below). The first thing these cells would do will be to send out a cytokine alarm to other cells. These cytokines include the acute-phase cytokines (IL-1, TNF-α, and IL-6), interferon-α (IFN-α), IL-12 and chemokines. The acute-phase cytokines promote systemic effects such as fever,15 and TNF-α will activate surrounding DCs in an autocrine manner.16 Next, the DCs pack up the evidence of the infection, change their appearance to optimize their ability to initiate antigen-specific immunity, and take antigenic evidence of the infection to the draining lymph node for presentation to T cells. Dendritic cells are the only type of antigen-presenting cells capable of initiating an immune response. The IL-12 produced by DC1 cells will activate natural killer (NK) cells and will also steer the subsequent immune response so that it will return to the skin with both cell-mediated and antibody-mediated protections (a helper T cell class 1 [TH1] type of immune response [see below]).
Other Cells of Immunity
The skin contains other cells that are part of innate protections including macrophages, NK cells, and NK T cells. These cells are a militia that can be directly activated by microbes. Natural killer cells are also activated by IL-12 and IFN-α, which is produced by DCs and macrophages. Interferon-γ released by the NK cells and the NK T cells can activate macrophages, making them angry. The angry macrophages increase levels of phagocytosis and produce NO and other antibacterial chemicals to kill bacteria more efficiently. Interferon-γ was initially called macrophages activation factor and is the key component of TH1 immune responses. The IFN-γ-treated macrophages also produce more IL-12 to reinforce the response.
Acquired Immunity: Enhancing the Ability to Call Reinforcements
A break in the skin wall and an invasion of the village require specific military reinforcement. Reinforcements come from the lymph node in the form of antigen-specific T cell-mediated and antibody immune responses. The DC1 sentry cells coming from the site of infection bring a peptide piece of a microbial protein and present this evidence to helper T cells (CD4 cells) and cytotoxic T cells (CD8 cells). The call to arms for the helper T cells is provided by a combination of cell-to-cell and cytokine interactions. The evidence of the infection presented to the T cells consists of representative peptides from the microbe displayed on the MHC molecules. To the CD4 T cells, 11 amino acid peptides are displayed on MHC II molecules, whereas the CD8 T cells see 8 amino acid peptides on MHC I molecules. The IL-6 produced by the DCs awakens the T cells from their bureaucratic (unreactivity due to T-regulator cells) slumber, whereas IL-12 directs the response to a TH1-type response. Helper T cell class 1 responses are characterized by IL-2 and IFN-γ-driven responses. The development of an immune response can be described as a drama7 about a battle between microbe and host.17
Once activated, the T cells move from the lymph node into the circulation. Chemokines produced at the site of infection and modifications to the capillary endothelium near the site promote the accumulation of T cells at the site of infection. Further interactions of the T cells with antigen-presenting macrophages keep the T cells on site. As long as the macrophages continue to present antigen and make IL-12, these T cells will continue to make IFN-γ to activate the macrophages and reinforce the cycle. Production of NO and cytokines by the macrophages can be protective and also contribute to immunopathology.
THE ROLE OF THE NORMAL FLORA
Just as villages of loyal subjects, pacified neighbors, and even squatters provide a buffer against attack of a centrally located castle, normal flora (commensal organisms) on the surface of the skin develops a symbiotic relationship with the host and plays a critical role in defending the body against microbiological attack. The environment which supports normal flora is modulated by the condition of the stratum corneum and eccrine, apocrine, and sweat gland secretions. Glandular secretions contain lipids and fatty acids that maintain a pH of 3 to 5 on the cutaneous surface, creating an inhospitable surface for the growth of many organisms. The normal flora of humans consists of more than 200 bacterial species and a few eukaryotic fungi. Common normal flora of the skin in dry areas includes Staphylococcus epidermidis, Micrococcus organisms, propionobacteria, hair follicle mites, and Pityrosporon yeast.18 Moist areas, such as the toe webs and axilla, harbor a greater diversity of skin flora including corynebacteria, mycobacteria, Staphylococcus aureus, and gram-negative bacteria.18 Cleanliness and the use of antibiotic soaps also play a role in adjusting the normal flora and suppressing the pathogens.
Many of the normal flora have specialized enzymes and processes that allow them to "live off the land" and use the dead skin cells as food. Among the fungi, the dermatophytes make keratinases, enzymes that digest keratin for food.
The normal flora suppresses the growth of pathogenic bacteria in several ways. First, niche occupancy and competition for nutrients limit the growth of bacteria other than the normal flora. Second, secretion of inhibitory metabolic products, including acetic and propionic acids, potentiates the low pH favored by the normal flora but inhibit many pathogenic bacteria. The normal flora may also secrete antibiotic-like substances such as penicillin and azelaic acid,5 antifungal volatiles, and surfactins that can dissolve the lipid membrane or envelope of competing pathogenic viruses and bacteria.19 As previously mentioned, the normal flora also stimulates production of antimicrobial substances by keratinocytes and other cells to which they are impervious.
The importance of normal flora is demonstrated by the increased susceptibility of sterile gnotobiotic animals raised in a sterile environment19 to Salmonella compared with animals with a normal flora. In humans, this is analogous to the overgrowth of Candida yeast after the reduction of normal vaginal flora because of a systemic antibiotic treatment.20
In some cases, pathogens are maintained as a normal flora in a carrier state until a situation arises for them to establish a pathogenic infection. For example, S. aureus may be found in the nares of some individuals and will lead to recurrent impetigo in areas of the skin, often on the face, where the skin barrier has been damaged, such as in excoriated eczema.21 Only when the nares are treated with mupirocin ointment twice a day for 7 to 14 days is the staphylococcal carrier state eliminated and recurrent impetigo prevented.21
Erosion of the Walls Can Lead to Cutaneous Infection
Erosion of the wall by microbial or other means can enlarge the microbial village or provide opportunities to new microbial squatters. Some fungi produce enzymes in addition to keratinases that disrupt the epidermis to provide access into the stratum corneum but rarely below the granular layer of the epidermis.22 These superficial and cutaneous dermatophytes are keratinophilic and keratinolytic and include Trichophyton, Epidermophyton, Microsporum, and Epidermophyton species, which cause tinea disease of various sites on the body.
The structure of the skin can also be compromised by bacterial proteins. S. aureus produces exfoliative toxins that sabotage the skin structure from the inside. These toxins specifically act on the granular cell layer to produce subcorneal blisters.23 On a local level, the toxin produces bullous impetigo. In babies with immature kidneys or patients with poor renal function the toxin spreads to cause blistering on the entire body called staphylococcal-scalded skin syndrome.24 The loss of the stratum corneum in these patients removes the primary barrier to defense against other pathogenic bacteria. Newer strains of community-acquired methicillin-resistant S. aureus also contain the Panton-Valentine leukocidin. This toxin can enhance the severity of skin and soft tissue infections, especially furuncules, cutaneous abscesses, and cellulitis by compromising the protection afforded by the host's neutrophil and macrophage militia.[ 25,26]
Patients with atopic dermatitis (also known as eczema) are at increased risk to skin infections because their skin barrier walls are compromised. Atopic dermatitis is associated with mutations in genes that control the production of filaggrin.27,28 A deficiency in the filaggrin protein compromises the barrier function of the skin, increases TEWL,29 and permits bacteria to invade the skin more easily, producing impetiginization and cellulitis. These individuals with abnormal stratum corneum are also at increased risk of widespread herpes simplex virus infections (eczema herpeticum/Kaposi varicelliform eruption). With the reinitiation of smallpox vaccination programs for military personnel, individuals who have the aforementioned abnormal skin barrier function may experience serious vaccinia virus infections (eczema vaccinatum). Close contacts of vaccinated military personnel who have atopic eczema are also at risk for this widespread blistering condition if fluid from the vaccination site comes in contact with their skin.
Lack of Innate Ammunition Can Lead to Cutaneous Infection
In addition to the structural problems that cause eczema patients to have increased TEWL, virtually all eczema patients develop secondary bacterial infections, which are associated with low levels of defensins.30 The importance of defensins is highlighted by patients with psoriasis vulgaris with increased levels of TEWL, but normal levels of defensins and these individuals rarely have secondary skin infections. The infections present in eczema patients potentiate the cycle of inflammation and certainly aggravate the pruritus, redness, swelling, and scaling associated with eczema. This explains the improvement that eczema patients experience when treated with intermittent courses of broad-spectrum antibiotics. Anti-inflammatory treatments (eg, topical steroids) when used, will also suppress the overactive innate/immune system in patients with eczema to help break the scratch-itch cycle, which is central to this condition that has been termed the itch that rashes.
Individuals who are deficient in CD4 T-cell function, for example, acquired immunodeficiency syndrome patients, can develop a variety of opportunistic infections of the skin because of their inability to reinforce local protections. These include severe seborrheic dermatitis from overgrowth of Pityrosporon yeast, persistent Candida intertrigo, widespread dermatophyte infections and impetigo, especially staphylococcal impetigo.31 Correction of the immune deficiency diminishes the susceptibility to these infections.31 In another condition, the inability to elicit protective cytokines because of a specific primary defect in recognizing Candida antigens in the skin can cause mucocutaneous candidiasis.31 This results in oral candidiasis (thrush), Candida infections beneath the proximal nail fold (paronychia), grotesque thickening of nails, and destructive changes in the skin and subcutaneous tissues but with little propensity for systemic dissemination.32 Because the T cells cannot be recruited to marshal an inflammatory response to kill the Candida species, systemic antifungal agents, like fluconazole, are required to clear the cutaneous disease.
Breaks in the Wall Allow Entry of Microbial Marauders
Surgical wounds, intravenous sites, abrasions, insect bites, lacerations, and excoriated dermatitis provide portals of entry for microbial infection. Community-acquired methicillin-resistant S. aureus in football players routinely involve these epidermal defects.33 Similarly, abrasion of the cornified layer of skin and contact with an opponent's facial herpetic vesicular lesions in wrestlers and rugby players commonly lead to infections termed herpes gladiatorum and herpes rugbiorium. Many hospital-acquired infections involve these induced portals of entry and would not occur in areas of intact skin.
The skin barrier requires complex structural elements, mechanisms to repair any structural defects, an intact immune system, and a healthy growth of resident normal flora to successfully fend off pathogenic bacteria and other infections. The analogy of a castle is an appropriate metaphor. The fortified walls are strong in their construction, and defense is aided by villages of symbiotic organisms (normal flora). The defenses are personned by sentries, some of whom have the innate ability to recognize and attack invaders. These defenses can be bolstered by a host of specialized immune soldiers that respond to the alarm. Some microbes can overcome these barriers and defenses, whereas others take advantage of breaks in the barrier or reductions in the host defenses to establish pathogenic infections and disease. In the end, the battle between microbes and the body can resolve the infection but may leave scars on battlefield.17
The authors thank Robert Brodell for his careful reading of the manuscript.
1. Marks R. The stratum corneum barrier: the final frontier. J Nutr. 2004;134(8 suppl):2017S-2021S.
2. Lee SH, Jeong SK, Ahn SK. An update of the defensive barrier function of skin. Yonsei Med J. 2006;47(3):293-306.
3. Elias PM. Stratum corneum defensive functions: an integrated view. J Invest Dermatol. 2005;125:183-200.
4. Sandilands A. Filagrin's fuller figure: a glimpse into the genetic architecture of atopic dermatitis. J Investigative Derm. 2007;127:1282-1284.
5. Elias PM. The skin barrier as an innate immune element. Semin Immunopathol. 2007;29(1):3-14.
6. Bruckner-Tuderman L, Hopfner B, Hammami-Hauasli N. Biology of anchoring fibrils: lessons from dystrophic epidermolysis bullosa. Matrix Biol. 1999;18(1):43-54.
7. Rosenthal KS. Vaccines Make Good Immune Theater: immunization as described in a three act play. Infect Dis Clin Pract. 2006;14:35-45.
8. Pivarcis A, Nagy I, Kemeny L. Innate immunity in the skin: how keratinocytes fight against pathogens. Curr Immunol Rev. 2005;1:29-42.
9. Salmon JK, Armstrong CA, Ansel JC. The skin as an immune organ. West J Med. 1994;160(2):146-152.
10. Braff MH, Bardan A, Nizet V, et al. Cutaneous defense mechanisms by antimicrobial peptides. J Invest Dermatol. 2005;125(1):9-13.
11. Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat Rev Immunol. 2003;3(9):710-720.
12. Murakami M, Ohtake T, Dorschner RA, et al. Cathelicidin anti-microbial peptide expression in sweat, an innate defense system for the skin. J Invest Dermatol. 2002;119(5):1090-1095.
13. Harder J, Schroder JM. RNase 7, a novel innate immune defense antimicrobial protein of healthy human skin. J Biol Chem. 2002;277(48):46779-46784.
14. Sallenave JM. Antimicrobial activity of antiproteinases. Biochem Soc Trans. 2002;30(2):111-115.
15. Kapetanovic R, Cavaillon JM. Early events in innate immunity in the recognition of microbial pathogens. Expert Opin Biol Ther. 2007;7(6):907-918.
16. Saalbach A, Klein C, Sleeman J, et al. Dermal fibroblasts induce maturation of dendritic cells. J Immunol. 2007;178(8):4966-4974.
17. Rosenthal KS. Are microbial symptoms "self-inflicted"? The consequences of immunopathology. Infect Dis Clin Pract. 2005;13:306-311.
18. Wolff K, Goldsmith LA, Katz SI, et al. Fitzpatrick's Dermatology in General Medicine. New York, NY: McGraw Hill, Health Prof Div;2008.
19. Que JU, Hentges DJ. Effect of streptomycin administration on colonization resistance to Salmonella typhimurium in mice. Infect Immun. 1985;48(1):169-174.
20. Pirotta MV, Garland SM. Genital Candida species detected in samples from women in Melbourne, Australia, before and after treatment with antibiotics. J Clin Microbiol. 2006;44(9):3213-3217.
21. Wenzel RP, Perl TM. The significance of nasal carriage of Staphylococcus aureus and the incidence of postoperative wound infection. J Hosp Infect. 1995;31(1):13-24.
22. Simpanya M. Dermatophytes: their taxonomy, ecology, and pathogenicity. In: Kushwaha RKS, Guarro J, eds. Biology of Dermatophytes and Other Keratinophilic Fungi. Bilbao, Spain: Revista Iberoamericana de Micología; 2000:1-12.
23. Curran JP, Al-Salihi FL. Neonatal staphylococcal scalded skin syndrome: massive outbreak due to an unusual phage type. Pediatrics. 1980;66(2):285-290.
24. Ladhani S, Joannou CL, Lochrie DP, et al. Clinical, microbial, and biochemical aspects of the exfoliative toxins causing staphylococcal scalded-skin syndrome. Clin Microbiol Rev. 1999;12(2):224-242.
25. Voyich JM, Otto M, Mathema B, et al. Is Panton Valentine Leukocidin the Major Virulence Determinant in Community-Associated Mehicillin-Resistant Staphylococcus aureus disease? J Inf Dis. 2006;194:1761-1770.
26. Voyich JM, Braghton KR, Sturdevant DE, et al. Insights into Mechanisms Used by Staphylococcus aureus to Avoid Destruction by Staphylococcus aureus to Avoid Destruction by Human Neutrophils. J Immunol. 2005;175:3907-3919.
27. Seguchi T, Chang-Yi C, Kusuda S, et al. Decreased expression of filaggrin in atopic skin. Arch Derm Res. 1996;288(8):442-446.
28. Sandilands A. Fillagrin's fuller figure: a glimpse into the genetic architecture of atopic dermatitis. J Investigative Derm. 2007;127:1282-1284.
29. Sugarman JL, Fluhr JW, Fowler AJ, et al. The objective severity assessment of atopic dermatitis score: an objective measure using permeability barrier function and stratum corneum hydration with computer assisted estimates for extent of disease. Arch Dermatology. 2003;39:1417-1422.
30. Ong PY, Ohtake T, Brandt C, et al. Endegenous Antimicrobial Peptides and Skin Infections in Atopic Dermatitis. NEJM. 2002;347(15):1151-1160.
31. Rigopoulos D, Paparizos V, Katsambas A. Cutaneous markers of HIV infection. Clin Dermatol. 2004;22(6):487-98.
32. Lilic D. New perspectives on the immunology of chronic mucocutaneous candidiasis. Curr Opin Infect Dis. 2002;15:143-147.
33. Nguyen DM, Mascola L, Bancroft E. Recurring methicillin-resistant Staphylococcus aureus infections in a football team. Emergency Med News. 2005;27(3):54-59.
© 2008 Lippincott Williams & Wilkins, Inc.