Germinal center architecture disturbance during Plasmodium berghei ANKA infection in CBA mice
© Carvalho et al; licensee BioMed Central Ltd. 2007
Received: 29 January 2007
Accepted: 16 May 2007
Published: 16 May 2007
Immune responses to malaria blood stage infection are in general defective, with the need for long-term exposure to the parasite to achieve immunity, and with the development of immunopathology states such as cerebral malaria in many cases. One of the potential reasons for the difficulty in developing protective immunity is the poor development of memory responses. In this paper, the potential association of cellular reactivity in lymphoid organs (spleen, lymph nodes and Peyer's patches) with immunity and pathology was evaluated during Plasmodium berghei ANKA infection in CBA mice.
CBA mice were infected with 1 × 106 P. berghei ANKA-parasitized erythrocytes and killed on days 3, 6–8 and 10 of infection. The spleen, lymph nodes and Peyer's patches were collected, fixed in Carson's formalin, cut in 5 μm sections, mounted in glass slides, stained with Lennert's Giemsa and haematoxylin-eosin and analysed with bright-field microscopy.
Early (day 3) strong activation of T cells in secondary lymphoid organs was observed and, on days 6–8 of infection, there was overwhelming activation of B cells, with loss of conventional germinal center architecture, intense centroblast activation, proliferation and apoptosis but little differentiation to centrocytes. In the spleen, the marginal zone disappeared and the limits between the disorganized germinal center and the red pulp were blurred. Intense plasmacytogenesis was observed in the T cell zone.
The observed alterations, especially the germinal center architecture disturbance (GCAD) with poor centrocyte differentiation, suggest that B cell responses during P. berghei ANKA infection in mice are defective, with potential impact on B cell memory responses.
Malaria remains one of the major public health problems in the developing world, with an estimate of 300–500 million cases and 1–3 millions deaths every year. The development of a vaccine has been one of the research strategies to face this threat, especially with the spreading of parasite resistance to many of the drugs currently available to treat the disease. However, despite over 30 antigens having been identified as vaccine candidates and many of them having been tested in pre-clinical and up to phase III clinical trials, none of them has so far generated a solid perspective for a vaccine to be available in the coming years . One of the reasons malaria vaccine development has been hindered is the fact that the immune responses and the mechanisms responsible for acquisition of immunity to malaria are largely unknown. Acquisition of immunity to malaria in highly endemic areas such as sub-Saharan Africa is considered to be a slow process requiring many years to take place. Exposed children below five years of age acquire protection against severe manifestations of the disease, remaining susceptible to infection and milder morbidity. As age increases, the frequency of clinical attacks decreases and after puberty most individuals (except pregnant women) present a complete immunity against clinical manifestations of the disease. Yet, most individuals remain susceptible to infection, but the parasite load is greatly decreased and very low parasitaemia are prevalent in adulthood [2, 3]. In addition, it seems that this partial, non-sterile, immunity is rapidly lost if the contact with the parasite is discontinued. Many factors seem to contribute to poor immunity in malaria. Among them, it is widely believed that blood stages of the parasite induce immunosuppression and impair the development of immunological memory. This has been questioned by Struik and Riley , who argued that some paradigms, especially the lack of memory in the immune response to the malaria parasite, do not have a solid scientific evidence and may be misleading.
In mice, both cellular and humoral responses play important roles in the immunity against blood stage malaria infection . But besides being involved in protection, immune responses in malaria can also trigger immunopathology . In fact, complications such as cerebral malaria and severe anaemia have a strong immunological component in humans , as well as in experimental models [8–10]. The Plasmodium berghei ANKA infection of CBA mouse is an established model of malaria with neurological involvement (the so-called experimental cerebral malaria), and infection with blood stage parasites leads to 100% lethality. The immune response in this model is not only ineffective against parasite growth, but also responsible for the 60–80% incidence of CM, mainly characterized by strong Th1 T cell responses , macrophage hyperactivation , and also CD8+ T cell cytotoxicity .
Given these characteristics of the immune responses during blood stage malaria, the understanding of the mechanisms leading to poor immunity and immunopathology is crucial for the rational development of prophylactic and therapeutic interventions, such as vaccines. Although immune responses to malaria and physiopathogenesis of cerebral malaria in mice have been widely studied [14–17], detailed analysis of changes in lymphoid compartments, which is basic to understand how the immune system respond to a challenge, has not been widely approached, with a few works mainly on the Plasmodium chabaudi model [18–20]. It has recently been shown that P. berghei ANKA-infected CBA mice develop various degrees of thymocyte apoptosis, leading in some cases to thymocyte depletion and severe thymus cortical atrophy . In the present work, attention was focused on the sequential immunohistopathological changes caused by P. berghei ANKA infection in several lymphoid compartments and their possible implications in pathology and immunity of the infection.
Materials and methods
Animals, parasite and infection
Four to 8 week-old female CBA mice (CECAL/Fiocruz, Rio de Janeiro) were inoculated by intraperitoneal injection with 1 × 106 P. berghei ANKA-parasitized erythrocytes obtained from a donor infected CBA mouse. A total of 71 mice in three separate experiments were infected and lymphoid organs of 32 mice were examined by histology, on the following days after infection: three (ten mice); six (two mice); seven (eight mice); eight (two mice); and 10 (ten mice). Seventeen uninfected control mice (age and sex-matched) were examined on days: three (six mice); seven (six mice); and 10 (five mice). Mice were anesthetized with ketamine plus midazolan and bled. Organs were removed immediately after death and transferred to Carson's formalin-Millonig . Organs from mice that died spontaneously were not collected. Protocols were reviewed and approved by the Fiocruz Ethical Committee on Animal Experimentation (CEUA protocol number P-0155-03).
Parasitaemia and clinical parameters
Density and maturation of circulating parasites were daily determined by examination of Giemsa-stained blood smears collected from the tail vein. Clinical CM was defined by the presentation of at least one of the following signs: limb paralysis, convulsions, stupor and roll over.
Spleen, inguinal and periaortic lymph nodes and Peyer's patches were fixed in Carson's formalin-Millonig . After fixation, the spleen was cut in four pieces in transversal sections, mounted in a box together with four lymph nodes and four Peyer's patches and embedded in paraffin. Sections (5 μm) were stained with haematoxylin-eosin and Lennert's Giemsa , and analyzed by bright-field microscopy. Lennert's Giemsa is the best stain to characterize, by bright-field, the cellular populations in lymphoid organs because stain RNA and DNA in blue (allowing to infer the presence of ergastoplasm and the characteristic of nucleoli), acidophilic substances in red by the eosin (discriminates eosinophils) and acid mucopolisaccharides in red-violet (labels mast cell granules).
Histopathology of the spleen
Histopathology of the lymph nodes
Histopathology of the Peyer's patches
Non-infected control mice presented quiescent Peyer's patches, with resting small lymphocytes, prominent dome region and several dispersed macrophages (Figure 4C–D). Overall, the alterations during infections were similar to those observed in spleen white pulp and lymph nodes, with many immunoblasts on day 3, intense centroblast activation, proliferation and apoptosis in disorganized GCs on days 6–8. In addition, there was a great depletion of lymphocytes in the dome region on days 6–8, turning the dome monocytes confluent and more evident, resulting in shrunk patches (Figure 4E–F). This histological finding is in accordance with the atrophy of the Peyer's patches observed macroscopically.
P. berghei ANKA infection in CBA mice leads to CM and multiorgan pathology, and much of its physiopathogenesis is owed to players of the immune system, with critical involvement of cytokines, T lymphocytes, macrophages and other mediators. The activation of the immune system in this model, whose intensity leads to the pathological consequences, is however largely ineffective to halt parasite growth and even mice that do not develop CM eventually die of hyperparasitaemia. This study shows that P. berghei ANKA infection in CBA mice leads to striking changes in secondary lymphoid organs.
Early (day 3) in the infection, there was a strong T cell response in lymph nodes, spleen and Peyer's patches, evidenced by the presence of many immunoblasts in the T cell areas of these organs. CM pathology is dependent on T cells, mainly CD4+ cells , outlining the importance of this early widespread activation of T cells. The most remarkable changes were coincident with the time of CM occurrence, on days 6–8 of infection, consisting mainly of striking activation and proliferation of B cells accompanied by GC architecture disturbance (from now called GCAD). Although polyclonal B cell activation and proliferation lead to high immunoglobulin production in malaria , and a possible role for immunoglobulins in the expression of human CM has been reported , it is unlikely that the atypical GC reaction plays a role in CM pathogenesis in this murine model, since it has been shown that B cell-knockout mice develop CM . Indeed, the GCAD was observed in mice with and without clinical signs of CM on days 6–8, and also in those killed on day 10, which did not develop CM.
The major defect in the GCAD seems to be B cell differentiation (centrocyte transformation), not activation and proliferation (centroblast reaction), which, instead, are overwhelming. In the P. chabaudi model, the presence of Peanut Agglutinin (PNA) positive cells in spleen follicles of infected mice has been considered evidence that a normal GC reaction occurs during infection . However, PNA staining does not differentiate cell types (centroblasts, centrocytes), so it cannot detect the poor cell differentiation as it was detected here by conventional Giemsa staining.
The defect in centrocyte transformation is a phenomenon of utmost importance. The GC is a tightly controlled microenvironment for B cell expansion, differentiation or death [28–30]. Centrocyte transformation is related to affinity-maturation of B cells and to the switch of immunoglobulin isotype. B cells that undergo somatic mutation in the Ig variable region genes and show improved affinity for the antigen are rescued from death at this stage through cooperation with follicular dendritic cells and become centrocytes. If the somatic mutation leads to lower affinity to antigen, the B cell cannot receive the rescue signals from follicular dendritic cells and die by apoptosis. Centrocytes then represent cells that ultimately may become the IgG-producing plasma cells when further stimulated. And more, they will be responsible for the secondary response to reinfection, playing a central role in the immunological memory. Struik and Riley  have made a critical and elegant review of the immunological memory in malaria, challenging the widespread belief that memory response to malaria is defective. However, the data presented here, although derived from a murine model, strengthen the notion that memory B cell response during acute plasmodial infection must in fact be affected, and deficient centrocyte transformation may then be one of the reasons for the difficulty in developing effective immune responses after repeated malaria infections. Therefore, overcoming this deficiency must be crucial for developing immunoprophylactic interventions such as vaccines.
Another aspect of the B cell response in GCAD is the occurrence of intense apoptosis. Some papers describe that apoptosis is increased during malarial infections in humans [31, 32] and in mice [18, 19]. By using flow cytometry, Helmby and coworkers  showed that the vast majority of apoptotic cells in the spleen of P. chabaudi-infected mouse were B cells, consistent with the present histological data. Different interpretations are given to this event, such as a physiological feedback mechanism to control the number of cells after clonal expansion and also an escape mechanism through which the parasite would cause immune unresponsiveness by causing lymphocyte death. Given the GCAD, apoptosis was interpreted as a consequence of the defective centrocyte transformation. Since proliferation is intense and differentiation with affinity maturation is affected, intense apoptosis should indeed be expected to occur as a consequence.
Although centrocyte transformation seems to be impaired, plasmacytogenesis is overwhelming. Plasmablasts could be derived from the marginal zone B cells, known to respond rapidly to blood-borne antigens due to their location adjacent to the marginal sinuses and their proximity to antigen-trapping macrophages and dendritic cells . Following activation, marginal zone B cells become plasmablasts and exit the marginal zone, undergoing a burst of proliferation and forming plasma cell foci in extrafollicular regions such as the periarteriolar lymphoid sheath , as in the present case. This could help explaining the disappearance of the marginal zone on days 6–8 of infection. This possibility has been proposed by others, who identified cells with phenotypic characteristics of marginal zone B cells spread throughout the spleen during the acute phase of the P. chabaudi infection . It is important to point out that marginal zone B cells do not represent postgerminal center T-dependent B cells and proliferate much faster in response to very low doses of polyclonal mitogens and terminally differentiate, without the participation of germinal center, into plasma cells within hours of activation .
Some works have addressed the lymphoid reactions in malaria infections, most data coming from the P. chabaudi model and focused in spleen changes. Sanchez-Torres and coworkers  have described the enlargement and disarray of the white pulp during acute P. chabaudi infection in CB6F1 mice. The disarray was characterized by the lack of definition of the limits between white and red pulps and, at the time of greater number of apoptotic cells, they described an almost complete disappearance of the red pulp. In that model, mice spontaneously recover of infection and white pulp reacquired organization three weeks after the control of parasitaemia. Although that work did not go into detail on the description of the specific white pulp compartments where the disarray occurred, it is likely that this event may be similar to that observed in the P. berghei ANKA model. However, in the latter, the red pulp was enlarged, contrary to what was described for the P. chabaudi model.
Urban and coworkers  have performed an integrated histological and immunochemistry analysis of spleens from patients that died of severe P. falciparum malaria, including cases of cerebral malaria. They describe, among other alterations, the cell depletion in the marginal zone, the lack of GC formation and a decrease in the number of B cells in the spleen. The authors interpret the lack of GCs and overall disorganization in spleen architecture as playing possible roles in the poor memory responses in malaria. Similar conclusions were reached in the present work but, as discussed above, based in different data and interpretations. First, in the murine model, GC formation is not absent, it is indeed remarkable, but it is highly disturbed, particularly the B cell differentiation. Second, in contrast to the data of Urban and coworkers, B cell numbers are not decreased, actually follicular B cell proliferation and plasmacytogenesis (GC dependent and/or independent?) are overwhelming and, despite the red pulp enlargement, B cell expansion seems to be the main responsible for the observed splenomegaly. These differences may be real, due to differences in the lymphoid reactions in humans and in the mouse, but can most probably be related to the method of study. Urban and coworkers based their definition of lymphoid compartments and cell types on HE staining and immunophenotyping (in the case of B cells, by using an anti-CD20 monoclonal antibody). In the present work, findings were based on morphology revealed by Giemsa staining. Both approaches present limitations. Immunohistochemistry can provide a good definition of the distribution of different cell types. However, since this is based on the detection of specific cell markers by monoclonal antibodies, subpopulations that may change the expression of a given marker during differentiation can become undetectable depending on the monoclonal antibody used. This is a main difficulty in characterizing mature B cell populations, because changes in morphological characteristics due to activation, proliferation and differentiation are accompanied by phenotypic changes in cell surface and cytoplasmic markers, such as IgD, IgM, IgG, CD19, CD20, B220 and others. For instance, CD20 is not uniformly expressed in the mature B cell lineage, being expressed in naïve and activated B cells but not in plasmablasts or plasma cells , and it is then possible that these cells can be missed in immunohistochemistry studies based on this marker. A panel of monoclonal antibodies is then necessary to overcome this kind of limitation. Similarly, conventional histological stainings such as Giemsa, although providing definition of cell types with a high degree of confidence based on differential morphology, do not guarantee 100% certainty. To clarify this point and further characterize the phenomenon, different and complementary techniques (Giemsa-based analysis associated to immunochemistry, for instance) performed in tissues of the same human or animal samples can be envisaged.
The phenomena and the interpretations described in this paper deserve further investigation in order to help unraveling the mechanisms leading to pathogenesis or poor immunity in malaria. The use of other host-parasite combinations and further characterization of cell types can provide a better understanding of the phenomena described here. This is important to allow improved therapeutic interventions and rational designs of effective vaccines.
To Belmira Santos from the Fiocruz Animal House for providing the CBA mice, and Bruno Eschenazi from the Laboratory of Production and Treatment of Images/IOC/Fiocruz for processing the images. This work was supported with funds of the Laboratory of Malaria Research and the Department of Pathology (POM).
- Carvalho LJM, Daniel-Ribeiro CT, Goto H: Malaria vaccine: candidate antigens, mechanisms, constraints and prospects. Scand J Immunol. 2002, 56: 327-343. 10.1046/j.1365-3083.2002.01160.x.View ArticlePubMed
- Druilhe P, Perignon JL: A hypothesis about the chronicity of malarial infection. Parasitol Today. 1997, 13: 353-357. 10.1016/S0169-4758(97)01095-8.View ArticlePubMed
- Baird JK: Age-dependent characteristics of protection v. susceptibility to Plasmodium falciparum. Ann Trop Med Parasitol. 1998, 92: 367-390. 10.1080/00034989859366.View ArticlePubMed
- Struik SS, Riley EM: Does malaria suffer from lack of memory?. Immunol Rev. 2004, 201: 268-290. 10.1111/j.0105-2896.2004.00181.x.View ArticlePubMed
- Perlmann P, Perlmann H, Berzins K, Troye-Blomberg M: Selected problems of malaria blood stage immunity. Tokai J Exp Clin Med. 1998, 23: 55-62.PubMed
- Daniel-Ribeiro CT, Ferreira-da-Cruz MF: The new and the old in malaria immunopathology. Ciencia e Cultura. 2000, 52: 269-281.
- Mackintosh CL, Beeson JG, Marsh K: Clinical features and pathogenesis of severe malaria. Trends Parasitol. 2004, 20: 597-603. 10.1016/j.pt.2004.09.006.View ArticlePubMed
- De Souza JB, Riley EM: Cerebral malaria: the contribution of studies in animal models to our understanding of immunopathogenesis. Microbes Infect. 2002, 4: 291-300. 10.1016/S1286-4579(02)01541-1.View ArticlePubMed
- Wassmer SC, Combes V, Grau GE: Pathophysiology of cerebral malaria: role of host cells in the modulation of cytoadhesion. Ann N Y Acad Sci. 2003, 992: 30-38.View ArticlePubMed
- Carvalho LJM, Alves FA, Oliveira SG, Valle RR, Fernandes AAM, Muniz JAPC, Daniel-Ribeiro CT: Severe anemia affects both splenectomized and non-splenectomized Plasmodium falciparum -infected Aotus infulatus monkeys. Mem Inst Oswaldo Cruz. 2003, 98: 679-686.View ArticlePubMed
- Hunt NH, Grau GE: Cytokines: accelerators and brakes in the pathogenesis of cerebral malaria. Trends Immunol. 2003, 24: 491-99. 10.1016/S1471-4906(03)00229-1.View ArticlePubMed
- Carvalho LJM, Lenzi HL, Pelajo-Machado M, Oliveira DN, Daniel-Ribeiro CT, Ferreira-da-Cruz MF: Plasmodium berghei: cerebral malaria in CBA mice is not clearly related to plasma TNF levels or intensity of histopathological changes. Exp Parasitol. 2000, 95: 1-7. 10.1006/expr.2000.4508.View ArticlePubMed
- Nitcheu J, Bonduelle O, Combadiere C, Tefit M, Seilhean D, Mazier D, Combadiere B: Perforin-dependent brain-infiltrating cytotoxic CD8+ T lymphocytes mediate experimental cerebral malaria pathogenesis. J Immunol. 2003, 170: 2221-2228.View ArticlePubMed
- Li C, Seixas E, Langhorne J: Rodent malarias: the mouse as a model for understanding immune responses and pathology induced by the erythrocytic stages of the parasite. Med Microbiol Immunol. 2001, 189: 115-126. 10.1007/s430-001-8017-8.View ArticlePubMed
- Good MF, Xu H, Wykes M, Engwerda CR: Development and regulation of cell mediated immune responses to the blood stages of malaria: implications for vaccine research. Annu Rev Immunol. 2005, 23: 69-99. 10.1146/annurev.immunol.23.021704.115638.View ArticlePubMed
- Schofield L, Grau GE: Immunological processes in malaria pathogenesis. Nat Rev Immunol. 2005, 5: 722-735. 10.1038/nri1686.View ArticlePubMed
- Lackner P, Beer R, Helbok R, Broessner G, Engelhardt K, Brenneis C, Schmutzhard E, Pfaller K: Scanning electron microscopy of the neuropathology of murine cerebral malaria. Malar J. 2006, 5: 116-120. 10.1186/1475-2875-5-116.PubMed CentralView ArticlePubMed
- Sanchez-Torres L, Rodriguez-Ropon A, Aguilar-Medina M, Favila-Castillo L: Mouse splenic CD4+ and CD8+ T cells undergo extensive apoptosis during a Plasmodium chabaudi AS infection. Parasite Immunol. 2001, 23: 617-626. 10.1046/j.1365-3024.2001.00422.x.View ArticlePubMed
- Helmby H, Jonsson G, Troye-Blomberg M: Cellular changes and apoptosis in the spleens and peripheral blood of mice infected with blood-stage Plasmodium chabaudi chabaudi AS. Infect Immun. 2000, 68: 1485-1490. 10.1128/IAI.68.3.1485-1490.2000.PubMed CentralView ArticlePubMed
- Achtman AH, Khan M, MacLennan IC, Langhorne J: Plasmodium chabaudi chabaudi infection in mice induces strong B cell responses and striking but temporary changes in splenic cell distribution. J Immunol. 2003, 171: 317-324.View ArticlePubMed
- Carvalho LJM, Ferreira-da-Cruz MF, Pelajo-Machado M, Daniel-Ribeiro CT, Lenzi HL: Plasmodium berghei ANKA infection induces thymocyte apoptosis and thymocyte depletion in CBA mice. Mem Inst Oswaldo Cruz. 2006, 101: 523-528.View ArticlePubMed
- Carson FL, Martin JH, Lynn JA: Formalin fixation for electron microscopy: a re-evaluation. Am J Clin Pathol. 1993, 59: 365-373.
- Lennert K: Malignant lymphoma other than Hodgkin's disease. 1978, Springer-Verlag, Berlin, 833-View Article
- Grau GE, Piguet PF, Engers HD, Louis JA, Vassali P, Lambert PH: L3T4+ T lymphocytes play a major role in the pathogenesis of murine cerebral malaria. J Immunol. 1986, 137: 2348-2354.PubMed
- Donati D, Zhang LP, Chene A, Chen Q, Flick K, Nystrom M, Wahlgren M, Bejarano MT: Identification of a polyclonal B-cell activator in Plasmodium falciparum. Infect Immun. 2004, 72: 5412-5418. 10.1128/IAI.72.9.5412-5418.2004.PubMed CentralView ArticlePubMed
- Scholander C, Treutiger CJ, Hultenby K, Wahlgren M: Novel fibrillar structure confers adhesive property to malaria-infected erythrocytes. Nat Med. 1996, 2: 204-208. 10.1038/nm0296-204.View ArticlePubMed
- Yanez DM, Manning DD, Cooley AJ, Weidanz WP, van der Heyde HC: Participation of lymphocyte subpopulations in the pathogenesis of experimental murine cerebral malaria. J Immunol. 1996, 157: 1620-1624.PubMed
- van Eijk M, Defrance T, Hennino A, De Groot C: Death-receptor contribution to the germinal-center reaction. Trends Immunol. 2001, 22: 677-682. 10.1016/S1471-4906(01)02086-5.View ArticlePubMed
- Guzman-Rojas L, Sims-Mourtada JC, Rangel R, Martinez-Valdez H: Life and death within germinal centres: a double-edged sword. Immunology. 2002, 107: 167-175. 10.1046/j.1365-2567.2002.01494.x.PubMed CentralView ArticlePubMed
- Calame KL, Lin KI, Tunyaplin C: Regulatory mechanisms that determine the development and function of plasma cells. Annu Rev Immunol. 2003, 21: 205-230. 10.1146/annurev.immunol.21.120601.141138.View ArticlePubMed
- Toure-Balde A, Sarthou JL, Aribot G, Michel P, Trape JF, Rogier C, Roussilhon C: Plasmodium falciparum induces apoptosis in human mononuclear cells. Infect Immun. 1996, 64: 744-750.PubMed CentralPubMed
- Riccio EK, Junior IN, Riccio LR, Alecrim MG, Corte-Real S, Morgado M, Daniel-Ribeiro CT, Ferreira-da-Cruz MF: Malaria associated apoptosis is not significantly correlated with either parasitemia or the number of previous malaria attacks. Parasitol Res. 2003, 90: 9-18.PubMed
- Martin F, Kearney JF: B-cell subsets and the mature preimmune repertoire. Marginal zone and B1 B cells as part of a "natural immune memory". Immunol Rev. 2000, 175: 70-79. 10.1111/j.1600-065X.2000.imr017515.x.View ArticlePubMed
- Oliver AM, Martin F, Kearney JF: IgMhighCD21high lymphocytes enriched in the splenic marginal zone generate effector cells more rapidly than the bulk of follicular B cells. J Immunol. 1999, 162: 7198-7207.PubMed
- Urban BC, Hien TT, Day NP, Phu NH, Roberts R, Pongponratn E, Jones M, Mai NTH, Bethell D, Turner GDH, Ferguson D, White NJ, Roberts DJ: Fatal Plasmodium falciparum malaria causes specific patterns of splenic architectural disorganization. Infect Immun. 2005, 73: 986-1994.
- Hoyer BF, Mantz RA, Radbruch A, Hiepe F: Long lived plasma cells and their contribution to autoimmunity. Ann N Y Acad Sci. 2005, 1050: 124-133. 10.1196/annals.1313.014.View ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.