Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Feb 12.
Published in final edited form as: Discov Med. 2018 Sep;26(142):79–83.

Hiding in Plain Sight: Time to Unlock Autoimmune Clues in Human CD5+ B Cells by Using NextGen Technology

Rizwan Ahmed 1, Zahra Omidian 1, Thomas Donner 2, Abdel Rahiam A Hamad 1,*
PMCID: PMC7879960  NIHMSID: NIHMS1006183  PMID: 30399325

Abstract

CD5+ B cells expand in many autoimmune diseases, including type 1 diabetes (T1D), rheumatoid arthritis (RA), and systemic lupus erythematosus (SLE). Furthermore, CD5+ B cells contain important subsets: IL-10-producing B-reg cells, FasL-expressing subset, and the majority of pre-naive B cells. In addition, they are major sources of natural autoantibodies, which are polyreactive and autoreactive. Thus, CD5+ B cells are clearly loaded with autoimmune clues that are yet to be unlocked and understood. We hypothesize that human CD5+ B cells are likely to yield enormously important novel information about the role of B cells in autoimmune disease if analyzed using the new technological advances in molecular biology and genomics. Use of high-throughput sequencing of B cell receptors (BCR) of CD5+ B cells could reveal public BCRs associated with autoimmune diseases, whereas transcriptional analysis of CD5+ B cells using single-cell RNA-seq may delineate distinct sublineages and their relationship to conventional B cells. If it turns out that autoimmune repertoires are concentrated in CD5+ B cells, given that CD5+ B cells are clearly identifiable by flow cytometry, therapeutic strategies can be developed to safely remove CD5+ B cells to mitigate ongoing autoimmunity and protect at-risk individuals.

It Is Time to Directly Study Human CD5+ B Cells Instead of Their Murine Surrogates

Unlocking the mystery of human CD5+ B cells is a decades-long quest. It began with the discovery that the pan-T cell marker CD5 is expressed on most human B-type chronic lymphatic leukemia (CLL) cells (Lanier et al., 1981). Analyses over ensuing years, as discussed below, have shown a clear association of CD5+ B cells with autoimmune diseases. However, the use of traditional methods and assays has failed to answer questions as basic as the composition of CD5+ B cells and their relationship to their CD5− counterparts, let alone their role in disease pathogenesis. Furthermore, it remains unclear whether human CD5+ B cells represent a distinct sublineage, similar to murine CD5+ B1 cells, or merely represent activated B cells (Baumgarth, 2017). Sophisticated analysis of the immune cells is now doable not only from peripheral blood but also from lymphoid and non-lymphoid organs of humans. Thus, direct analysis of CD5+ B cells from healthy individuals and patients could be very useful in delineating their composition and lineage specification. In addition, it is expected to provide novel information about their roles in the pathogenesis of autoimmune diseases. Generated clues could then be used to design mechanistic studies in mouse models. This clinic-to-the-bench and back-to-the-clinic approach could streamline and limit preclinical studies to those highly relevant to human disease, thereby minimizing disappointments and agony caused by failure to translate otherwise highly successful preclinical studies into successful treatments to the human diseases.

We hypothesize that using the new technological advances in molecular biology and genomics will be key to better understanding of human CD5+ B cells. Use of high-throughput sequencing of B cell receptors (BCR) of CD5+ B cells could reveal public BCRs associated with autoimmune diseases and use of single-cell RNA-seq analysis may delineate distinct sublineages.

Here we briefly discuss described subsets of CD5+ B cells, their association with human autoimmune diseases, and how use of new technologies could lead to better understanding of their composition, lineages, and role in disease pathogenesis.

Subsets of Human CD5+ B Cells

CD5+ B cells are heterogeneous and several subsets have been described to date. The most notable subset of CD5+ B cells is the IL-10-producing B-reg subset (referred to as B10 cells) that downregulates detrimental inflammatory responses through production of IL-10 (Saxena et al., 2014; Yanaba et al., 2008). However, many questions still remain regarding the origin of B10 cells and their mechanism of actions and whether the expression of CD5 is an indicator of specific lineage or merely a sign of activation. Another subset of CD5+ B cells is characterized by constitutive expression of the apoptosis-inducing Fas ligand (Lundy and Klinker, 2014; Saxena et al., 2017). This subset is expanded in type 1 diabetes (T1D) patients (Saxena et al., 2017) and appears to negatively regulate homeostasis of IL-10-producing B-reg cells in the NOD mouse model of T1D (Xiao et al., 2011). Many questions remain unanswered regarding the origin of FasL-expressing B cells and their relationship to conventional and pathophysiological roles. In this regard, Lundy et al. (2015) showed that experimental transformation of human B cells by a non-replicative variant of Epstein-Barr virus (EBV) consistently resulted in high expression of functional FasL protein, thereby providing another mechanism for the existence of FasL+ B cells. On the other hand, a prenaive B cell population that exists between transitional and naive B cells in human peripheral blood is reported to account for the most circulating CD5+ B cell population (Lee et al., 2009). Pre-naive CD5+ B cells display some of the functional properties of normal mature naive B cells. Because of this ability, it is proposed that pre-naive B cells pose the risk of causing autoimmunity when inappropriately stimulated (Lee et al., 2009). However, it is notable that since the original description of pre-naive B cells several years ago, there have been no follow-up studies that illuminate their role in host defense or autoimmunity settings. However, one study by the same group showed that production of IL-10 by pre-naive B cells limits their participation in normal immune responses, but they rather differentiate into autoantibody-secreting plasma cells in lupus patients (Sim et al., 2015). Human transitional T1 B cells, defined by high expression of CD24 and CD38, also express CD5 (Sims et al., 2005) and possess immunoregulatory properties to inhibit differentiation of CD4 T cells and suppress production of tumor necrosis factor (TNF-α) and IFN-γ by CD4 T cells in healthy, but not, lupus patients (Blair et al., 2010). However, CD5 is also upregulated on B cells by BCR stimulation (Youinou et al., 1999), making it difficult to determine whether CD5 expression denotes specific subsets, marks activated B cells, or does both. Use of single-cell RNA-seq will be useful in investigating heterogeneity of CD5+ B cells and their relationship to one another and their identity. On the other hand, analysis of their BCR repertoires using high-throughput immunosequencing will determine whether their expansion in autoimmune diseases is associated with specific clonal and autoantigens.

Association of CD5+ B Cells with Autoimmune Diseases

Different autoimmune diseases are associated with the expansion of CD5+ B cells. Evidence that CD5+ B cells expand in T1D has been described more than two decades ago (De Filippo et al., 1997; Munoz et al., 1995; Schatz et al., 1991; Smerdon et al., 1994). More recently, we linked expansion of CD5+ B cells in T1D subjects to a specific subset that is characterized by constitutive expression of FasL (Saxena et al., 2017). Interestingly, FasL and IL-10-producing subsets appear by flow cytometry to represent two distinct subsets (Hamad et al., 2012; Saxena et al., 2017). However, very little is known about their transcriptional profiles and relationship to one another and to other CD5+ B cells and conventional B cells. Use of deep sequencing and single cell analysis will help answer these questions. Rheumatoid factor-producing B cells are significantly enriched within CD5+ than in CD5− B cells isolated from the peripheral blood of RA patients (Youinou et al., 1987). There are also increased numbers of CD5+ B cells in spondylarthritis (Cantaert et al., 2012). CD5+ B cells have also been reported to be increased in myasthenia Gravis patients (Araga et al., 1995), and CD5+ B-reg cells have been implicated in suppressing experimental myasthenia gravis by an IL-10-dependent mechanism (Sheng et al., 2014; 2015). In addition, prenaive CD5+ B cells are found in increased proportions in peripheral blood of lupus patients (Lee et al., 2009). Selective expansion of CD5+ B cells in these autoimmune diseases provides identifiable subpopulations of disease-associated and presumably autoreactive cells that are ripe for high-throughput repertoire sequencing and single-cell RNA-seq to understand their sublineages and their roles in driving autoimmunity. If CD5+ B cells turn out to be drivers of autoimmunity, new strategies can be developed to spare normal productive B cells and target depletion to pathogenic subsets, thereby mitigating one of the most serious side effects of pan-B cell depletion immunotherapy.

CD5+ B Cells Are Major Sources of Natural Autoantibodies in Humans

Like murine CD5+ B1 cells, human CD5+ B cells produce polyreactive and autoreactive Abs (Casali and Notkins, 1989). These natural antibodies afford hosts rapid protection immediately following infections and prior to the development of adaptive immune responses. Natural antibodies also facilitate elimination of apoptotic cells thereby maintaining homeostasis and preventing inflammation. However, due to their polyreactivity and autoreactivity, natural antibodies can recognize self-antigens and promote autoimmune diseases in susceptible individuals. Studies in mice show that they can be positively selected by self-antigens (Hayakawa et al., 1999; Kasaian and Casali, 1993). Transgenic mice expressing a BCR specific for Thy-1 antigen produce specific antibodies only when mice express Thy-1 (Hayakawa et al., 1999). Given that most natural antibodies are considered non-pathogenic, the question of whether CD5+ B cells produce natural antibodies that promote autoimmune diseases in humans remains unanswered. In addition, specificities of BCRs of natural antibodies in humans are not well-understood. Applying new technologies to analyze BCR repertoires of CD5+ B cells could prove useful in elucidating whether CD5+ B cells harbor public BCRs associated with specific autoimmune diseases.

Use of Next-generation Technologies to Unlock Mysteries of CD5+ B Cells

Use of high-throughput sequencing and application of single-cell genomics technologies have revolutionized the study of immune cells. Both types of technology are highly appropriate and needed for breakthrough analysis of CD5+ B cells and their mysteries. Below, we briefly describe how these technologies can advance our understanding of human CD5+ B cells and their role in disease.

Use of high-throughput analysis of CD5+ B cells may lead to identification of elusive public BCRs associated with autoimmunity

High-throughput next-generation sequencing provides a powerful tool to profile TCR and BCR repertoires at the single-cell level (Sherwood et al., 2011). As noted above, CD5+ B cells expand in several autoimmune diseases. Yet there have been, to our knowledge, limited and rather no deliberate attempts to analyze immunoglobulin heavy chain variable regions (IGHV) repertoire of CD5+ B cells in the steady state or patients with major autoimmune diseases using high-throughput deep immunosequencing. Several properties support the ideas that IGHV repertoires of CD5+ B cells, rather than those of conventional B cells, may be enriched for disease-relevant public BCRs. Generation of highly diverse repertoires is critical for equipping the immune system with the ability to recognize and respond to various invading pathogens. The high-level diversity of the immune repertoire of conventional B cells has made it difficult to identify public BCRs that are clonally expanded in autoimmune settings Please cite Hamad et al (Discovery Medicine 2016). One reason could be limiting the analysis to conventional B cells and exclusion of CD5+ B cells. We argue that public BCRs could be particularly enriched within autoreactive CD5+ B cells. BCR repertoire of CD5+ B cells, unlike that of conventional B cells, is generally highly restricted (Hardy and Hayakawa, 2001; Sanz et al., 1989; Schutte et al., 1991). N-region addition and somatic hypermutation are highly restricted among CD5+ B cells (Gu et al., 1990; Hardy et al., 1989; Pennell et al., 1989). Consequently, natural antibodies derived from CD5+ B cells, unlike those derived from conventional B cells, tend to have germline-like sequences with broadly reactive and autoreactive repertoires (Forster et al., 1988; Hayakawa and Hardy, 1988; Hardy et al., 1989; Pennell et al., 1989; Gu et al., 1990). This leads to the focus of autoreactive receptors among CD5+ B cells and might explain their expansion in autoimmune diseases as a result of the recognition of self-antigens. Thus, in light of the close association of CD5+ B cells with autoreactivity and limited BCR repertoires, CD5+ B cells might be enriched for public BCRs that drive autoimmunity.

Use of single-cell RNA-seq could reveal composition of CD5+ B cells and their lineage relationship to CD5− B cells

Single-cell RNA-Seq has proven to be an effective method for comprehensive cellular decomposition of heterogeneous immune populations (Jaitin et al., 2014; 2015). CD5+ B cells are found at high frequencies in fetal cord blood and in serous cavities of healthy adult individuals (Berland and Wortis, 2002; Duan and Morel, 2006). In addition, they constitute the majority of pre-naive B cells (Lee et al., 2009). However, conventional B cells were also reported to upregulate CD5 expression upon BCR activation (Youinou et al., 1999). Thus, the relationship between CD5+ and CD5− B cells is unclear. Another unresolved question is focused on the identification of the human equivalent of the mouse B1 cells (Griffin et al., 2011a; 2011b). Resolution of these questions is of great importance because characterization of CD5+ B cells and understanding of their relationship to conventional CD5− B cells are prerequisites for understanding the overall role of B cells in host defense and autoimmune diseases.

Concluding Remarks

Recent genomics technological advances have allowed analysis of repertoires of T and B cells and their identities at the single-cell level, generating troves of information that are being mined for understanding roles of various immune cells in host defense, autoimmunity, and cancers. Consequently, different cell types and subtypes are being subjects of new analysis. Given that CD5+ B cells are some of the least understood subpopulations, we expect that there will be high rewards and great returns by subjecting their antigen receptors to high-throughput sequencing and analyzing their transcriptional profiles using single-cell RNA-seq. Such analysis is expected to be done under different platforms by different investigators with interests in different diseases. Generated information could solve problems that have been difficult to solve by subjecting the analysis of conventional B cells to high-throughput sequencing such as identifying public BCRs associated with different autoimmune diseases, which could be concentrated in expanded CD5+ B cells. Likewise, subjecting CD5+ B cells to single RNA-seq is expected to clarify their relationship to conventional B cells, their composition, and roles of their subsets in health and disease.

Acknowledgment

This work is supported by the NIH grant # 1R01AI099027-01A1 (to A.H.) and Norma Raab Foundation (AH and TD).

Footnotes

Disclosure

The authors declare no conflicts of interest.

References

  1. Araga S, Kishimoto M, Adachi A, Nakayasu H, Takenaka T, Takahashi K. The CD5+ B cells and myasthenia gravis. Autoimmunity 20(2):129–134, 1995. [DOI] [PubMed] [Google Scholar]
  2. Baumgarth N A hard(y) look at B-1 cell development and function. J Immunol 199(10):3387–3394, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Berland R, Wortis HH. Origins and functions of B-1 cells with notes on the role of CD5. Annu Rev Immunol 20:253–300, 2002. [DOI] [PubMed] [Google Scholar]
  4. Blair PA, Norena LY, Flores-Borja F, Rawlings DJ, Isenberg DA, Ehrenstein MR, Mauri C. CD19(+)CD24(hi)CD38(hi) B cells exhibit regulatory capacity in healthy individuals but are functionally impaired in systemic Lupus Erythematosus patients. Immunity 32(1): 129–140, 2010. [DOI] [PubMed] [Google Scholar]
  5. Cantaert T, Doorenspleet ME, Francosalinas G, Paramarta JE, Klarenbeek PL, Tiersma Y, Van Der Loos CM, De Vries N, Tak PP, Baeten DL. Increased numbers of CD5+ B lymphocytes with a regulatory phenotype in spondylarthritis. Arthritis Rheum 64(6): 1859–1868, 2012. [DOI] [PubMed] [Google Scholar]
  6. Casali P, Notkins AL. CD5+ B lymphocytes, polyreactive antibodies and the human B-cell repertoire. Immunol Today 10(11):364–368, 1989. [DOI] [PubMed] [Google Scholar]
  7. De Filippo G, Pozzi N, Cosentini E, Cavalcanti M, Carel JC, Tamasi S, Franzese A, Pignata C. Increased CD5+ CD19+ B lymphocytes at the onset of type 1 diabetes in children. Acta Diabetol 34(4):271–274, 1997. [DOI] [PubMed] [Google Scholar]
  8. Duan B, Morel L. Role of B-1a cells in autoimmunity. Autoimmun Rev 5(6):403–408, 2006. [DOI] [PubMed] [Google Scholar]
  9. Griffin DO, Holodick NE, Rothstein TL. Human B1 cells are CD3-: A reply to “A human equivalent of mouse B-1 cells?” and “The nature of circulating CD27+ CD43+ B cells”. J Exp Med 208(13):2566–2569, 2011a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Griffin DO, Holodick NE, Rothstein TL. Human B1 cells in umbilical cord and adult peripheral blood express the novel phenotype CD20+ CD27+ CD43+ CD70. J Exp Med 208(1):67–80, 2011b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gu H, Forster I, Rajewsky K. Sequence homologies, N sequence insertion and JH gene utilization in VHDJH joining: implications for the joining mechanism and the ontogenetic timing of Ly1 B cell and B-CLL progenitor generation. EMBO J 9(7):2133–2140, 1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hamad AR, Arcara K, Uddin S, Donner T. The potential of Fas ligand (apoptosis-inducing molecule) as an unconventional therapeutic target in type 1 diabetes. Front Immunol 3:196, 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hardy RR, Carmack CE, Shinton SA, Riblet RJ, Hayakawa K. A single VH gene is utilized predominantly in anti-BrMRBC hybridomas derived from purified Ly-1 B cells. Definition of the VH11 family. J Immunol 142(10):3643–3651, 1989. [PubMed] [Google Scholar]
  14. Hardy RR, Hayakawa K. B cell development pathways. Annu Rev Immunol 19:595–621, 2001. [DOI] [PubMed] [Google Scholar]
  15. Hayakawa K, Asano M, Shinton SA, Gui M, Allman D, Stewart CL, Silver J, Hardy RR. Positive selection of natural autoreactive B cells. Science 285(5424):113–n6, 1999. [DOI] [PubMed] [Google Scholar]
  16. Jaitin DA, Kenigsberg E, Keren-Shaul H, Elefant N, Paul F, Zaretsky I, Mildner A, Cohen N, Jung S, Tanay A, Amit I. Massively parallel single-cell RNA-seq for marker-free decomposition of tissues into cell types. Science 343(6172):776–779, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jaitin DA, Keren-Shaul H, Elefant N, Amit I. Each cell counts: Hematopoiesis and immunity research in the era of single cell genomics. Semin Immunol 27(1):67–71, 2015. [DOI] [PubMed] [Google Scholar]
  18. Kasaian MT, Casali P. Autoimmunity-prone B-1 (CD5 B) cells, natural antibodies and self recognition. Autoimmunity 15(4):315–329, 1993. [DOI] [PubMed] [Google Scholar]
  19. Lanier LL, Warner NL, Ledbetter JA, Herzenberg LA. Expression of Lyt-1 antigen on certain murine B cell lymphomas. J Exp Med 153(4):998–1003, 1981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lee J, Kuchen S, Fischer R, Chang S, Lipsky PE. Identification and characterization of a human CD5+ prenaive B cell population. J Immunol 182(7):4116–4126, 2009. [DOI] [PubMed] [Google Scholar]
  21. Lundy SK, Klinker MW. Characterization and activity of Fas ligand producing CD5(+) B cells. Methods Mol Biol 1190:81–102, 2014. [DOI] [PubMed] [Google Scholar]
  22. Lundy SK, Klinker MW, Fox DA. Killer B lymphocytes and their fas ligand positive exosomes as inducers of immune tolerance. Front Immunol 6:122, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Munoz A, Gallart T, Usac EF, Fernandez-Alvarez J, Vinas O, Somoza N, Barcelo J, Gomis R. Anti-islet cell and anti-insulin antibody production by CD5+ and CD5− B lymphocytes in IDDM. Diabetologia 38(1):62–72, 1995. [DOI] [PubMed] [Google Scholar]
  24. Pennell CA, Sheehan KM, Brodeur PH, Clarke SH. Organization and expression of VH gene families preferentially expressed by Ly-1+ (CD5) B cells. Eur J Immunol 19(11):2115–2121, 1989. [DOI] [PubMed] [Google Scholar]
  25. Sanz I, Casali P, Thomas JW, Notkins AL, Capra JD. Nucleotide sequences of eight human natural autoantibody VH regions reveal apparent restricted use of VH families. J Immunol 142(11):4054–4061, 1989. [PubMed] [Google Scholar]
  26. Saxena A, Khosraviani S, Noel S, Mohan D, Donner T, Hamad AR. Interleukin-10 paradox: A potent immunoregulatory cytokine that has been difficult to harness for immunotherapy. Cytokine, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Saxena A, Yagita H, Donner TW, Hamad ARA. Expansion of FasL-expressing CD5+ B cells in type 1 diabetes patients. Front Immunol 8:402, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Schatz DA, Lang F, Cantor AB, Riley WJ, Maclaren NK, Sleasman JW, Barrett DJ. CD5+ B lymphocytes in high-risk islet cell antibody-positive and newly diagnosed IDDM subjects. Diabetes 40(10):1314–1318, 1991. [DOI] [PubMed] [Google Scholar]
  29. Schutte ME, Ebeling SB, Akkermans KE, Gmelig-Meyling FH, Logtenberg T. Antibody specificity and immunoglobulin VH gene utilization of human monoclonal CD5+ B cell lines. Eur J Immunol 21(5):1115–1121, 1991. [DOI] [PubMed] [Google Scholar]
  30. Sheng JR, Quan S, Soliven B. CD1d(hi)CD5+ B cells expanded by GM-CSF in vivo suppress experimental autoimmune myasthenia gravis. J Immunol 193(6):2669–2677, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sheng JR, Quan S, Soliven B. IL-10 derived from CD1dhiCD5(+) B cells regulates experimental autoimmune myasthenia gravis. J Neuroimmunol 289:130–138, 2015. [DOI] [PubMed] [Google Scholar]
  32. Sherwood AM, Desmarais C, Livingston RJ, Andriesen J, Haussler M, Carlson CS, Robins H. Deep sequencing of the human TCRgamma and TCRbeta repertoires suggests that TCRbeta rearranges after alpha-beta and gamma-delta T cell commitment. Sci Transl Med 3(90):90ra61, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sim JH, Kim HR, Chang SH, Kim IJ, Lipsky PE, Lee J. Autoregulatory function of interleukin-10-producing prenaive B cells is defective in systemic lupus erythematosus. Arthritis Res Ther 17:190, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Sims GP, Ettinger R, Shirota Y, Yarboro CH, Illei GG, Lipsky PE. Identification and characterization of circulating human transitional B cells. Blood 105(11):4390–4398, 2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Smerdon RA, Peakman M, Hussain MJ, Wong FS, Watkins PJ, Leslie RD, Vergani D. CD5+ B-cells at the onset of type I diabetes and in the prediabetic period. Diabetes Care 17(7):657–664, 1994. [DOI] [PubMed] [Google Scholar]
  36. Xiao Z, Mohamood AS, Uddin S, Gutfreund R, Nakata C, Marshall A, Kimura H, Caturegli P, Womer KL, Huang Y, Jie C, Chakravarti S, Schneck JP, Yagita H, Hamad AR. Inhibition of Fas ligand in NOD mice unmasks a protective role for IL-10 against insulitis development. Am J Pathol 179(2):725–732, 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Yanaba K, Bouaziz JD, Haas KM, Poe JC, Fujimoto M, Tedder TF. A regulatory B cell subset with a unique CD1dhiCD5+ phenotype controls T cell-dependent inflammatory responses. Immunity 28(5):639–650, 2008. [DOI] [PubMed] [Google Scholar]
  38. Youinou P, Jamin C, Lydyard PM. CD5 expression in human B-cell populations. Immunol Today 20(7):312–316, 1999. [DOI] [PubMed] [Google Scholar]
  39. Youinou P, Mackenzie L, Jouquan J, Le Goff P, Lydyard PM. CD5 positive B cells in patients with rheumatoid arthritis: phorbol ester mediated enhancement of detection. Ann Rheum Dis 46(1):17–22, 1987. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES