These results confirm the observations made by Nemazee and colleagues, who showed that receptor editing in the spleen is marginal and that IgD-positive T2 cells undergo apoptosis upon BCR cross-linking 36. Collectively, our results suggest that BAFF-R expression is regulated by BCR signaling and that the outcome of BCR signaling on BAFF-R expression is B-cell developmental stage dependent, namely a down-modulation on immature B cells

and up-regulation on mature B cells. Recently, Dabrafenib cost we could show that expression of BAFF-R on mature B cells is required for their maintenance and not only for their development beyond transitional type 1 B cells 20. This suggests that for survival, mature B cells do not

rely on surface expression of BCR alone 37. As already mentioned, triggering of both receptors mediates activation of NF-κB, suggesting a potential and elegant mechanism for B cells to determine their lifespan also within the mature compartment. Up-regulation of BAFF-R upon BCR ligation could ensure only on mature B cells an increased survival and allow them to undergo the necessary final differentiation stages within the B-cell follicles. Findings in support of this assumption come from the observations made in mice lacking both Rac-1 and Rac-2. Such mice have defective BCR signaling, resulting in diminished numbers of splenic B cells, but normal numbers of BM B cells. Furthermore, this impaired BCR signaling also leads to reduced levels of BAFF-R, pointing to a direct regulation of BAFF-R expression by BCR signaling via the Rac-1 and Rac-2 pathway 38. Collectively, we suggest a mechanism PI3K Inhibitor Library concentration by which BAFF-BAFF-R signaling determines the survival acetylcholine time window for B cells beyond the immature B-cell stage, and in particular upon rearrangement and expression of their BCR. The tight control of surface BAFF-R expression by BCR ligation according to the developmental stage supports our hypothesis. Thus, B cells can exploit the same signaling mechanisms for two different outcomes according to the biological requirements, namely reduced survival/deletion of auto-reactive B cells

within immature B cells and increased survival within mature B cells. In addition, our data allowed us to link mouse and human B-cell biology in regard to BAFF-R expression. In both species, BAFF-R expression starts at the immature B-cell stage and a correlation exists between BAFF-R and surface IgM expression, suggesting that for human B cells as well, the BCR is controlling BAFF-R up-regulation. Moreover, we show that recombination, by means of RAG2 expression, is almost exclusively confined to the BAFF-R negative fraction. Thus, for immature B cells in the mouse, BAFF-R expression is induced on positively selected cells. Female C57BL/6 mice were purchased from RCC (Füllinsdorf). Mice were used at 6–8 weeks of age.

As with CCR7, we showed previously that the level of CD38 expression does not correlate ABT-888 clinical trial with chemotaxis towards CCL19 [24]. Nevertheless, we could see that DC stimulated with bromelain

or with bromelain in combination with the cytokine cocktail without PGE2 had noticeably higher MFI values for CD38 (Fig. 2B). Addition of reduced amounts of PGE2 did not increase the MFI. Thus, PGE2 had an inhibitory effect of CD38 expression on DC, similar to IL-12p70 production. Interestingly, a correlation between CD38 expression and IL-12p70 secretion of DC has been described previously [33], in agreement with our data. The only DC population capable of producing higher amounts of IL-12p70 was DC stimulated https://www.selleckchem.com/products/z-ietd-fmk.html with bromelain in combination with the cytokine cocktail without PGE2. We expected to find a higher secretion of IL-12p70 in the group stimulated with the cytokine cocktail without PGE2, as PGE2 has been claimed to be responsible for the lack of this cytokine, but our results indicate that it is not enough to only remove PGE2. In addition to not producing any notable amounts of IL-12p70, these DC also showed a less mature phenotype compared with the other groups, so obviously PGE2 is necessary for inducing (phenotypic) maturation. However, addition of bromelain could overcome this lack of stimulation. On the other hand, bromelain alone was not potent enough to induce both phenotypic

maturation and high IL-12p70 production. The lack of IL-12p70 production was not a result of a general inability of the DC, as we detected large amounts of IL-12p70 after stimulation with the bacterial compound OK432

using DC from the same preparation [24]. Comparing the functionality of the generated DC populations in a MLR, we could show that PGE2 also influenced the T cell stimulatory capacity of the DC. When DC stimulated with the modified cytokine cocktail without PGE2 were cocultured with lymphocytes, fewer proliferative T cells were detected. Addition of ¼ of PGE2 to the cocktail improved this stimulatory capacity. This was also true regarding the phenotype of the cells. Use of ¼ of the amount Tenoxicam of PGE2 in the cocktail increased the expression of surface maturation markers, and some markers had even higher surface expression using this stimulation than with the original cytokine cocktail. Addition of bromelain to both the original and the modified cytokine cocktail with reduced PGE2 resulted in an even more mature phenotype, but this phenotype had an insufficient secretion of IL-12p70. Because IL-12p70 is essential for a strong induction of cytotoxic T lymphocyte (CTL) responses, several other attempts to generate DC with high IL-12p70 secretion have been made by other research groups. Stimulation with polyriboinosinic polyribocytidylic acid (poly I:C) has shown to generate DC capable of producing high amounts of IL-12p70 [34, 35].

The association of MCL and FcεRI-γ is surprising given that MCL lacks the canonical motif — a positively charged amino acid in the transmembrane

domain — for binding activating adaptors, and others have tried and failed to demonstrate this association [4]. The Thr38 residue of MCL that they postulate mediates the association with FcεRI-γ is conserved in the rat, but we have been unable to demonstrate any direct association of rat MCL to FcεRI-γ. The direct recognition of TDM that Miyake et al. [13] describe suggests that MCL can play a role in TDM recognition independently of its association with Mincle. In our hands, rat MCL reporters are not stimulated by mycobacteria, while Mincle reporters are stimulated by mycobacteria (Supporting SAR245409 cost Information Fig. 1). Although it is unknown AZD1152-HQPA chemical structure exactly how TDM is recognized by Mincle, both TDM and the Malassezia ligand for Mincle [21] are glycolipids. Although the presence of both the saccharide and lipid portions of TDM is important for recognition by Mincle [10], it is likely that the sugar moiety is the major antigen determinant. Sugar recognition is mediated by the lectin domain, and within this domain,

a tripeptide motif is thought to heavily influence the type of sugar moieties that can be recognized. An EPX motif (where X is usually asparagine) mediates binding to glucose moieties such as found in TDM [22]. The EPN tripeptide motif is conserved in Mincle from rat, mouse, and human, and Mincle from all three species is able to mediate recognition of Malassezia and mycobacterial cord factor ([8, 10, 11] and our unpublished data). For MCL, the EPX motif is conserved in rat and human (although X is D in human and K in rat), but in mouse only the E is conserved. This suggests that there is little selection pressure on this motif in MCL or that different ligands are recognized by the different species. In addition, MCL has previously been shown to have very weak sugar binding [23]. One possible explanation for the differences we

see is that MCL binds rather to the lipid Oxalosuccinic acid portion. Although lipid binding by C-type lectins is unusual, it is not unheard of — surfactant proteins A and D are both able to bind to a range of lipids via their carbohydrate recognition domains [24]. In their experimental system with purified TDM, the lipid portion is presumably exposed and available for binding to MCL reporter lines; in our system with intact mycobacteria, the lipid portion may be buried in the membrane and thus unable to stimulate our MCL reporters. If this hypothesis is correct, the Mincle/MCL heterodimer described here could allow co-ordinate binding to the TDM molecule, with Mincle binding to the sugar moiety and MCL to the lipid. The congenic rat strains DA.APLEC (APLEC gene complex from PVG) [25] and DA.NKCB (NK complex from PVG) [26] were maintained under conventional conditions.

4C, D). However, not only γδCD8αα+ iIEL but also αβCD8α+ iIEL cells showed a basal [Ca2+]i decrease. This was unlikely to be a direct effect of the GL3 mAb on αβ iIEL but may be due to changes in the composition of αβCD8α+ iIEL, e.g. through attraction of systemic αβ+CD8+ cells with lower basal [Ca2+]i levels into the gut epithelium 40. In contrast, basal [Ca2+]i levels of neither systemic CD8− p-γδ nor CD8− i-γδ were altered by GL3-treatment (Fig. 4C and D). These data suggest that the observed high basal [Ca2+]i levels of γδCD8αα+ Small molecule library price iIEL reflect a constant TCR-specific activation in vivo,

which could be partially blocked by anti-γδ TCR mAb treatment. Next, we investigated how γδ T cells from GL3-treated γδ reporter mice responded to TCR stimulation. As shown in Fig. 4A, the TCR complex was down-regulated

but still present at residual levels on the cell surface of these γδ T cells. We found that anti-CD3 and anti-γδ Sirolimus TCR mAb clustering still elicited Ca2+-fluxes in CD8− p-γδ and CD8− i-γδ from mice injected with GL3, albeit with lower or almost flat amplitudes compared with those from mock-treated animals. The iIEL populations CD8+ i-γδ and CD8+ i-αβ only showed a decrease of basal [Ca2+]i, without evident mAb-induced Ca2+-flux neither in PBS nor in GL3 treated mice (Fig. 5A). The quantification of these changes, displayed as fold of basal [Ca2+]i PTK6 levels after anti-CD3 and anti-γδ TCR mAb clustering, showed that CD8− p-γδ and CD8− i-γδ were affected by the GL3 treatment (Fig. 5B). In addition, iIEL from PBS- and GL3-treated γδ reporter mice were analyzed

for responsiveness to ex vivo stimulation with GL3 and GL4, a different anti-γδ TCR mAb. In vivo treatment with GL3 reduced the TCR-dependent CCL4 and IFN-γ production of γδ iIEL (Fig. 5C). Surprisingly, the CCL4 and IFN-γ production capability of γβ iIEL from GL3-treated γδ reporter mice stimulated ex vivo with the anti-αβ TCR (H57) was increased (Fig. 5D). In conclusion, γδ iIEL suffered a loss of function in response to TCR stimuli when their TCR was modulated by GL3 treatment for 6 days. Together, this suggests that the iIEL do not become exhausted and do not change their activated phenotype with repeated high-dose anti-γδ TCR treatment. However, the down-modulation of their surface TCR in combination with the decoration of residual surface γδ TCR is likely to be the reason for the diminished TCR responsiveness and cytokine production. This further implies a role for the TCR in the physiology of γδ T cells. However, it is at present not clear to what extent the responsiveness of γδ T cells to other stimuli, e.g. engagement of other receptors such as NKG2D or TLR, may be also altered by TCR modulation. The question whether, after thymic selection, the TCR on γδ T cells had a physiological role at all was not unanticipated 19, 23.

Whether vascular calcification can be prevented or reversed with strategies INK-128 aimed at maintaining phosphate homeostasis is as yet unknown. One recent study also determined an association between serum phosphate within the normal range and vascular and valvular calcification.21 This study of 439 young and middle-age participants from the Multi-Ethnic Study of Atherosclerosis (MESA) with both normal renal function and CKD, and no known CVD, reported that after adjustment for eGFR, each 1 mg/dL increase in serum phosphate concentration was significantly associated with a 21%, 33%,

25% and 62% greater prevalence of coronary artery, thoracic, aortic valve and mitral valve calcification respectively. The CARDIA study, described earlier, also showed that phosphate levels within the reference range were significantly associated with coronary artery calcium levels in a young healthy adult population.19 Elevations in serum phosphate have been associated with structural changes and renal decline in animal models.68 In human observational studies, hyperphosphataemia is associated with progression of established CKD and the development of ESKD (end-stage Copanlisib in vivo kidney

disease)23,69–71 and studies of renal transplant recipients describe an association between higher serum phosphate and renal allograft loss.27,28 Serum phosphate levels in the upper-normal range have also recently been reported to be associated with an increased risk of developing incident CKD and ESKD.6,24 One study involving 2269 participants from the Framingham Heart Study showed that those in the highest phosphate category had an increased risk of CKD with OR 2.14 (95% CI 1.07–4.28) 4��8C when compared with the reference group with serum phosphate 2.5–3.49 mg/dL.6 The same study also analysed 13 372 participants

from the Third National Health and Nutrition Examination Survey (NHANES III) and reported that phosphate ≥4 mg/dL was associated with an increased risk of incident ESKD (RR 1.90 (95% CI 1.03–3.53)). Zoccali et al. recently evaluated the relationship between baseline serum phosphate, disease progression and response to angiotensin-converting enzyme (ACE) inhibition in 331 patients with proteinuric CKD in the prospective Ramipril Efficacy In Nephropathy (REIN) trial.72 Phosphate levels in the highest two quartiles were significantly associated with faster progression to both ESKD and to a composite end-point of doubling of serum creatinine or ESKD compared with patients with phosphate levels below the median. Therefore, with higher serum phosphate levels the renoprotective effect of ramipril decreased, despite adjustment for potential confounders such as GFR and urinary protein. This suggests that phosphate may potentially modify the protective effect of the only real therapeutic class of agents used in CKD. FGF-23 is the most potent hormone regulating phosphate homeostasis.73 In health, FGF-23 is secreted by osteocytes and osteoblasts in response to dietary phosphate intake.

Induction of in vitro Treg cells was most easily accomplished with anti-CD3 mAb mitogen-based stimulation. Therefore, to control for the use of mitogen-based stimulation, it was necessary to confirm that n-butyrate anergized mitogen-stimulated CD4+ T cells similarly to antigen-stimulated CD4+ T cells. Primary cultures of isolated C57BL/6 CD4+ T cells were stimulated with plate-bound anti-CD3 mAb and soluble

anti-CD28 mAb for 7 days in the presence or absence of n-butyrate. As seen in Fig. 1A, n-butyrate reduced proliferation of CD4+ T cells by approximately 95% in mitogen-stimulated primary cultures. To test whether n-butyrate induced unresponsiveness was retained after the removal of the HDAC inhibitor, the CD4+ T cells from the primary culture were re-stimulated in secondary cultures that did not contain n-butyrate. As shown in Fig. 1B, control CD4+ T cells JAK inhibitors in development from the

primary cultures proliferated vigorously when re-stimulated in secondary cultures. In contrast, CD4+ T cells from the n-butyrate-treated primary cultures proliferated 83–91% less than untreated CD4+ T cells. The retention of proliferative unresponsiveness in the secondary cultures demonstrated that the CD4+ T cells from the n-butyrate-treated mitogen-stimulated primary cultures were anergic. Anergy in CD4+ T cells usually involves an inability to generate IL-2 in association with proliferative unresponsiveness. Consequently, IL-2 secretion RAD001 solubility dmso by the CD4+ T cells was also examined to confirm the onset of anergy (Fig. 1C). CD4+ T cells from control primary cultures secreted IL-2 in secondary cultures stimulated with anti-CD3 mAb. In contrast, IL-2

secretion Non-specific serine/threonine protein kinase was inhibited in CD4+ T cells from the n-butyrate-treated primary cultures. The anergic CD4+ T cells did not generate any additional IL-2 beyond the detected background levels in response to anti-CD3 mAb stimulation in the secondary cultures. The decreased IL-2 concentration within the anergic CD4+ T cell culture supernatants had no bearing upon proliferation in the n-butyrate-treated CD4+ T cells as seen in Fig. 1B. Taken together, the results in Fig. 1 revealed that n-butyrate induced anergy within mitogen-stimulated CD4+ T cells as determined through significant reduction of proliferation and IL-2 secretion. To determine if n-butyrate increased the percentage of FoxP3+ Treg cells in primary or secondary cultures, CD4+ T cells from transgenic FoxP3EGFP C57BL/6 mice were stimulated in primary cultures with or without n-butyrate. Natural Treg cells as determined by the presence of FoxP3EGFP comprised approximately 8% of isolated lymphoid CD4+ T cells (data not shown). TGF-β was added to additional primary cultures to generate FoxP3+ T cells as a positive control [21]. Percentages of FoxP3+ T cells were quantified daily over the course of 5 days (Fig. 2A). The percentage of CD4+FoxP3+ T cells increased only in the primary cultures stimulated in the presence of TGF-β, as shown on Day 4 in Fig.

Clinical data from the group of patients are listed in Table 1. The age varied between 20 and 85 years (median 66 years). Almost all patients presented various comorbidities, mainly manifestations of the metabolic

syndrome like diabetes mellitus (40.2%), hypertension (58.7%), peripheral arterial occlusive disease (20.6%) or coronary heart disease (27.2%). 17.4% suffered from malignancies, and 19.6% showed various degrees of renal disease including end-stage renal failure reflecting the frequently observed comorbidity status of patients with invasive S. aureus infections (Laupland et al., 2003). Serum samples from specific pathogen-free (SPF) mice, juvenile mice and human sera from healthy adults and umbilical cord blood (UCB) were analyzed by Western blots for the presence of anti-Eap antibodies (Fig. 1a). Antibodies could be detected in various concentrations in all human LY2109761 concentration sera. However, SPF mice as well as juvenile mice did not show any anti-Eap antibody response. Further analysis of the human sera revealed IgM, IgG and IgA antibodies in adult samples, while in UCB, only IgG antibodies were found (Fig. 1b). For further analysis, anti-Eap antibodies were quantified by ELISA. In all blood donors, antibodies could be detected with a considerable variability in titers for IgM and IgG (Fig. 2a). No correlation was found between IgG and IgM antibody titers within individuals (correlation coefficient r2: 0.0074; Fig. 2b). this website Also, when comparing

the results for IgA and IgG from Pomalidomide purchase Western blot analysis, no correlation could be found (data not shown). All 92 patients suffering from S. aureus infections showed anti-Eap antibodies. Both IgM as well as IgG anti-Eap antibody titers were significantly higher in patients compared with healthy individuals (IgM, P=0.007; IgG, P<0.0001, Fig. 2a). However, no correlation could be established between IgM and IgG antibody titers. The avidities of anti-Eap antibodies from

healthy controls and patients were high in both groups, with patients displaying significantly higher avidity indices compared with healthy controls (patients mean 0.805, controls mean 0.696; P<0.0001, Fig. 2c). Because transcription of eap by S. aureus in deep wounds was promoted compared with the superficial wounds (Joost et al., 2009), we determined whether the extent of anti-Eap antibody response also differs as a function of infection type (Table 2). Patients with deep infections showed significantly higher anti-Eap antibody titers than those with superficial infections (P=0.001). Detailed analysis revealed significantly higher titers for patients suffering from abscesses compared with other types of infection (P<0.001). Extremely high titers were found in patients presenting with spondylodiscitis (mean 361.2), although in comparison with patients with other types of infections, these did not reach statistical significance (P=0.057), most likely due to the small number of patients (n=4).

Indeed, when PBMCs derived from IFN-β-treated patients were depleted of monocytes, the strong induction of IL-6 observed in total PBMCs was completely lost. In addition, a strong reduction of BAFF expression was observed in in vivo IFN-β-conditioned PBMCs after the depletion of monocytes. In a similar fashion, in the absence of monocytes, there was no induction of TLR7-driven IgM and IgG production, indicating that IFN-β treatment could exert its therapeutic effects this website by fine-tuning monocyte functions, in the context of TLR7 stimulation, that act through bystander mechanisms on the differentiation of

B lymphocytes. Taking into account that TLR7 is crucial for type I IFN release from pDC [41] and is, at the same time, an IFN-inducible gene [22], we can envisage the existence of a tight relation between IFN-β response and TLR7 responsiveness of MS monocytes, whose full comprehension deserves further investigation. In line with this view, recent data obtained by Molnarfi and collaborators showed that monocytes from RRMS patients exhibited a reduced ability to produce HGF, a neuroprotective and neuroinflammation-suppressive mediator, when compared with HD [42]. Treatment with IFN-β significantly enhanced

HGF PF-6463922 synthesis and secretion by blood monocytes, contributing to the clinical benefit of IFN-β in RRMS via the combined HGF-mediated neuroprotective and anti-inflammatory mechanisms. In this context, it is also important to remind that monocytes are abundant in inflammatory MS brain lesions and displayed also altered functions and an activated innate immune signature Glutamate dehydrogenase in MS patients with clinically more severe course [43]. In particular,

the type I IFN pathway is dysregulated in these monocytes, which may contribute to more active disease. In addition to that, conditional genetic knockout of IFNAR1 in monocytes, but not in T cells, B cells, or central nervous system cells, leads to enhanced disease severity in the animal model of MS [44]. All these evidences indicate that perturbations of the type I IFN signaling pathway and response in monocytes could represent crucial events in MS immunopathology and, at the same time, a key target of IFN-β therapy. On the other hand, we cannot exclude that the replenished TLR7 responsiveness in PBMCs and monocytes of IFN-β-treated MS patients could be related to the rescue or prevention of TLR7 tolerance, that is generally induced by specific ligands of this receptor and leads to a reduced cytokine and Ig production [45]. Indeed, Poovassery and Bishop [45] recently demonstrated that IFN-β controls TLR7 tolerance and activation through the PI3K/Akt/mammalian target of rapamycin signaling pathway but also enhancing TLR7 expression in human B cells.

Strains lacking either of these two mediators

have been shown to be more sensitive to pro-oxidants such as hydrogen peroxide, menadione and methyl viologen or paraquat (7, 9), suggesting that oxyR and rpoS are essential for survival and growth under oxidative conditions. Similar results have been found in other bacterial species and the role of OxyR in the response to oxidative stress is well established. Midostaurin datasheet For example, oxyR mutants of Pseudomonas aeruginosa are hypersensitive to pro-oxidants including H2O2 and paraquat (16) while E. coli with deletions of oxyR are hypersensitive to hydrogen peroxide and have increased rates of spontaneous mutation during aerobic growth (17). Similarly, oxyR mutants of Brucella abortus, Erwinia carotovora and Xanthomonas campestris, all show increased sensitivity to pro-oxidants (17–20). Negative regulation of oxyR by RpoS has been reported in E. coli (21). In particular the degree of β-galactosidase expression from a single-copy oxyR::lacZ fusion in a RpoS-defective strain has been shown to be higher than in its parental strain as the cells enter into, and remain in, the stationary phase growth (21). Additionally, increased expression of RpoS prevents the normal expression of oxyR (21). However, in contrast to this,

Schellhorn observed a significant reduction in oxyR expression in an E. coli rpoS::Tn10 mutant (22), a result supported by our own observations with B. pseudomallei in EPZ-6438 cost which Edoxaban low amounts of CAT activity were observed in oxyR::CAT/rpoS−, which contains a chromosomal oxyR::CAT fusion and is null for rpoS. More significantly, isogenic replacement of RpoS in strain oxyR::CAT/rpoS−/RpoS restored oxyR::CAT expression to the extent seen in the parental strain (oxyR::CAT), suggesting that RpoS acts as a positive regulator of oxyR transcription in

B. pseudomallei. Three genes have been shown to be under transcriptional control of OxyR, namely dpsA (23), katG (24) and gorA (25). The expression pattern of katG during growth of B. pseudomallei has been previously examined using a chromosomal katG::CAT fusion as a reporter. CAT activity was observed to increase during early exponential growth, reaching a maximum value in the early stationary phase growth, after which it declined in the late stationary phase growth (6). Significantly, expression was greater in an oxyR mutant strain during all phases of growth, suggesting that katG expression is negatively regulated by OxyR during normal growth, although further studies showed that katG was positively regulated by OxyR during oxidative stress (6). The negative regulation of katG by oxyR was confirmed in this study, a greater degree of CAT expression being seen in katG::CAT as compared to katG::CAT/oxyR−.

Adoptively transferred p14 CD8+ T cells coexpressed CD44, PD-1 and IL-7Rα as analyzed by FACS analysis of blood (Fig. 2G, Supporting Information Fig. 2C) and spleens (data not shown) 5 days after transfer. Thus, CML-specific CTL display an activated phenotype but retain IL-7Rα Afatinib purchase expression. The fact that specific CTL downregulate IL-7Rα expression in the presence of a chronic infection but maintain IL-7Rα expression in the presence of CML expressing the same viral antigen was surprising and led

to the question if IL-7 production is increased in CML mice. To analyze this, we compared IL-7 expression in mRNA isolated from spleen of CML and naïve C57BL/6 mice by RT-PCR. The thymus as organ with documented high IL-7 production served as a positive control. IL-7 mRNA was detectable in the spleen of CML and of naïve C57BL/6 mice (Fig. 3A and Supporting Information Fig. 3). Next, we analyzed whether IL-7 mRNA is detectable in CML granulocytes and in control granulocytes. We therefore quantitatively compared IL-7 mRNA production of sorted GFP+ granulocytes from CML mice with sorted granulocytes from C57BL/6 mice.

Surprisingly, IL-7 mRNA was detectable in both malignant and control granulocytes (Fig. 3B). Moreover, this experiment revealed that IL-7 mRNA was not differently expressed in malignant and in normal granulocytes. However, the total number of granulocytes in the spleen of mice with CML is three to four-fold higher than that found in C57BL/6 control mice (Fig. 3C). These findings were confirmed by quantification of IL-7 protein levels per Metformin order milligram spleen of naïve C57BL/6 mice and CML mice (Fig. 3D). Furthermore, IL-7 was detectable by intracellular staining of brefeldin-treated malignant (GFP+) and normal (GFP−) granulocytes but not in granulocytes from IL-7-deficient mice (MFI increase of IL-7 in Cyclooxygenase (COX) GFP− granulocytes (12.4±2.9%) and GFP+ granulocytes (11.4±2.9%)

(Fig. 3E and F)). Taken together, the malignant granulocytes produce IL-7 and are increased in numbers in secondary lymphoid organs such as the spleen. To study the role of IL-7 produced by leukemic cells in more detail, H8×IL-7-deficient mice were used as bone marrow donors (H8×IL-7−/−-CML mice) to establish CML disease in C57BL/6 recipients. In this experiment, the leukemic cells will not produce IL-7. However, stromal and epithelial cells of the recipient mouse are capable of IL-7 secretion. Purified p14 CD8+ T cells (CD45.1+CD8+Vα2+) were adoptively transferred to H8×IL-7−/−-CML mice, H8-CML and naïve C57BL/6 mice. P14 CD8+ T cells expanded similarly in H8×IL-7−/−-CML mice and in H8-CML mice (Fig. 4A). However, significantly more p14 CTL survived long term in H8-CML mice than H8×IL-7−/−-CML mice (analyzed in blood: H8-CML: 8.2±3.7%; H8×IL-7−/−-CML: 1.2±0.6%; p=0.04).