This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Background. Endemic Burkitt's lymphoma (eBL) is a B cell neoplasm that is thought to arise from arrested maturation of germinal center B cells. However, information is currently lacking on the peripheral B cell compartment and whether changes in B cell subsets mirror elevated Epstein-Barr virus (EBV) loads in eBL patients. Methods. We used flow cytometry to analyze peripheral B cell subset distribution and real-time quantitative polymerase chain reaction (Q-PCR) to analyze EBV viral loads in 32 children presenting with eBL and 25 age-matched controls from western Kenya. Results. No cells with a BL tumor phenotype (CD19+CD10+CD77+CD38+) were found in the peripheral circulation of either eBL patients or controls. The frequencies of naive (CD19+IgD+CD27-) and classical memory (CD19+IgD-CD27+) B cells were comparable in the two groups, while there were significantly higher frequencies of B cell receptor (BCR)-deficient (CD19+κ-λ-) and CD19+CD27-CD5-CD10- memory B cells in eBL patients as relative to controls (p=0.0038 and p=0.0001, respectively). Importantly, EBV viral load was more elevated in eBL patients as compared to controls (p<0.0001), and was positively correlated with the frequency of CD19+CD27-CD5-CD10- B cells (r=0.4952, p=0.0191). Conclusions. Taken together, these data suggest that there is profound perturbation of B cell homeostasis in eBL patients with concomitant elevation of EBV viral loads being associated with CD19+CD27-CD5-CD10- B cells, and not with peripheral circulation of tumor cells.
Endemic Burkitt's lymphoma (eBL) is a high grade B cell lymphoma, and is the most common pediatric malignancy in malaria-endemic regions of sub-Saharan Africa and Papua New Guinea . Epstein Barr virus (EBV) can be found in up to 95% of eBL tumors, implicating the virus as critical in the etiology of this cancer [2,3]. In addition, there is a profound dysregulation of EBV persistence and immunity in children [4-6] in malaria-endemic areas, suggesting that alterations in EBV persistence due to repeated malaria infection precede the emergence of a malignant clone.
Germinal center (GC) reactions play an important role in the generation of effective humoral immune responses . Somatic hypermutation and class-switch recombination are crucial steps in germinal center reactions for generation of functional B cell receptors (BCR), which, in turn, are important in the survival of B cells in peripheral circulation [8,9]. Ideally, all B cells that fail to develop functional BCR through acquisition of deleterious mutations during somatic hypermutation are effectively removed through apoptosis [9,10]. However, there are studies demonstrating that EBV can rescue BCR-deficient B cells from apoptosis [11-14]. Defects in BCR expression in mouse models have been implicated in arrested maturation of B cells and in B cell lymphomagenesis [15,16]. Interestingly, several studies have reported that BL tumor cells may be derived from BCR-deficient germinal center B cells, with the EBV viral proteins, latent membrane protein (LMP)-1 and LMP-2a playing critical roles in the survival of these pre-malignant B cell subsets [13,14,17]. BL tumors express classic markers of germinal center B cells, such as CD10, CD38 and CD77 [16-18] and molecular profiling indicates that many genes expressed in germinal centers are also expressed in BL tumors [7,19]. However, the cell type is controversial and some have argued that they have post-germinal center memory-cell origin .
We and others have found elevated EBV viral loads in DNA extracted from whole blood of eBL patients [4,21,22], but it is unknown whether this elevated viral load is due to presence of a leukemic phase of the disease or simply release of EBV viral DNA into circulation from necrotic tumor cells. In healthy EBV sero-positive individuals, EBV is strictly cell-associated and found in the CD19+IgD-CD5-CD27+ memory B cell compartment [12,23,24], but in patients with EBV-associated post-transplant lympho-proliferative disease, chronically elevated viral loads are indicative of an increase in circulation of immunoglobulin (Ig)-null B cells that only express CD19 and no other classic B cell markers such as CD10 [11,12,25]. In this study, we examined the peripheral B cell phenotype and EBV viral load in eBL patients and controls from a malaria-endemic region with high incidence of eBL. Our results revealed that the EBV viral load does not correlate with an increase in circulating BL tumor cells, but rather with elevation of CD19+CD5-CD27-CD10- B cell population.
Information about the study population and design has been reported elsewhere . Briefly, following informed consent by the study participant's parents/guardians, we enrolled 34 children presenting with eBL at the Nyanza Provincial General Hospital (NPGH), Kisumu, Kenya. Two of these patients were excluded from further analysis; one patient had acute leukemia and another was diagnosed to be HIV+. We also enrolled 25 age-matched controls (herein referred to as 'controls') from Kanyawegi village along the shores of Lake Victoria in Kisumu West District, Nyanza Province. Kanyawegi is a rural district that has a high incidence of eBL and experiences persistent P. falciparum transmission . This protocol was approved by the Kenya Medical Research Institute (KEMRI) Ethical Review Committee, the Institutional Review Board for Human Studies at the University Hospitals of Cleveland and Case Western Reserve University (Dr. Moormann's institutional affiliation at the time of sample collection) and at SUNY Upstate Medical University, USA.
Blood collection and processing
Finger-prick blood was collected in EDTA tubes and measurements of hemoglobin (Hb) levels were determined using a portable B-hemoglobin photometer (Hemocue AB Angelholm, Sweden). Complete blood counts were performed with a Beckman coulter AcT diff2 (Beckman-coulter corporations, Miami, FL, USA) on finger-prick blood, collected in the EDTA-containing vials. An aliquot of finger-prick blood was stored at -80oC for subsequent isolation of DNA. In addition, 2-5 ml of venous blood was drawn from both eBL patients and controls. Ficoll-hypaque density gradient centrifugation of blood was carried out within 1 hour of blood collection. Peripheral blood mononuclear cells (PBMC) were immediately analyzed for B cell phenotype by flow cytometry.
Preparation of blood and flow cytometric analysis
Cell viability was determined using the Trypan Blue dye exclusion assay and all cells were routinely greater than 98% viable. Following PBMCs isolation, immunofluorescence labeling was done as previously described . The cell surface staining was done using the following antibodies to lymphocyte surface receptors; CD45-APC, CD19-PerCP, CD3-APC, CD8-PE, CD4-FITC, CD3-FITC, CD27-FITC, CD10-PE, CD5-APC, Kappa (κ)-FITC, Lambda (λ)-PE, CD10-APC, CD77-FITC and IgD-PE. Isotypic controls were IgG2a, K-FITC (mouse), IgG1, K-PE (mouse), IgG1, K-PerCP (mouse) and IgG1, K-APC (mouse) (BD Pharmigen, San Diego, CA, USA). Data was collected using a FACsCalibur flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, USA) and analysed using FlowJo software (Tree Star Inc. San Carlos, CA, USA).
Human Immunoglobulin Free Light Chains κ and λ ELISA
Levels of serum free κ and λ light chains assays were determined using Human Immunoglobulin Free Light Chains κ and λ ELISA kit (BioVendor Laboratorni Medicina, Modrice, Czech Republic), according to the manufacturers' protocol.
Quantitative PCR to quantify EBV DNA
DNA was extracted from 200 l of finger-prick whole blood using Qiagen DNAeasy kit (Qiagen, Valencia, CA, USA) according to the manufacturers' protocol. DNA was eluted from the column in 200 l of H20 and stored at -20oC. Primers and probes to detect a 70 bp region of the EBV BALF5 gene and the ¢-actin gene were previously described . The viral load was normalized to the ¢-actin copy number, log-transformed and then calculated based on copies of EBV genome/µg of whole blood. The EBV viral load was reported as log EBV copies/µg DNA.
Differences in non-parametric values between two defined groups were determined using Mann-Whitney U test. Correlations were examined using Spearman's rank correlation test. A p<0.05 was considered statistically significant. Data was analyzed using GraphPad prism version 5 (GraphPad Software, Inc, La Jolla, CA, USA).
General characteristics of the study population
We enrolled 32 eBL patients (21 males and 11 females) and 25 healthy age-matched controls (16 males and 9 females). The demographic, parasitological and hematological characteristics of the study populations are summarized in Table 1. Age (p=0.7584) and distribution of gender (p=0.8822) were comparable between the two groups. The proportions of children who were parastemic was significantly higher in controls relative to eBL patients (p<0.0001). However, most children presenting with eBL at NPGH have been referred from another hospital, and in most cases it was reported that these patients have been treated with anti-malarial drugs prior to admission at NPGH. Leukocyte counts were comparable in eBL patients and controls, as there were no significant differences in white blood cell counts (WBC) (p=0.1356), lymphocyte counts (LY) (p=0.3220), monocytes counts (MO) (p=0.2458) and granulocyte counts (GR) (p=0.0569). Red blood cell counts (RBC) were also comparable (p=0.4661). In contrast, Hb levels were significantly lower in the eBL patients compared to controls (p=0.0009) and a higher frequency of eBL patients were anemic compared to controls (p<0.0001). Lactate dehydrogenase (LDH) levels, which were used as a clinical marker of tumor burden were significantly higher in eBL patients compared to controls (p<0.0001).
Elevated EBV viral loads in eBL patients are not associated with circulating tumor cells in peripheral blood
Elevated EBV viral loads in BL patients are thought to result from increased circulation of BL tumor cells in peripheral circulation . To test this possibility, we quantified EBV viral loads from whole blood using Q-PCR. Consistent with earlier observations [4,22], we found significantly higher EBV viral loads in eBL patients compared to controls (median 4.14 log EBV copies/µg DNA [range 0.00-4.11] versus 0.90 log EBV copies/ µg DNA [range 0.00-5.23], p<0.0001) (Fig. 1).
To determine if the elevated EBV viral loads observed in peripheral blood from eBL patients were due to the presence of circulating tumor cells, we analyzed peripheral B cells for evidence of tumors cells. PBMC were stained with a panel of mAbs specific for CD19, CD10, CD38 and CD77. The CD19 is a pan-B cell marker found on both normal and malignant B cells. The CD10, CD38 and CD77 are expressed by BL cells [19,27] such that if there were circulating BL cells, we would anticipate high proportions of cells expressing these markers. As these markers are also found in normal peripheral B cells albeit at low levels, we also analyzed the peripheral B cell phenotype of the controls. Results demonstrated that there were no cells with atypical BL phenotype (CD19+CD10+CD38+CD77+) circulating in peripheral blood of either eBL patients or controls (Fig A) or eBL patients (Fig. 2B). It is possible that EBV down-regulation of the CD10 and CD38 germinal center B cell markers, also used as BL tumor cell marker [11,16,17], can lead to lack of detection of these cell types using flow cytometry. To check for the presence of these cell types in peripheral circulation, we analyzed PBMCs for surface expression of these markers. As shown in Table 2 and Fig.2 C and D, respectively, we were able to detect (CD19+CD38+CD10+77-) B cells expressing these germinal center markers in peripheral circulation, although at a lower frequency in eBL patients compared to controls (median 8.37% [range 1.81-18.24] versus 21.93% [13.65-28.78]; p=0.0005), suggesting that down-regulation or loss of BL tumor markers may not explain the lack of detection of BL tumor cells in peripheral circulation.
As a second marker, to check for clonality of circulating cells indicative of tumor cells, we also tested PBMC for the surface expression of the kappa («) or lambda (¬) chain on B cells. For the presence of monoclonal cell populations, we would observe skewing of distribution of surface immunoglobulin light chain towards either κ or λ light chains. Typically, in healthy adult blood, approximately 46.0% of B cells are «-positive and 39.0% are ¬-positive . In accordance with this, in the samples of control study population, we observed a median frequency of 47.7% of B cells were κ-positive, 32.3% were λ-positive while 2.48% expressed both «- and ¬-light chains. In addition, 17.76% neither expressed «- or ¬- light chains (Fig.3 A and C). In contrast, in eBL patients, there was a median frequency of 4.36% of B cells expressing «-positive, 17.32% were ¬-positive while 1.01% expressed both «- and ¬-light chains. A higher proportion 73.9% neither expressed «- or ¬-light chains (BCR-deficient) (Fig. 3 B and D)€®€ € Remarkably, eBL patients had significantly higher frequency of CD19+κ-λ- (BCR-deficient) B cells relative to controls (median 73.39% [range 11.18-95.09] versus 17.76% [range 10.82-19.76]; p=0.0038), while the controls had normal distribution of «- or ¬-light chains (Table 2). These data suggest that the elevated EBV viral loads reported in eBL patients are not associated with circulating BL tumor cells.
Endemic Burkitt's lymphoma patients have normal levels of serum free κ and λ light chains in their plasma
We next explored whether low frequencies of B cells expressing a functional B cell receptor (BCR) on B cells was due to cytogenic aberrancies that impair heavy and light chains pairing, or as a result of down-regulation of heavy chain but not light chain transcription . Therefore, we assayed for serum free κ- and λ- light chains in plasma of both eBL patients and controls. There was no significant difference in the levels of serum free κ light chains in controls relative to eBL patients (median 13.03mg/L [range 5.31- 28.83] versus 9.11 mg/L [range 1.23-40.07]; p=0.0678) (Fig.3E). In addition, the levels of serum free λ lights were also comparable between controls and eBL patients (median 9.73mg/L [range 6.48-54.29] versus 11.16mg/L [3.95-45.06]; p=0.5625) (Fig.3F). These levels were within the normal ranges for detection of serum free κ light chains (normal range, 3.3-19 mg/L) and λ light chains (normal range, 5.7-26.3 mg/L) in healthy individuals . These findings suggest that lack of expression of functional BCR on these B cell subsets is not due to down-regulation of the transcription of heavy chain genes.
Endemic Burkitt's lymphoma patients have elevated frequencies of atypical (CD19+IgD-CD27-) memory B cells and decreased non-class switched (CD19+IgD+CD27+) memory B cells
We further determined the frequencies of naive and memory B cell populations based on flow cytometric staining with IgD and CD27. As shown in Table 2, frequencies of the total B cell population (CD19+), naive (CD19+IgD+CD27-) and classical memory (CD19+IgD-CD27+) B cell populations were comparable in eBL patients and controls. Further analysis revealed significantly higher frequencies of mature naive B cells (CD19+IgD+CD10-) in eBL patients relative to controls (median 60.75% [range 50.59-68.48] versus 54.06 [47.55-59.38]; p=0.0346). However, the frequency of immature transitional B cells (CD19+IgD+CD10+) was significantly higher in controls relative to eBL patients (median 21.42% [range 15.83-27.72] versus 11.86% [range 4.28-18.33]; p=0.0023).
Within the memory B cell compartment, the frequency of non-class switched memory (CD19+IgD+CD27+) B cells, important in bacterial or viral defense , was significantly higher in controls as compared with eBL patients (median 13.66% [ 12.84-16.50] versus 7.90% [range 6.13-12.46]; p<0.0001). However, the frequency of atypical (CD19+IgD-CD27-) memory B cells was significantly higher in eBL patients as compared to controls (median 15.63% [range 10.23-22.20] versus 10.77% [range 6.99-14.29]; p=0.0304). These data suggest that eBL patients have perturbed memory B cell homeostasis.
Following the observations that the memory B cell pool is perturbed in eBL patients, we hypothesized that there would be perturbations in the frequency of memory B cells subsets which are associated with control of blood-borne pathogens . To test this hypothesis, we carried out a flow cytometric analysis based on the expression of CD19, CD27, CD10 and CD5. As shown in Table 2, there was no statistical difference in the frequency of CD19+CD27+CD5- between the controls and eBL patients, however, the frequency of CD19+CD27+CD5+ memory B cell subsets was significantly higher in the controls relative to eBL patients (median 11.42% [range 8.64-14.51] versus 6.91% [range 3.34-10.51]; p=0.0058). Likewise, the frequency of CD19+CD27-CD5+ memory B cells was also significantly higher in controls relative to the eBL patients (median 33.40% [28.85-38.23] versus 21.90% [range 14.97-29.06]; p<0.001).
Further characterization of CD5+ memory B cells based on co-expression of CD10 revealed that the frequency of CD19+CD5+CD10- B cells were significantly higher in controls relative to the eBL patients (median 26.70% [range 22.80-30.15] versus 22.40% [range 17.65-27.91]; p=0.0486). In contrast, the frequency of CD19+CD27-CD5-CD10- memory B cells were significantly higher in eBL patients as compared to the controls (median 79.74% [range 68.92-85.45] versus 64.67% [range56.01-69.74]; p<0.0001).
Correlation between EBV viral loads and B cell sub-populations
Elevated EBV viral loads have been associated with immunoglobulin (Ig)-null B cells in other EBV-associated conditions [11,25]. However, contrary to this previous observation, we did not observe any association between BCR-negative B cells and EBV viral loads (r=-0.2332, p=0.2842), suggesting that the elevated EBV viral loads reported in eBL patients may not be restricted to EBV persistence in BCR-negative B cells alone.
Many studies have shown that EBV is restricted to the CD5- B cell memory B cell subsets in peripheral blood [11,12,24,31]. Therefore, we investigated whether there is an association between EBV viral loads and frequencies of different B cell subsets. As shown, EBV viral loads were negatively correlated with frequencies of IgD+CD27+ (r=-0.5880, p=0.0064) (Fig.4A), CD27+CD5+ (r=-0.5948, p=0.0045) (Fig.4B) and CD27+CD10- (r=-0.4376, p=0.0417) (Fig.4C) and positively correlated with the frequency of CD27-CD5-CD10-(r=0.4952, p=0.0191) (Fig.4D). These results illustrate that elevated EBV viral loads in eBL patients are associated with higher frequencies of CD19+27-CD5-CD10- B cells.
Endemic Burkitt's lymphoma is a highly aggressive B cell lymphoma thought to arise from arrested maturation of germinal center B cells [7,16,19]. Previous observations raised the possibility that elevated EBV viral loads reported in eBL patients may be due to increased circulation of BL tumor cells in peripheral circulation . In addition, recent studies suggest that BL tumor cells may be derived from BCR-deficient B cell precursors [13,14,17]. To test these possibilities, we analyzed B cell phenotypes and EBV viral loads in eBL patients and in age-matched controls. Our results demonstrate that although there were striking differences in peripheral blood B cell subsets and EBV viral loads in eBL patients compared to controls, BL tumors cells do not circulate in peripheral blood of eBL patients. These findings also provide a direct demonstration of increased frequency of BCR-deficient B cells in peripheral blood of eBL patients and controls, further suggesting that there is profound perturbation of B cell homeostasis in eBL patients and children from malaria-endemic regions.
While B cell receptors are a prerequisite for the generation and survival of mature B cells and in determining the size or the frequency of lymphoid compartments during immune responses [15,19], the presence of BCR-deficient B cells in the peripheral blood of patients with EBV-associated disorders have been reported [11,12]. Our data further confirm these previous findings, and extend those observations by demonstrating the presence of BCR-deficient B cells in the peripheral blood of eBL patients. This population represents apoptosis-prone B cell subsets that are normally eliminated through negative selection during germinal center reactions [9,17]. Although the exact mechanisms that lead to the expansion of these B cell subsets in peripheral circulation of eBL patients are still not clearly understood, both EBV infection and/or tumor-associated immunosuppression are both factors that may be interfering with germinal center or peripheral negative selection check-points, resulting in peripheral circulation of Ig-null B cells [13,14,17]. However, our data show that elevated EBV viral load was not associated with the frequency of BCR-deficient B cell subset.
We also report observation of BCR-deficient B cells in the peripheral circulation of children with asymptomatic P. falciparum malaria infection, suggesting that children from malaria-endemic regions also experience aberrancies in B cell development that may predispose them to develop eBL [26,32]. Although the exact mechanisms that lead to accumulation of these subsets in peripheral blood of these children is unclear, one possibility is that continuous antigenic stimulation due to chronic infection with P. falciparum may interfere with peripheral or germinal center negative selection check-points, allowing the release of BCR-deficient B cells into peripheral blood [8,11]. Alternatively, the long-term effects of EBV viral reactivation, common in malaria-endemic settings , may lead to increased release of EBV virions that infect immature BCR-negative B cells and rescue them from apoptosis, thus leading to development of malignant B cell clones. In fact, elevated frequencies of immature B cells have been reported in children from malaria-endemic settings . This subset is not only susceptible to EBV infection as compared to memory B cells, but also expresses elevated levels of microRNA (hsa-miR-127), which is important in blocking B cell differentiation [16,33]. Furthermore, stimulation of this sub-population with unmethylated bacterial DNA (CPG) through Toll-like receptor 9 (TLR-9) leads to constitutive expression of activation-induced deaminase (AID) genes, critical in potentiating c-myc translocation important in eBL lymphomagenesis [19,33]. P. falciparum also generates a ligand for TLR-9 important in recognition of unmethylated double-stranded DNA-hemozoin complex and this may drive expansion of latently infected B cells, thus predisposing individuals from malaria-endemic settings to develop lymphomas .
Although, we report a positive correlation between EBV viral loads with frequencies of CD19+27-CD5-CD10- B cells in eBL patients, a recent study in HIV+ patients demonstrated association of EBV viral load with expansion of immature transitional B cells . Moreover, EBV infection of natural killer cells, monocytes and T-cells is common feature in patients suffering from EBV lympho-proliferative disorders . Together, these data suggest that in patients with elevated EBV viral load, EBV infection is not restricted to memory B cells. Hence, further studies are needed to elucidate the cellular origin of elevated EBV viral loads reported in eBL patients.
Previous studies have shown that expansion of CD19+IgD-CD27- memory B cells is associated with impaired immune responses during pathogenic infections and in autoimmune diseases . Recently, studies in both HIV and P. falciparum infections have associated expansion of this subset with immune hypo-responsiveness [32,37]. We also report expansion of this subset in eBL patients as compared to controls, with frequencies in eBL patients comparable to those observed in autoimmune diseases and bacterial infections . In addition, consistent with previous reports in HIV and autoimmune patients [31,36], we report reduced frequency of non-class-switched memory B cells in eBL patients, further suggesting that chronic immune activation may lead to B cell dysfunctions in eBL patients. However, the effects of these perturbations on functional immunity in eBL patients were beyond the scope of the current study.
In conclusion, we show that there are major perturbations in peripheral B cell homeostasis in eBL patients, with our results supporting the notion that malignant B cell clones in eBL tumors may be derived from BCR-deficient B cells. In addition, elevated EBV viral loads in eBL patients are associated with CD19+CD27-CD5-CD10- B cell sub-populations but not due to increment of BL tumor cells in peripheral circulation. Hence, future studies on eBL lymphomagenesis should focus on the molecular and phenotypic markers of the BCR-deficient B cell subsets described in the current study, in addition to identifying the niche of EBV. This will provide critical insight into eBL pathogenesis and also help in refining therapeutic approaches.