- Open Access
Altered Ca2+ homeostasis in polymorphonuclear leukocytes from chronic myeloid leukaemia patients
Molecular Cancervolume 5, Article number: 65 (2006)
In polymorphonuclear leukocytes (PMNL), mobilization of calcium ions is one of the early events triggered by binding of chemoattractant to its receptors. Besides chemotaxis, a variety of other functional responses are dependent on calcium ion mobilization. PMNL from chronic myeloid leukaemia (CML) patients that were morphologically indistinguishable from normal PMNL were found to be defective in various functions stimulated by a chemoattractant – fMLP. To study the mechanism underlying defective functions in CML PMNL, we studied calcium mobilization in CML PMNL in response to two different classical chemoattractants, fMLP and C5a.
Release of calcium estimated by flow cytometry and spectrofluorimetry using fluo-3 as an indicator showed that the [Ca2+]i levels were lower in CML PMNL as compared to those in normal PMNL. But, both normal and CML PMNL showed maximum [Ca2+]i in response to fMLP and C5a at 10 sec and 30 sec, respectively. Spectrofluorimetric analysis of the total calcium release in chemoattractant treated PMNL indicated more and faster efflux of [Ca2+]i in CML PMNL as compared to normal PMNL.
Fine-tuning of Ca2+ homeostasis was altered in CML PMNL. The altered Ca2+ homeostasis may contribute to the defective functions of CML PMNL.
In polymorphonuclear leukocytes (PMNL), changes in intracellular calcium, i.e. [Ca2+]i are associated with multiple cellular events, including activation of cellular kinases and phosphatases, degranulation, phagosome-lysosome fusion, regulation of cytoskeleton binding proteins, transcriptional control and modulation of surface receptors . Migration of leukocytes through the extracellular matrix to the site of action is the first step in host defence and role of calcium in this process is well reviewed by Maxfield . Although no stable [Ca2+]i gradients were detected in migrating human PMNL, a transient global increase in [Ca2+]i was found to be important for chemotaxis . PMNL migration can be induced by binding of chemoattractants to their receptors present on PMNL surface. The classical chemoattractants for PMNL are n-formyl peptides that are analogous to bacterial secretion  and anaphylatoxin C5a, which is formed upon complement activation . Specific receptors for n-formyl peptides and C5a are present on PMNL and they share common structural motifs . Mobilization of [Ca2+]i is one of the early events triggered by binding of a chemoattractant to its receptor.
Chronic myeloid leukaemia (CML) is a clonal, pluoripotent stem cell disorder characterized by the occurrence of Philadelphia chromosome (Ph1) and presence of a large number of mature and immature myeloid cells in the peripheral blood . Earlier work from our laboratory has shown that PMNL from CML patients were defective in actin dependent functions such as chemotaxis, degranulation, endocytosis, etc. [8–12]. Chemotaxis was found to be defective in all the phases of the disease . Calcium plays a central role in these functions. Calcium regulates cell motility by regulating polymerization of actin – one of the major motile machinery proteins in PMNL. Increased [Ca2+]i levels lead to fragmentation of actin network by disrupting the cross bridges of actin network. Increased [Ca2+]i levels cause fragmentation of F-actin by activation of actin severing and capping proteins such as gelsolin and macrophage capping protein . Fibroblasts transfected with gelsolin, a calcium activated actin severing and capping protein, display increased motility . In view of the role of calcium in various motility related events, the present studies are aimed to study mobilization of Ca2+ in CML PMNL. Mobilization of Ca2+ by fMLP and C5a was studied in these cells. PMNL from healthy normal individuals were used as control. We found that fine-tuning of Ca2+ homeostasis in CML PMNL was altered as compared to that in normal PMNL.
Measurement of [Ca2+]i by flow cytometry
Basal [Ca2+]i levels in PMNL
Fluo-3 loaded normal PMNL showed a broad bell shaped plot, indicating considerable variation in the basal [Ca2+]i levels of the normal PMNL population (Fig. 1). Fluo-3 loaded unstimulated CML PMNL showed a broad plot with a long tail near Y-axis, containing 9–10% of the population. Thus, 10% of the population had very low levels of [Ca2+]i. Comparison between the two populations showed that basal levels of [Ca2+]i in CML PMNL were lower and showed more variation. However, this difference was not statistically significant (Table 1).
Stimulation with fMLP
In fMLP stimulated normal PMNL, plots were negatively skewed and broader. Hence the heterogeneity in the PMNL population with respect to [Ca2+]i levels increased. On stimulation, the peaks shifted to right (Fig. 1A; c, d, e) leading to a significant increase in [Ca2+]i levels at 10 sec, 30 sec, and 60 sec (Table 1 and Fig. 2), maximum increase in [Ca2+]i being at 10 sec. On further treatment of these cells with calcium ionophore A23187, [Ca2+]i levels increased significantly (Fig. 1A; f, Table 1), but heterogeneity in PMNL with respect to [Ca2+]i decreased. Addition of EGTA resulted in building up of a small peak towards extreme left on the X-axis indicating that only 8–10% of the total population, was sensitive to EGTA (Fig. 1A; g). But the resulting decrease in total [Ca2+]i levels was considerable. On further addition of MnCl2, [Ca2+]i levels decreased significantly and were below the basal levels (Fig. 1A; h, Table 1). The PMNL population was extremely heterogeneous as far as sensitivity to quenching of calcium by MnCl2 was considered.
To compare the extent of stimulation in PMNL, the ratio of [Ca2+]i levels before and after fMLP stimulation were calculated. In normal PMNL, 10 sec of fMLP treatment resulted in 1.25 to 4.4 times (mean ± SEM = 1.6 ± 0.21) increase in [Ca2+]i over the basal levels. Further treatment with ionophore increased the mean ratio to 2.5 ± 0.36. Subsequent addition of EGTA and MnCl2 decreased [Ca2+]i levels to 21% and 64% of the maximum levels, respectively; ultimately quenching down the fluorescence to about 30% lower than the basal levels (Fig. 3).
CML PMNL showed narrowing of the PMNL peak after fMLP stimulation, indicating decrease in heterogeneity in the population with respect to [Ca2+]i levels (Fig. 1C; b, c, d, e). At all the time points studied after fMLP stimulation, levels of [Ca2+]i were significantly higher than the basal [Ca2+]i levels (Table 1, Fig. 2). The peak levels were seen at 10 sec. On further treatment of these cells with calcium ionophore A23187, the [Ca2+]i levels increased further, but shape of the peak was unaltered (Fig. 1C; f). On addition of EGTA, the total [Ca2+]i content in the population was not altered considerably (Fig. 1C; g). On further addition of MnCl2, the peak width increased, the peak shifted towards Y-axis and an additional very small peak appeared adjacent to Y-axis (Fig. 1C; h). Thus, the PMNL population was extremely heterogeneous as far as sensitivity to quenching of calcium by MnCl2 was considered. The ratio of maximum fluorescence in fMLP stimulated cells ranged from 1.42 to 5.62. On addition of ionophore the mean ratio increased to 2.6. Addition of EGTA led to 10% decrease in the fluorescence intensity where as MnCl2 resulted in a statistically significant 60% decrease as compared to the maximum fluorescence intensity. Thus, EGTA and MnCl2 together brought down the fluorescence to the basal level (Fig. 3).
Both, normal and CML PMNL, showed significantly higher [Ca2+]i levels on fMLP stimulation as compared to that in the respective unstimulated PMNL(Fig. 1A and 1C). But on fMLP stimulation, the heterogeneity with respect to [Ca2+]i levels was lower in CML PMNL than in normal PMNL. The [Ca2+]i levels both before and after fMLP stimulation were lower in CML than in normal PMNL. However, this difference was not statistically significant. In fMLP stimulated CML PMNL, the drop in levels of [Ca2+]i at 60 sec was rapid and more as compared to that in normal PMNL. This was evident from the higher ratio over the basal level at 60 sec in normal PMNL as compared to that in CML PMNL. On the addition of calcium ionophore A23187, fMLP stimulated CML PMNL showed lower levels of [Ca2+]i as compared to normal PMNL. But the extent of [Ca2+]i mobilization was higher in CML as reflected in the ratios (Fig. 3). Though the extent of [Ca2+]i stimulation with fMLP and ionophore was higher in CML PMNL than that in normal, it was not statistically significant.
Quenching of [Ca2+]i with EGTA showed considerable decrease in [Ca2+]i levels in normal PMNL but not in CML PMNL. On further quenching of [Ca2+]i by MnCl2, the [Ca2+]i levels were maintained above the basal levels in CML PMNL whereas, in normal PMNL these were lower than the basal levels (Table 1). Significant quenching of [Ca2+]i is seen on addition of MnCl2 in both the populations. When ratios of these EGTA and MnCl2 treated normal and CML populations were compared they were significantly higher in CML PMNL than the respective normal PMNL. Thus, it shows that though the levels of [Ca2+]i and quenching of [Ca2+]i were always lower in CML PMNL, the extent of stimulation, i.e. the ratios of these [Ca2+]i levels to the basal levels were maintained at higher levels in CML PMNL as compared to that in normal PMNL.
Stimulation with C5a
C5a stimulated normal PMNL showed broad negatively skewed peaks (Fig. 1B; c, d, e). The [Ca2+]i levels were significantly higher than the basal [Ca2+]i levels. The maximum increase was at 30 sec after C5a stimulation (Fig. 2 and Table 2). Treatment of these cells with calcium ionophore A23187 resulted into two peaks. A major peak was seen with a modal channel shifted to right as compared to that seen at 30 sec (Fig. 1B; f), indicating a further increase in [Ca2+]i levels. A small population of PMNL that formed a minor peak was probably non-respondent to the ionophore treatment. The addition of EGTA resulted in a significant shift of the modal channel of the major peak to the left and its broadening (Fig. 1B; g). This showed increased heterogeneity in normal PMNL with respect to quenching of [Ca2+]i by EGTA.
Though after EGTA treatment, [Ca2+]i levels reduced significantly than that in ionophore treated PMNL, they remained at a significantly higher level than the basal [Ca2+]i levels (Table 2). On further addition of MnCl2, the major peak broadened further and both the peaks shifted towards Y-axis showing heterogeneity in the PMNL population as far as sensitivity to quenching of calcium by MnCl2 was considered (Fig. 1B; h). After MnCl2 treatment the [Ca2+]i levels reached below the basal [Ca2+]i levels and these were significantly lower than that in ionophore treated PMNL. To compare the extent of stimulation in PMNL, the ratio of [Ca2+]i levels before and after C5a stimulation was calculated. In normal PMNL, it ranged from 1.2 to 2.89. On further treatment with ionophore though the mean ratio increased to 2.57 ± 0.25, this increase was not statistically significant. Further additions of EGTA and MnCl2 led to 14% and 55% quenching of fluorescence. This decrease in fluorescence was statistically significant (Fig. 3 and Table 2).
C5a stimulated CML PMNL showed negatively skewed peaks that were shifted towards right (Fig. 1D; c, d, e). These increases in [Ca2+]i levels after C5a stimulation were significantly higher than the basal [Ca2+]i levels. Maximum increase in the [Ca2+]i levels was seen at 30 sec after C5a stimulation (Fig. 2 and Table 2). On treatment of these cells with calcium ionophore A23187 [Ca2+]i levels increased further. The addition of EGTA resulted in the building up of an extended tail on the left side of the peak indicating heterogeneity in PMNL population as far as sensitivity to quenching of calcium by EGTA was considered. About 20% of the total population was lying in this tail and hence the decrease in total [Ca2+]i levels was considerable (Table 2). On further addition of MnCl2, a major bell shaped peak along with a minor peak towards extreme left was seen (Fig. 1D; g). Thus, PMNL population was extremely heterogeneous as far as sensitivity to quenching of calcium by MnCl2 was considered. This further reduced [Ca2+]i levels to considerably lower levels (Table 2). In CML PMNL the ratio over basal level ranged from 1.45 to 3.77. Though addition of ionophore increased the mean ratio to 2.71 ± 0.23, it was not statistically significant. In contrast to this, sequential addition of EGTA and MnCl2 decreased the fluorescence intensity significantly, by 15% and 58%, respectively (Fig. 3).
Both normal and CML PMNL showed significantly higher [Ca2+]i levels on C5a stimulation as compared to the respective basal [Ca2+]i levels. But the heterogeneity with respect to [Ca2+]i levels was higher in CML PMNL. Though at 10 sec after C5a stimulation the levels of [Ca2+]i, were comparable in normal and CML PMNL, at later time points these were higher in normal PMNL. However, these differences were not statistically significant. Maximum stimulation was seen at 30 sec in both the populations. In CML PMNL the [Ca2+]i levels increased by 2.33 times whereas in normal PMNL it increased by 2.13 times the basal level (Table 2 and Fig. 3). The extents of stimulation were higher in C5a stimulated CML PMNL at 10 sec and 30 sec as compared to normal PMNL. At 60 sec, a steep decrease in [Ca2+]i levels was seen in CML PMNL while in normal PMNL it was gradual. This resulted in reversing the scenario (Fig. 3). On addition of calcium ionophore A23187 subsequent to C5a stimulation, CML PMNL showed a lower amount of [Ca2+]i as compared to normal PMNL (Table 2). But extent of stimulation was higher in CML PMNL than that in normal PMNL. However, this was not statistically significant. Similarly differences in the decrease in the [Ca2+]i by EGTA and MnCl2 in normal and CML PMNL were non-significant (Table 2, Fig. 3).
In both CML and normal PMNL, the [Ca2+]i levels in C5a stimulated PMNL were higher than fMLP stimulated PMNL. The maximum [Ca2+]i levels were seen at 10 sec and 30 sec after fMLP and C5a stimulation, respectively (Fig. 2). When the maximum [Ca2+]i levels obtained by treatment of PMNL with calcium ionophore A23187 and minimum [Ca2+]i levels obtained by quenching with EGTA and MnCl2, in fMLP and C5a stimulated normal and CML PMNL were compared, they were not significantly different. Thus, though levels of [Ca2+]i seen in fMLP and C5a stimulated normal and CML PMNL differed, the behavioural pattern, i.e. mode of alterations in [Ca2+]i levels with respect to time, was similar in both the populations.
Measurement of total Ca2+ by spectrofluorimetry
Mobilization of [Ca2+]i eventually leads to efflux of [Ca2+]i. Therefore, to estimate total mobilization of Ca2+ we have quantitated Ca2+ levels in fluo-3 loaded normal and CML PMNL by spectrofluorimetry.
Basal Ca2+ levels in PMNL
The basal levels of Ca2+ in normal PMNL ranged from 224 to 386 nM whereas in CML PMNL it ranged from 175 to 342 nM. As seen in flow cytometric assay the basal levels of Ca2+ were lower in CML PMNL as compared to that in normal PMNL (Table 3). However, this difference was statistically non-significant.
Stimulation with fMLP
On fMLP stimulation of normal PMNL, the total amount of Ca2+ increased significantly at 10 sec and 30 sec as compared to the basal levels; whereas at 60 sec it was significantly lower as compared to the basal level (Table 3). Peak Ca2+ levels were seen at 10 sec (Fig. 4). To compare the extent of stimulation in PMNL, the ratio of Ca2+ levels before and after fMLP stimulation were calculated. In normal PMNL, this ratio ranged from 1.2 to 2.09 (Fig. 5). On addition of EGTA little quenching of the fluo-3 fluorescence was seen as compared to the fluorescence intensity of the calcium ionophore A23187 treated cells. However, the levels of Ca2+ were still significantly higher as compared to the basal levels. The ratio of the two was 1.11 ± 0.23. On addition of MnCl2, a significant quenching of fluorescence occurred, bringing down the ratio to 0.44 ± 0.11 (Fig. 5).
In CML PMNL, the total amount of Ca2+ was significantly higher at 10 sec and 30 sec after fMLP stimulation as compared to basal levels. Whereas at 60 sec, Ca2+ levels were higher than the basal level but it was statistically non-significant. Peak Ca2+ levels were seen at 10 sec (Table 3 and Fig. 4). This was also evident from the ratio over the basal level at different time points after fMLP stimulation, which ranged from 1.45 to 2.70. Similar to normal PMNL, about 30% quenching of fluo-3 was seen on the addition of EGTA as compared to the [Ca2+]i levels achieved after ionophore addition. The Ca2+ levels were 1.28 ± 0.33 times higher than the basal levels. This difference between Ca2+ levels was statistically significant. On addition of MnCl2 the Ca2+ levels decreased significantly, lowering down the ratio to 0.46 ± 0.06 (Fig. 5).
In fMLP stimulated CML PMNL, the Ca2+ levels as well as extent of stimulation were higher than those in normal PMNL (Fig. 5). However, the differences between the two populations were statistically non-significant. The drop in Ca2+ levels after reaching the peak levels was higher in CML PMNL as compared to that in normal PMNL. But, since the Ca2+ levels had reached much higher in CML PMNL as compared to normal PMNL these remained higher than the basal levels for a longer time.
Stimulation with C5a
In normal PMNL, total amount of Ca2+ was significantly higher at all the time points after C5a stimulation as compared to the basal levels, the peak Ca2+ levels being at 10 sec (Table 4 and Fig. 4). The ratio of Ca2+ levels of C5a stimulated normal PMNL over basal Ca2+ levels ranged from 1.25 to 2.56. On addition of EGTA, little quenching of the fluo-3 fluorescence was seen as compared to the fluorescence intensity of the calcium ionophore A23187 treated cells. Though the levels of Ca2+ were higher as compared to the basal levels, they were statistically non-significant. The ratio of the two was 1.22 ± 0.33 (Fig. 5). On addition of MnCl2, a significant quenching of fluorescence occurred lowering down the ratio to 0.59 ± 0.12 (Fig. 5).
Similar to normal PMNL, CML PMNL showed significantly higher Ca2+ levels on C5a stimulation that peaked at 10 sec (Table 4 and Fig. 4). In these cells, the ratio over basal Ca2+ levels ranged from 1.69 to 2.75. Quenching of fluo-3 was seen on addition of EGTA as compared to the fluorescence intensity of the ionophore treated cells. The Ca2+ levels were 1.57 ± 0.33 times higher than the basal levels. However, this difference was not statistically significant. On further addition of MnCl2, the Ca2+ levels decreased significantly, thereby reducing the ratio to 0.73 ± 0.12 (Fig. 5). Thus on C5a stimulation, both, CML and normal PMNL showed significant increase in the Ca2+ levels reaching maximum at 10 sec (Table 4). The concentration of Ca2+ and extent of stimulation were higher in CML PMNL as compared to that in normal PMNL (Fig. 4). But these differences were not statistically significant. The drop in Ca2+ concentrations after reaching the peak stimulation was rapid and more in CML PMNL as compared to that in normal PMNL (Fig. 4).
In normal PMNL, at all the time points after stimulation the levels of Ca2+ were higher in C5a stimulated PMNL than that in fMLP stimulated PMNL (Fig. 4 and Table 4). In CML PMNL, at 10 sec and 30 sec after stimulation Ca2+ levels were higher in fMLP stimulated PMNL than those in C5a stimulated PMNL. At 60 sec calcium levels were higher in C5a stimulated CML PMNL as compared to fMLP stimulated CML PMNL (Fig. 4 and Table 4). Thus, alterations were seen in the levels and time kinetics of Ca2+ mobilization in CML PMNL in response to fMLP and C5a stimulation. Levels of Ca2+ reached after addition of EGTA and MnCl2 were slightly higher in fMLP stimulated PMNL as compared to that in C5a stimulated PMNL (Table 3 and 4). However, this difference was statistically non-significant. Thus, flow cytometric and spectrofluorimetric estimation of Ca2+, indicated a rapid and greater efflux of [Ca2+]i in CML PMNL as compared to that in normal PMNL. But in spite of more efflux of [Ca2+]i, the [Ca2+]i levels in CML PMNL after stimulation were maintained above the basal level. Whereas in normal PMNL, the [Ca2+]i levels after stimulation either dropped down below basal level or were slightly higher. In summary, the fine-tuning of Ca2+ homeostasis is altered in CML PMNL as compared to that in normal PMNL.
Binding of the chemoattractants to their receptors on PMNL transmits signals across the plasma membrane, initiating phosphophorylation cascade and changes in [Ca2+]i concentration, that are crucial to cell activation. Both the release of calcium from the intracellular stores and the influx of calcium from the extracellular space contribute to the rise in [Ca2+]i levels. This [Ca2+]i in turn regulates chemoattractant receptor availability in PMNL .
Calcium from the extracellular space enters the cell cytoplasm through various types of channels, i.e. voltage operated Ca2+ channels (VOCCs), ligand-gated non-specific cation channels (LGCCs) and receptor-activated Ca2+ channels (RACCs). Calcium can also be released from internal Ca2+ stores through inositol 1,4,5-triphosphate (IP3) or ryanodine receptors and is replenished by store-operated channels (SOCs) (Fig. 6) . Presence of ryanodine-sensitive calcium stores that might be involved in receptor mediated chemotaxis has been reported in human PMNL .
During chemotaxis localised increases in [Ca2+]i occur. Two distinct [Ca2+]i storage sites have been identified in PMNL. One site was located peripherally under the plasma membrane and the other in the juxtanuclear space. The central [Ca2+]i storage site released [Ca2+]i in response to fMLP, whereas engagement and clustering of CD11b/CD18 integrins caused [Ca2+]i release from the peripheral stores . The peripheral stores of [Ca2+]i are known to be regulated by the cytoskeleton interactions . The rise in cytosolic [Ca2+]i concentration is caused by the mobilization of Ca2+ from intracellular stores to the cytosol followed by an immediate increase in plasma membrane permeability to extracellular calcium. Efflux of [Ca2+]i is mediated by calmodulin dependent calcium adenosine triphosphatase (Ca-ATPase). Ca-ATPase probably serves as the regulatory and homeostatic mechanism required to maintain low [Ca2+]i concentrations. When the cytosolic [Ca2+]i concentration is increased either by fMLP or ionophore, efflux of [Ca2+]i from the cell occurs as a result of calmodulin mediated activation of Ca-ATPase .
Mobilization of Ca2+ is upstream to various biochemical and functional events stimulated in PMNL by various chemoattractants. Therefore, to understand the mechanism of various defects seen in CML PMNL, we have studied mobilization of Ca2+ in response to chemoattractant stimulation. The intracellular free calcium and the total calcium release, i.e. calcium released intracellularly and extracellularly were studied in CML PMNL in comparison to normal PMNL.
Our studies showed altered calcium homeostasis in CML PMNL. The analysis of fluo-3 loaded normal and CML PMNL showed heterogeneity with respect to the basal [Ca2+]i levels and [Ca2+]i levels reached after stimulation. On further stimulation of the chemoattractant stimulated PMNL with calcium ionophore, the heterogeneity in [Ca2+]i levels decreased. Vast heterogeneity was seen in PMNL with respect to the sensitivity of quenching of [Ca2+]i by EGTA and MnCl2. Similar heterogeneity was reported earlier. Elsner et al reported heterogeneity in [Ca2+]i levels in human PMNL . Additionally, Elsner et al demonstrated heterogeneity in signal transduction pathways leading to mobilization of cytosolic calcium in human PMNL in response to fMLP, C5a and IL-8/NAP-1. Studies by Metzner et al suggested that this heterogeneity could be because individual PMNL required different threshold concentrations of stimulant to induce release of [Ca2+]i . Yee et al have shown that treatment of PMNL with lipopolysaccharide (LPS) leads to increase in [Ca2+]i levels and subsequent stimulation with fMLP resulted in further elevation of [Ca2+]i. LPS augmented the overall responsiveness of a population of PMNL by causing a subpopulation of cells to become highly responsive and able to generate changes in [Ca2+]i upon low level stimulation . This heterogeneity in PMNL with respect to [Ca2+]i mediated signal transduction may be important in the fine control of the non-specific immune system in response to weak environmental signals.
The pattern of [Ca2+]i release on fMLP stimulation differed from that of C5a stimulation. The response to C5a was delayed as compared to that seen for fMLP. This was true for both CML and normal PMNL. This could be because of different time kinetics for internalization and recycling of FPR and C5aR in human PMNL . We have seen that in normal PMNL fMLP receptor internalization occurred by 2–5 min whereas C5a receptor internalization was delayed and occurred by 5–10 min. The reported values of Ca2+ in fluo-3 loaded PMNL are 825 ± 94 nM and 798 ± 102 nM on stimulation with fMLP and C5a respectively . These values are much higher than what we have obtained. These differences in calcium concentrations could be due to the differences in the experimental conditions. Moreover, the levels of calcium reported by Lepidi et al were presumably in the Caucasian population. The differences in calcium levels could be because of the racial differences.
Ca2+ levels are regulated by the interplay of kinases and phosphatases. Phosphorylation is reported to down regulate both agonist induced Ca2+ entry and Ca2+ mobilization . Bcr-abl, a chimeric protein expressed by the Philadelphia chromosome has a high and unregulated tyrosine kinase activity. Many different proteins are aberrantly phosphorylated in the cells expressing bcr-abl . Unique phosphatases that can be modulated by α-interferon are also involved in the abnormalities in CML . High phosphotyrosyl phosphatase activity was observed in PMNL from CML patients in the chronic phase. The activity may be characteristic of mature cells and may regulate cellular events through dephosphorylation of p210bcr-abl . Protein tyrosine phosphatases (PTPs)-PTP1B is enhanced in cells expressing p210bcr-abl and has been shown to play a role in dephosphorylation of p210bcr-abl in vivo. PTP1B may function as a specific, negative regulator of p210bcr-abl signalling in vivo. CD45, a family of transmembrane PTPs is expressed in PMNL. Altered expression of CD45 isoforms has been reported in myeloid leukaemias . The altered expression of kinases and phosphatases may alter phosphorylation of various proteins in CML PMNL. Batliwala et al have reported altered phosphorylation pattern of proteins in CML PMNL. In unstimulated CML PMNL, pp1 and pp5 were extensively phosphorylated while phosphorylation of pp3 had reduced as compared to that in normal PMNL. Upon PMA stimulation, normal PMNL showed phosphorylation of pp1 and pp4. But, in contrast to normal PMNL, pp1 and pp4 in CML PMNL did not respond to PMA . Additionally, alterations in the surface proteins and glycoproteins of plasma membrane of CML PMNL have been reported [32, 33]. These alterations in various proteins and enzymes may contribute to the altered Ca2+ homeostasis in CML PMNL.
Ras related GTPases – rho, rac and cdc42 regulate polymerization of actin to produce stress fibres and lamellipodia . Expression of bcr-abl affects ras and ras related super family of small GTPases [35, 36]. Expression of p210bcr-abl also resulted in reorganization of the actin cytoskeleton in 32DC13 cells , suggesting that alteration of rasGTPases probably resulted in alteration of actin network. Altered actin network could affect release of [Ca2+]i from the intracellular stores and efflux of [Ca2+]i from the cell cytoplasm. Earlier we have observed that in CML PMNL actin polymerization in response to chemoattractant stimulation was delayed and significantly lower. Moreover, distribution of F-actin in CML PMNL was altered as compared to that in normal PMNL [10, 12]. Sachhi et al had reported that PMNL from neonates showed defective chemotaxis, failure in polymerization of actin and lower levels of [Ca2+]i, following stimulation with fMLP or zymosan activated serum . In PMNL, the peripheral stores of calcium are regulated by the cytoskeletal protein interactions . In turn, increased [Ca2+]i levels cause fragmentation of F-actin by activation of actin severing and capping proteins such as gelsolin and macrophage capping protein . In view of this, the lower basal levels of [Ca2+]i and the increased extent of [Ca2+]i release seen in CML PMNL in response to chemoattractants could be because of altered regulation of [Ca2+]i stores by altered actin network.
Mutations or functional abnormalities in the various Ca2+ transporters lead to plethora of diseases. In skeletal muscle pathology, mutations in ryanodine receptors cause malignant hyperthermia and porcine stress syndrome. Mutations in calcium pump cause Brody disease. Various abnormalities in calcium handling proteins have been reported in the heart during aging, hypertrophy, diabetes, etc. . Alterations in [Ca2+]i levels of PMNL are associated with various disorders. Lower [Ca2+]i levels were observed in PMNL from asthmatic patients, in response to PAF stimulation . These lower levels of [Ca2+]i were due to low influx of Ca2+. In localized juvenile periodontitis (JP), the basal levels of [Ca2+]i in PMNL were comparable to that of normal. But, fMLP and C5a induced [Ca2+]i levels were reduced in JP. This was due to decreased Ca2+ influx. These reduced [Ca2+]i levels were associated with reduced PKC levels and defective chemotaxis [40, 41]. Similarly in aged individuals, PMNL showed low [Ca2+]i levels that were associated with defective chemotaxis and oxidative burst . Thus, lower [Ca2+]i levels were often associated with altered chemotaxis. In contrast to this, calcium transients were not required for other PMNL functions such as an oxidative burst. For example, PMNL from Myelodisplastic syndrome (MDS) patients are defective in respiratory burst. Studies by Nakaseko et al showed that basal and fMLP stimulated [Ca2+]i levels in MDS PMNL were comparable to those in normal PMNL . However, O'Flaherty et al have reported that [Ca2+]i transients potentiated PMNL degranulation and respiratory burst . [Ca2+]i is also known to regulate chemoattractant receptor availability in PMNL. In PMNL from asthmatic patients, lower [Ca2+]i levels were associated with lower PAF receptor . O'Flaherty et al have reported that [Ca2+]i depleted PMNL showed decrease in receptors for fMLP, LTB4 and PAF . Our studies showed that though not statistically significant, levels of [Ca2+]i were lower in unstimulated CML PMNL as compared to normal PMNL. These [Ca2+]i levels in CML PMNL were lower up to 1 min after stimulation with fMLP and C5a. However, the total calcium release was more in CML PMNL than that in normal PMNL. This suggested that efflux of [Ca2+]i was more and faster in CML PMNL than that in normal PMNL. Earlier we have shown that CML PMNL showed a lower number of FPR, defects in actin and tubulin polymerization, pinocytosis and degranulation. But, stimulation of oxidative burst in CML PMNL was comparable to that in normal PMNL. Thus, our observations are in agreement with earlier reports that [Ca2+]i levels regulate availability of chemoattractant receptors and low [Ca2+]i levels are associated with defects in actin polymerization, chemotaxis and degranulation. However, [Ca2+]i transients are not essential for stimulation of an oxidative burst. Additionally our results suggest that the levels of [Ca2+]i maintained in the cell immediately after stimulation and not the amount of total calcium released are important in stimulation of various functional events in PMNL. PKC is one of the important messengers leading to various functional events in PMNL. In signal transduction pathway, PKC is downstream to Ca2+ mobilization. In CML PMNL, expression of calcium dependent PKC isoform α was lowered . Various functional events in PMNL are mediated partly or totally through PKC. Therefore, it could be that the defects in various functional events in CML PMNL are due to alterations in PKC caused by altered Ca2+ homeostasis or are the direct effect of altered Ca2+ homeostasis per se.
In summary, our studies show that Ca2+ homeostasis in CML PMNL is altered. This could be one of the contributing factors for the reduced responses seen in CML PMNL. Further studies of the calcium oscillations in CML PMNL and the Ca-ATPase would help in pin pointing the defects in calcium homeostasis in CML PMNL.
Ficoll-Hypaque, 4-(2-hydroxyethyl)-1-piperazine ethane sulphonic acid (HEPES), ethyleneglycol-bis-(3-aminoethyl)-tetra acetate (EGTA), manganese chloride (MnCl2), n-formyl-methionyl-leucyl-phenylalanine (fMLP), complement factor 5a (C5a) and Calcium ionophore A23187 were from Sigma. Dimethyl sulfoxide (DMSO) was from Seezle, Germany. Fluo-3-pentapotassium salt and fluo-3AM – the pentaacetoxymethyl ester of fluo-3 and calcium calibration kit were from Molecular Probes (Eugene, USA).
Patients were diagnosed for CML on the basis of standard clinical and haematological criteria. Heparinized peripheral blood was collected from CML patients in chronic phase of the disease, before commencement of therapy. Healthy individuals were used as controls.
Isolation of PMNL
PMNL from peripheral blood were isolated on a ficoll-hypaque gradient . To lyse RBCs the cell pellet was given a hypotonic shock using 3 parts of chilled Milli-Q water. Isotonicity was restored by the addition of 1/3rd volume of 0.6 M KCl. The PMNL were pelleted and suspended in phosphate buffered saline (PBS). Viability of cells was more than 99% as checked by the erythrocin B dye exclusion test. The PMNL fraction was 95% enriched as checked by Giemsa's staining.
One millimolar stock solution of fluo-3AM was prepared in dry DMSO. Fluo-3 loading of PMNL was carried out as described by Vandenberghe et al with some modifications . Briefly, PMNL were suspended at a density of 1× 107 cells per ml in fluo-3 loading buffer (10 mM HEPES, 137 mM NaCl, 5 mM KCl, 1 mM Na2HPO4, 5 mM glucose and 0.5 mM MgCl2, pH 7.4) containing 5 μM fluo-3AM. PMNL were incubated for 30 min at 37°C with gentle agitation. Cells were then washed twice with Ca2+ estimation buffer (i.e. fluo-3 loading buffer with 1 mM CaCl2) to remove extra cellular fluo-3AM. Finally the cell density was adjusted to 1 × 107 cells per ml in the calcium estimation buffer. The cells were kept in dark at 4°C till use.
Measurement of [Ca2+]i by flow cytometry
106 PMNL were suspended in one ml Ca2+ estimation buffer and incubated in a 37°C water bath, for 5 min before each assay. The samples were run on FACScalibur (Becton-Dickinson, USA) using Cell-Quest software. The samples were excited by an argon ion laser at 488 nm and emission was measured at 525 nm on a logarithmic scale. Before addition of the stimuli, fluorescence of the fluo-3 loaded PMNL was measured. Optimum stimulation of PMNL in suspension with fMLP was seen at a concentration of 10-8 M [10, 11] whereas with C5a it was seen at 10 fold lower concentration . Therefore, PMNL were stimulated with 10-8 M fMLP and 10-9 M C5a. Fluorescence was estimated at 10 sec, 30 sec and 60 sec after addition of the stimuli. For each acquisition, minimum 5000 events were collected. Fmax was estimated using PMNL treated with calcium ionophore A23187. Quenching of external and internal fluo-3 fluorescence was estimated using 0.4 M EGTA and 2 mM MnCl2, respectively. Median channel number was taken as a measure of fluorescence intensity of the sample.
To convert the arbitrary fluorescence units of fluorescence measured by spectrofluorimeter into absolute Ca2+, a calibration procedure described by Vandenberghe was used . The dissociation constant, i.e. Kd for Ca2+ bound fluo-3 was calculated using calibration kit. The maximum concentration of Ca2+ in the kit, i.e. 39.8 μM was taken as Fmax and buffer without Ca2+ was taken as Fmin. The calcium bound fluo-3 and free fluo-3 both showed excitation and emission peak at 506 nm and 526 nm, respectively. Therefore, Kd of fluo-3 was calculated by measuring fluorescence of fluo-3 pentapotassium salt (5 μM) using these excitation and emission wavelengths on a Shimadzu RF1501 spectrofluorimeter. The cuvette holder was maintained at 37°C.
The Kd of fluo-3 was calculated using the equation:
Kd = [Ca2+]i/[(F-Fmin)/(Fmax-F)] equation I 
The experiment was done four times independently.
Measurement of total Ca2+ by spectrofluorimetry
106 fluo-3 loaded PMNL were suspended in two ml Ca2+ estimation buffer and incubated in a water bath at 37°C for 5 min, before each assay. PMNL were then stimulated with 10-8 M fMLP or 10-9 M C5a. The fluorimetric reading was taken before stimulation and at 10 sec, 30 sec and 60 sec after stimulation.
Autofluorescence was calculated using unloaded PMNL. Fmax, the maximum fluorescence was obtained by treating the cells with 10 μM of calcium ionophore A23187. F represents the fluorescence of the test sample. Quenching of external and internal fluo-3 fluorescence was estimated using 0.4 M EGTA and 2 mM MnCl2, respectively. Fluo-3-Mn2+ complex is eight times less fluorescent as compared to fluo-3-Ca2+ complex. Under experimental conditions, the quenching of fluo-3 by MnCl2 was low as compared to autofluorescence of PMNL, hence autofluorescence was considered as Fmin . The spectrofluorimetric values were converted to concentration of Ca2+ by using equation I.
Non-parametric tests were applied for statistical analysis of the data. Wilcoxon signed rank test was used to compare the median fluorescence channel and absolute calcium concentrations within normal and CML samples. The Mann-Whitney Wilcoxon test was used to compare the median fluorescence channel and absolute calcium concentrations of normal and CML samples. The Mann-Whitney Wilcoxon test was also used to compare the median fluorescence channel and absolute calcium concentration values of fMLP stimulated PMNL with that of C5a stimulated PMNL.
Kaldjian EP, Stoolman LM: Regulation of L-selectin mRNA in Jurkat cells. Opposing influences of calcium- and protein kinase C-dependent signaling pathways. J Immunol. 1995, 154 (9): 4351-4362.
Maxfield FR: Regulation of leukocyte locomotion by Ca2+. Trends Cell Biol. 1993, 3 (11): 386-391. 10.1016/0962-8924(93)90088-I
Marks PW, Maxfield FR: Local and global changes in cytosolic free calcium in neutrophils during chemotaxis and phagocytosis. Cell Calcium. 1990, 11 (2–3): 181-190. 10.1016/0143-4160(90)90069-7
Schiffmann E, Corcoran BA, Wahl SM: N-formylmethionyl peptides as chemoattractants for leucocytes. Proc Natl Acad Sci USA. 1975, 72 (3): 1059-1062. 10.1073/pnas.72.3.1059
Fernandez HN, Henson PM, Otani A, Hugli TE: Chemotactic response to human C3a and C5a anaphylatoxins. I. Evaluation of C3a and C5a leukotaxis in vitro and under stimulated in vivo conditions. J Immunol. 1978, 120 (1): 109-115.
Syndermann RUR: Phagocytic cells: Stimulus-Response coupling mechanisms. Inflammation: Basic principles and clinical correlates. Edited by: Gallin JI GI, Snyderman R. 1992, 421-439. New York: Raven Press.
Galbraith PR, Abu-Zahra HT: Granulopoiesis in chronic granulocytic leukaemia. Br J Haematol. 1972, 22 (2): 135-143.
Anklesaria PN, Advani SH, Bhisey AN: Defective chemotaxis and adherence in granulocytes from chronic myeloid leukaemia (CML) patients. Leuk Res. 1985, 9 (5): 641-648. 10.1016/0145-2126(85)90146-8
Naik NR, Advani SH, Bhisey AN: PMN cells from chronic myeloid leukaemia (CML) patients show defective chemotaxis in remission. Leuk Res. 1989, 13 (11): 959-965. 10.1016/0145-2126(89)90002-7
Naik NR, Bhisey AN, Advani SH: Flow cytometric studies on actin polymerization in PMN cells from chronic myeloid leukaemia (CML) patients. Leuk Res. 1990, 14 (10): 921-930. 10.1016/0145-2126(90)90183-A
Radhika V, Thennarasu S, Naik NR, Kumar A, Advani SH, Bhisey AN: Granulocytes from chronic myeloid leukaemia (CML) patients show differential response to different chemoattractants. Am J Hematol. 1996, 52 (3): 155-164. 10.1002/(SICI)1096-8652(199607)52:3<155::AID-AJH4>3.0.CO;2-S
Radhika V, Naik NR, Advani SH, Bhisey AN: Actin polymerization in response to different chemoattractants is reduced in granulocytes from chronic myeloid leukaemia patients. Cytometry. 2000, 42 (6): 379-386. 10.1002/1097-0320(20001215)42:6<379::AID-CYTO1005>3.0.CO;2-Q
Mandeville JT, Maxfield FR: Calcium and signal transduction in granulocytes. Curr Opin Hematol. 1996, 3 (1): 63-70.
Cunningham CC, Stossel TP, Kwiatkowski DJ: Enhanced motility in NIH 3T3 fibroblasts that overexpress gelsolin. Science. 1991, 251 (4998): 1233-1236. 10.1126/science.1848726
O'Flaherty JT, Rossi AG, Jacobson DP, Redman JF: Roles of Ca2+ in human neutrophil responses to receptor agonists. Biochem J. 1991, 277 (Pt 3): 705-711.
Bootman MD, Berridge MJ: The elemental principles of calcium signaling. Cell. 1995, 83 (5): 675-678. 10.1016/0092-8674(95)90179-5
Elferink JG, De Koster BM: Ryanodine as inhibitor of chemotactic peptide-induced chemotaxis in human neutrophils. Biochem Pharmacol. 1995, 50 (7): 975-979. 10.1016/0006-2952(95)00222-L
Pettit EJ, Hallett MB: Two distinct Ca2+ storage and release sites in human neutrophils. J Leukoc Biol. 1998, 63 (2): 225-232.
Al-Mohanna FA, Pettit EJ, Hallett MB: Does actin polymerization status modulate Ca2+ storage in human neutrophils? Release and coalescence of Ca2+ stores by cytochalasins. Exp Cell Res. 1997, 234 (2): 379-387. 10.1006/excr.1997.3596
Perianin A, Synderman R: Analysis of calcium homeostasis in activated human polymorphonuclear leukocytes. Evidence for two distinct mechanisms for lowering cytosolic calcium. J Biol Chem. 1989, 264 (2): 1005-1009.
Elsner J, Kaever V, Emmendorffer A, Breidenbach T, Lohmann-Matthes ML, Roesler J: Heterogeneity in the mobilization of cytoplasmic calcium by human polymorphonuclear leukocytes in response to fMLP, C5a and IL-8/NAP-1. J Leukoc Biol. 1992, 51 (1): 77-83.
Metzner B, Elsner J, Dobos G, Kownatzki E, Parlow F, Schraufstatter I, Norgauer J: [Ca2+]i-transients and actin polymerization in human neutrophils under stimulation with GRO alpha and complement fragment C5a. Agents Actions. 1994, 42 (3–4): 101-106. 10.1007/BF01983473
Yee J, Christou NV: Neutrophil priming by lipopolysaccharide involves heterogeneity in calcium-mediated signal transduction. Studies using fluo-3 and flow cytometry. J Immunol. 1993, 150 (5): 1988-1997.
Van Epps DE, Simpson S, Bender JG, Chenoweth DE: Regulation of C5a and formyl peptide receptor expression on human polymorphonuclear leukocytes. J Immunol. 1990, 144 (3): 1062-1068.
Lepidi H, Zaffran Y, Ansaldi JL, Mege JL, Capo C: Morphological polarization of human polymorphonuclear leucocytes in response to three different chemoattractants: an effector response independent of calcium rise and tyrosine kinases. J Cell Sci. 1995, 108 (Pt 4): 1771-1778.
Montero M, Garcia-Sancho J, Alvarez J: Phosphorylation down-regulates the store-operated Ca2+ entry pathway of human neutrophils. J Biol Chem. 1994, 269 (6): 3963-3967.
Salgia R, Brunkhorst B, Pisick E, Li JL, Lo SH, Chen LB, Griffin JD: Increased tyrosine phosphorylation of focal adhesion proteins in myeloid cell lines expressing p210BCR/ABL. Oncogene. 1995, 11 (6): 1149-1155.
Seong DC, Sims S, Johnson E, Howard OM, Reiter B, Hester J, Talpaz M, Kantarjian H, Deisseroth A: A phosphatase activity present in peripheral blood myeloid cells of chronic myelogenous leukaemia patients but not normal individuals alters nuclear protein binding to transcriptional enhancers of interferon-inducible genes. J Clin Invest. 1990, 86 (5): 1664-1670.
Shibata K, Nishimura J, Takahira H, Nawata H: Phosphotyrosine phosphatase activity prevents the detection of P210bcr/abl protein in mature cells in chronic myelogenous leukaemia even by an immunoblotting technique. Leukaemia. 1989, 3 (9): 615-619.
Schiavone EM, Lo Pardo C, Di Noto R, Manzo C, Ferrara F, Vacca C, Del Vecchio L: Expression of the leucocyte common antigen (LCA, CD45) isoforms RA and RO in acute haematological malignancies: possible relevance in the definition of new overlap points between normal and leukaemic haemopoiesis. Br J Haematol. 1995, 91 (4): 899-906.
Batliwalla F, Advani S, Gothoskar B, Zingde S: Differential phosphorylation in normal and leukemic granulocytes in response to phorbol 12-myristate 13-acetate. Leuk Res. 1994, 18 (5): 327-336. 10.1016/0145-2126(94)90016-7
Baker MA, Kanani A, Hindenburg A, Taub RN: Changes in the granulocyte membrane following chemotherapy for chronic myelogenous leukaemia. Br J Haematol. 1986, 62 (3): 431-438.
Zingde SM, Mungikar AM, Chhajlani V, Advani SH, Gothoskar BP: Plasma membranes from normal and chronic myeloid leukemic granulocytes: isolation and two-dimensional polyacrylamide gel electrophoretic analysis. Cancer Biochem Biophys. 1985, 7 (4): 333-342.
Nobes CD, Hall A: Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell. 1995, 81 (1): 53-62. 10.1016/0092-8674(95)90370-4
Druker B, Okuda K, Matulonis U, Salgia R, Roberts T, Griffin JD: Tyrosine phosphorylation of rasGAP and associated proteins in chronic myelogenous leukaemia cell lines. Blood. 1992, 79 (9): 2215-2220.
Mandanas RA, Leibowitz DS, Gharehbaghi K, Tauchi T, Burgess GS, Miyazawa K, Jayaram HN, Boswell HS: Role of p21 RAS in p210 bcr-abl transformation of murine myeloid cells. Blood. 1993, 82 (6): 1838-1847.
Sacchi F, Augustine NH, Coello MM, Morris EZ, Hill HR: Abnormality in actin polymerization associated with defective chemotaxis in neutrophils from neonates. Int Arch Allergy Appl Immunol. 1987, 84 (1): 32-39.
Missiaen L, Robberecht W, van den Bosch L, Callewaert G, Parys JB, Wuytack F, Raeymaekers L, Nilius B, Eggermont J, De Smedt H: Abnormal intracellular ca(2+)homeostasis and disease. Cell Calcium. 2000, 28 (1): 1-21. 10.1054/ceca.2000.0131
Bialasiewicz P, Nowak D, Krol M, Antczak A: Altered intracellular calcium signalling after PAF stimulations in polymorphonuclear leukocytes from asthmatic patients. J Lipid Mediat Cell Signal. 1997, 17 (1): 21-30. 10.1016/S0929-7855(97)00018-7
Agarwal S, Reynolds MA, Duckett LD, Suzuki JB: Altered free cytosolic calcium changes and neutrophil chemotaxis in patients with juvenile periodontitis. J Periodontal Res. 1989, 24 (2): 149-154. 10.1111/j.1600-0765.1989.tb00870.x
Shibata K, Warbington ML, Gordon BJ, Kurihara H, Van Dyke TE: Defective calcium influx factor activity in neutrophils from patients with localized juvenile periodontitis. J Periodontol. 2000, 71 (5): 797-802. 10.1902/jop.2000.71.5.797
Lipschitz DA, Udupa KB, Boxer LA: The role of calcium in the age-related decline of neutrophil function. Blood. 1988, 71 (3): 659-665.
Nakaseko C, Asai T, Wakita H, Oh H, Saito Y: Signalling defect in FMLP-induced neutrophil respiratory burst in myelodysplastic syndromes. Br J Haematol. 1996, 95 (3): 482-488.
Balasubramanian N, Advani SH, Zingde SM: Protein kinase C isoforms in normal and chronic myeloid leukemic neutrophils. Distinct signal for PKC alpha by immunodetection on PVDF membrane, decreased expression of PKC alpha and increased expression of PKC delta in leukemic neutrophils. Leuk Res. 1998, 22 (7): 597-604. 10.1016/S0145-2126(98)00045-9
Boyum A: Separation of blood leucocytes, granulocytes and lymphocytes. Tissue Antigens. 1974, 4 (4): 269-274.
Vandenberghe PA, Ceuppens JL: Flow cytometric measurement of cytoplasmic free calcium in human peripheral blood T lymphocytes with fluo-3, a new fluorescent calcium indicator. J Immunol Methods. 1990, 127 (2): 197-205. 10.1016/0022-1759(90)90069-8
Mandecki W, Mollison KW, Bolling TJ, Powell BS, Carter GW, Fox JL: Chemical synthesis of a gene encoding the human complement fragment C5a and its expression in Escherichia coli. Proc Natl Acad Sci USA. 1985, 82 (11): 3543-3547. 10.1073/pnas.82.11.3543
Minta A, Kao JP, Tsien RY: Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores. J Biol Chem. 1989, 264 (14): 8171-8178.
Merritt JE, McCarthy SA, Davies MP, Moores KE: Use of fluo-3 to measure cytosolic Ca2+ in platelets and neutrophils. Loading cells with the dye, calibration of traces, measurements in the presence of plasma, and buffering of cytosolic Ca2+. Biochem J. 1990, 269 (2): 513-519.
We thank Dr. A. N. Bhisey for his encouraging support. We are thankful to Dr. I. Mittra for the use of flow cytometer. The work was supported by a grant from Indian National Science Academy. C.M.R was a fellow of University Grants Commission, India.
The author(s) declare that they have no competing interests.
CMR participated in the standardization of the techniques, acquisition of the data, analysis and drafting the manuscript. SHA was a clinical collaborator and was involved in diagnosis of the patients and providing clinical samples for the studies. NRN conceived the study, participated in the design of the study, standardization of the techniques, acquisition of the flow cytometry data, coordination and helped to draft the manuscript. All the authors have read and approved the final manuscript.