Responsive Monocyte Proteins and Decreases Inflammatory Cytokines in ESRD

CLINICAL RESEARCH www.jasn.org

Cholecalciferol Supplementation Alters Calcitriol- Responsive Monocyte Proteins and Decreases Inflammatory Cytokines in ESRD

Jason R. Stubbs,* Arun Idiculla,* Joyce Slusser,† Rochelle Menard,* and L. Darryl Quarles*

*The Kidney Institute, University of Kansas Medical Center, Kansas City, Kansas; and †Flow Cytometry Core Laboratory, University of Kansas Medical Center, Kansas City, Kansas

ABSTRACT In vitro, monocyte 1␣-hydroxylase converts 25-hydroxyvitamin D [25(OH)D] to 1,25-dihydroxyvitamin D to regulate local innate immune responses, but whether 25(OH)D repletion affects vitamin D–responsive monocyte pathways in vivo is unknown. Here, we identified seven patients who had 25(OH)D insufficiency and were undergoing long-term hemodialysis and assessed changes after cholecalciferol and paricalcitol therapies in both vitamin D–responsive proteins in circulating monocytes and serum levels of inflammatory cytokines. Cholecalciferol therapy increased serum 25(OH)D levels four-fold, monocyte vitamin D receptor expression three-fold, and 24-hydroxylase expression; therapy decreased monocyte ϩ 1␣-hydroxylase levels. The CD16 “inflammatory” monocyte subset responded to 25(OH)D repletion the most, demonstrating the greatest increase in vitamin D receptor expression after cholecalciferol. Cholecalciferol therapy reduced circulating levels of inflammatory cytokines, including IL-8, IL-6, and TNF. These data suggest that nutritional vitamin D therapy has a biologic effect on circulating monocytes and associated inflammatory markers in patients with ESRD.

J Am Soc Nephrol 21: 353–361, 2010. doi: 10.1681/ASN.2009040451

There are two sources of vitamin D in humans: Di- min D therapy have been unsuccessful,2–4 because etary sources, including ergocalciferol (D2) and chole- these patients are expected to lack sufficient residual calciferol (D3), and endogenous production by the renal CYP27B1 activity to support circulating serum skin in response to sunlight (D3 only). To become ac- 1,25(OH)2D levels. For this reason, therapeutic ap- tive, vitamin D (D2 and D3) must undergo hydroxy- proaches to treat vitamin D deficiency in patients with lation in the liver to form 25-hydroxyvitamin D ESRD have shifted away from the use of nutritional [25(OH)D], a prehormone that is further hydroxyvitamin D forms to favor the use of calcitriol or its lated via the 1␣-hydroxylase (CYP27B1) enzyme in associated analogues. the kidney to generate 1,25-dihydroxyvitamin D Although calcitriol and its analogues have CLINICAL RESEARCH 1 [1,25(OH)2D], the active form of this hormone. A proved to be important therapies for the treatment separate enzyme, 24-hydroxylase (CYP24), is a deac- of disordered mineral metabolism in patients with tivating enzyme that forms an inactive metabolite of this hormone that is excreted in the bile.1 As patients Received April 28, 2009. Accepted September 24, 2009. with chronic kidney disease (CKD) progress to ESRD, Published online ahead of print. Publication date available at renal CYP27B1 activity decreases and the formation of www.jasn.org.

1,25(OH)2D becomes impaired, resulting in hypocal- Correspondence: Dr. Jason R. Stubbs, 3901 Rainbow Boulevard, cemia and secondary hyperparathyroidism in many of MS 3002, Kansas City, KS 66160. Phone: 913-588-6074; Fax: these patients. Previous attempts to counteract these 913-588-3867; E-mail: [email protected] changes in mineral metabolism with nutritional vita- Copyright ᮊ 2010 by the American Society of Nephrology

J Am Soc Nephrol 21: 353–361, 2010 ISSN : 1046-6673/2102-353 353 CLINICAL RESEARCH www.jasn.org

ESRD, the sole use of these compounds for the correction of vita- functions in peripheral tissues unrelated to mineral metabo- min D deficiency in this setting has the potential for adverse effects lism, and it remains unclear whether this supplementation has and may be a fundamentally flawed treatment strategy.5 It is well any measurable physiologic effects in the setting of ESRD. To established that vitamin D has both “classical” actions to affect address this question, we set out to investigate the physiologic mineral metabolism and “nonclassical” actions at various other impact of nutritional vitamin D repletion on monocyte pro- tissues unrelated to mineral metabolism, including the heart, tein expression and monocyte-derived inflammatory cytokine prostate, and monocytes, among others.6,7 The existence of non- levels in patients who have ESRD and are on hemodialysis. Our classical actions of vitamin D is supported by the presence of vita- findings demonstrate that the administration of nutritional min D regulatory enzymes, such as CYP27B1 and CYP24, within vitamin D in the form of cholecalciferol to patients with non- these nonclassical tissues, where it is hypothesized that circulating functioning kidneys does indeed have biologic effects on cir- 25(OH)D serves as a necessary substrate for the local generation culating monocytes and decreases circulating levels of multiple of 1,25(OH)2D and the regulation of tissue-specific, biologic inflammatory cytokines that have been linked to increased pathways. It is believed that the local production of 1,25(OH)2D morbidity and mortality in this population. Our findings may in these nonclassical tissues forms a microenvironment for the serve as early evidence for in vivo, nonclassical effects of chole- regulation of vitamin D–responsive pathways at the cellular level, calciferol therapy in patients with ESRD and suggest a poten- ultimately leading to biologic effects that are independent of cir- tial benefit of 25(OH)D repletion in this patient population. culating 1,25(OH)2D. Existing associations between vitamin D deficiency and various disease processes, including cancer, diabe- RESULTS tes, infection, and cardiovascular health, have provided evidence for nonclassical actions of 1,25(OH) D.8–15 2 Cholecalciferol Therapy Increases 1,25(OH)2D An overwhelming amount of data supporting nonclassical ac- Production in Patients with ESRD tions of 1,25(OH)2D exists, but it remains unclear whether these We measured changes in the serum markers of mineral metab- actions are related to the local production of this hormone or olism before and after cholecalciferol therapy. We found that from circulating 1,25(OH) D produced by the kidneys. Despite 2 both 25(OH)D and 1,25(OH)2D levels significantly increased an overwhelming amount of ex vivo data suggesting the impor- from the precholecalciferol to postcholecalciferol time points tance of local vitamin D production by cells not involved in min- (Table 1). Not only did these levels rise in response to therapy, eral metabolism, in vivo assessments of the autocrine and para- but also the final 1,25(OH)2D levels fell within the reference crine functions of these cells have been difficult and are range of “normal” values for the general population. Despite a confounded by the concurrent systemic generation of significant rise in both 25(OH)D and 1,25(OH)2D levels, 1,25(OH)2D by the kidneys with nutritional vitamin D therapy. cholecalciferol therapy resulted in no significant change in the This controversy of local versus systemic regulation of the non- other markers of mineral metabolism, including calcium, classical actions of vitamin D has the potential for profound con- phosphorus, fibroblast growth factor 23 (FGF-23), and para- sequences in patients who have chronic kidney disease (CKD) and thyroid hormone (PTH). Two weeks after the discontinuation are treated with calcitriol analogues, because these patients often of cholecalciferol therapy, the levels of 25(OH)D remained sig- have a deficiency of both 25(OH)D and 1,25(OH)2D. Although nificantly elevated compared with pretreatment values. this scenario poses a difficult problem for clinicians developing treatment strategies for these patients, it affords a unique oppor- Cholecalciferol Effect on CYP27B1, CYP24, Vitamin D tunity to study the effects of extrarenal production of calcitriol in Receptor, Toll-Like Receptor 2, and Cathelicidin the setting of minimal renal CYP27B1 activity. Expression in Monocytes Thus, there is a gap in our knowledge about the effects of We used flow cytometry analysis to assess changes in monocyte nutritional vitamin D (D2 or D3) supplementation on cellular CYP27B1 and CYP24 expression as a measure of the intracel-

Table 1. Serum biochemistries for patients who had ESRD and were treated with cholecalciferol and paricalcitol Treatment Time Point Serum Variable Before After After Paricalcitol Cholecalciferol Cholecalciferol Restart Phosphorus (mg/dl) 4.9 Ϯ 0.5 4.6 Ϯ 0.3 5.3 Ϯ 0.8 Calcium (mg/dl) 8.6 Ϯ 0.3 8.4 Ϯ 0.4 8.5 Ϯ 0.3 25(OH)D (ng/ml) 13.9 Ϯ 2.2 53.9 Ϯ 3.3a 47.5 Ϯ 3.9a,b Ϯ Ϯ a Ϯ c 1,25(OH)2D (pg/ml) 9.4 1.1 23.2 1.2 16.9 3.4 Intact PTH (pg/ml) 366.5 Ϯ 86.1 358.9 Ϯ 94.3 351.4 Ϯ 102.8 FGF-23 (pg/ml) 6841.1 Ϯ 5686.6 5816.1 Ϯ 4968.4 12369.1 Ϯ 11292.5 Data are means Ϯ SEM (n ϭ 7); cholecalciferol therapy ϭ 8 wk; paricalcitol therapy ϭ 2 wk. aP Ͻ 0.0001 versus precholecalciferol levels. bP Ͻ 0.01 versus postcholecalciferol levels. cP Ͻ 0.05 versus precholecalciferol levels.

354 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 353–361, 2010 www.jasn.org CLINICAL RESEARCH

lular conversion of 25(OH)D to 1,25(OH)2D in response to cholecalciferol and paricalcitol therapies in patients with ESRD. In addition, we assessed changes in vitamin D receptor (VDR), Toll-like receptor 2 (TLR2), and cathelicidin expression by this same method to evaluate the potential impact of 25(OH)D repletion and paricalcitol therapy on monocyte VDR expression and downstream VDR-responsive pathways. Our flow cytometry analysis of monocytes in these patients revealed a decrease in the CYP27B1 expression after both 25(OH)D repletion with cholecalciferol and the restart of paricalcitol (Figure 1A). In addition, CYP24 expression increased after cholecalciferol therapy, whereas the CYP24 expression returned to baseline levels after the discontinuation of cholecalciferol and restart of paricalcitol therapy (Figure 1A). Combining these measurements to generate the CYP27B1-to-CYP24 ratio demonstrated a decrease in this ratio after both 25(OH)D repletion and paricalcitol therapy (Figure 1A). Our analysis of changes in VDR, TLR2, and cathelicidin expression revealed a significant increase in VDR expression after cholecalciferol administration and a further rise in VDR expression in response to the restart of Figure 1. paricalcitol (Figure 1B). There was no significant change in Change in the CYP27B1, CYP24, VDR, TLR2, and cathelicidin expression in monocytes in response to cholecalcif- TLR2 expression, whereas the cathelicidin expression ϩ erol and paricalcitol in patients with ESRD is shown. CD14 cells slightly decreased in response to cholecalciferol and parical- were isolated from patients (n ϭ 7) at three separate time points citol therapy (Figure 1B). as described previously. Monocytes were analyzed by flow cytometry after the addition of fluorescence-labeled antibodies specific for the targets of interest, and the mean fluorescence Differential Impact of Cholecalciferol Therapy on intensity (MFI) was measured as a quantification of the relative Monocyte Subpopulations protein expression. (A) The MFI for CYP27B1 decreased with both After cholecalciferol administration, flow cytometry analysis cholecalciferol (P ϭ 0.11) and paricalcitol (P ϭ 0.06) therapy when of monocytes demonstrated the emergence of a subpopulation compared with the previous time point, with the final postparicalcitol expression being significantly lower than the baseline of cells expressing very high levels of VDR (Figure 2A). To Ͻ characterize further this monocyte subpopulation, we per- expression (P 0.05). CYP24 expression increased with cholecalciferol (P ϭ 0.05) and decreased back to baseline after the dis- formed additional staining for the CD16 monocyte marker, continuation of this therapy and initiation of paricalcitol (P Ͻ which is expressed in a subset of monocytes exhibiting an in- 0.05), with no significant difference between postparicalcitol and flammatory phenotype. We found that this subpopulation ex- baseline CYP24 expression. The calculated Cyp27B1-to-CYP24 pressing high levels of VDR was strongly positive for CD16 ratio decreased significantly after 8 wk of cholecalciferol therapy (Figure 2B), whereas the larger population of monocytes dem- (P Ͻ 0.01), with no further change in this ratio after the restart of onstrating lesser degrees of VDR expression contained both paricalcitol therapy. (B) VDR expression increased significantly CD16ϩ and CD16Ϫ cells. Reinitiating paricalcitol treatment in after 25(OH)D repletion with cholecalciferol (P Ͻ 0.01) and 25(OH)D-replete patients resulted in further expansion of this seemed to increase further after the restart of paricalcitol (P ϭ CD16ϩ, high-VDR monocyte subpopulation (Figure 2A). 0.12). The final postparicalcitol VDR expression was significantly greater than the baseline expression (P ϭ 0.001). Total monocyte TLR2 expression seemed unchanged by both cholecalciferol and paricalcitol therapy, whereas cathelicidin expression seemed to Differential Expression of TLR2, Cathelicidin, and decrease with both of these therapies (NS), with the postparical- CYP24 in High- and Low-VDR Monocyte citol cathelicidin expression being significantly lower than the Subpopulations baseline expression (P Ͻ 0.05). After discovering the presence of two separate populations of monocytes in response to 25(OH)D repletion, one with very Cholecalciferol Therapy Alters Serum Inflammatory high VDR expression and one with lower VDR expression, we Cytokine Levels performed a subanalysis on TLR2, cathelicidin, and CYP24 To evaluate the impact of 25(OH)D repletion on the inflam- expression in each of these populations after cholecalciferol matory phenotype of these patients, we measured serum therapy. We found that TLR2, cathelicidin, and CYP24 expres- changes in inflammatory cytokines at three time points: Before sion were significantly higher in the high-VDR monocyte sub- cholecalciferol, after cholecalciferol and after paricalcitol re- set (Figure 3). start. We observed a decrease in IL-6, IL-8, and TNF-␣ after

J Am Soc Nephrol 21: 353–361, 2010 Vitamin D and Monocytes 355 CLINICAL RESEARCH www.jasn.org

Figure 3. High- and low-VDR monocyte subpopulations demonstrate unique levels of TLR2, cathelicidin, and CYP24 expression. Our previous finding of the existence of two distinct populations after cholecalciferol therapy prompted us to perform a subanalysis to evaluate the difference in TLR2, cathelicidin, and CYP24 expression within these two monocyte subsets. We found TLR2 expression to be five-fold greater in the high-VDR monocyte subset compared with the low-VDR monocytes (P Ͻ 0.001), whereas cathelicidin expression was two-fold greater (P Ͻ 0.01) and CYP24 expression was three-fold greater (NS) in the high- Figure 2. Flow cytometry analysis demonstrates the emergence VDR monocytes. of a subpopulation of monocytes with extremely high levels of VDR expression in response to cholecalciferol and paricalcitol therapy. (A) Frames 1 through 3 (left to right) illustrate flow ϩ cytometry analysis of isolated CD14 monocytes at three collection time points from two separate patients. Time point 1 (precholecalciferol) represents the postwashout collection before the initiation of cholecalciferol therapy. Time point 2 (postcholecalciferol) represents monocytes isolated after 8 wk of cholecalciferol therapy. Time point 3 (postparicalcitol) was taken 2 wk after stopping cholecalciferol and restarting paricalcitol. VDR expression by MFI is plotted on the y axis of each box, with CYP27B1 expression on the x axis. The large, lower population of monocytes (in blue) demonstrates lower levels of VDR expression when compared with the smaller population of cells outlined in the small box (in red), which demonstrates very high levels of VDR Figure 4. Change in serum IL-8, IL-6, and TNF-␣ levels after expression. After the administration of cholecalciferol, there was cholecalciferol and paricalcitol therapy. To assess the effects of a noticeable increase in the number of monocytes expressing cholecalciferol and paricalcitol therapies on the inflammatory high levels of VDR, with a further increase in this population after phenotype of patients with ESRD, we measured serum levels of paricalcitol restart. Similar findings were noted in other study several inflammatory cytokines that are generated by the mono- participants. (B) Further subanalysis of these two monocyte pop- cyte lineage. We observed a 55% decrease in serum IL-8 levels ulations demonstrating unique degrees of VDR upregulation re- (P ϭ 0.13), a 30% reduction in IL-6 levels (P ϭ 0.05), and a 60% vealed the population expressing high levels of VDR (in red) to be ␣ Ͻ ϩ decline in TNF- levels (P 0.05) after cholecalciferol therapy. strongly CD16 , whereas the population expressing lower levels ␣ ϩ Ϫ Serum IL-8 and TNF- levels remained lower after the discontin- of VDR (in blue) contained both CD16 and CD16 cells. uation of cholecalciferol and restart of paricalcitol, whereas the IL-6 levels seemed to return to near baseline values (NS, after 25(OH)D repletion with cholecalciferol (Figure 4). After the paricalcitol versus after cholecalciferol). discontinuation of cholecalciferol and reinitiation of paricalcitol therapy, serum levels of IL-8 and TNF-␣ remained low, pathways affecting immune function.10,16–21 In vitro and ex whereas IL-6 levels returned to levels near baseline. vivo studies have shown that cells in the monocyte lineage are exquisitely sensitive to vitamin D analogs through VDR-mediated pathways, and as terminal differentiation occurs toward DISCUSSION the macrophage phenotype, these cells exhibit an enhanced

ability to synthesize 1,25(OH)2D from the 25(OH)D precur- Monocytes are hypothesized to use locally produced sor.22–24 Whereas nutritional vitamin D deficiency has been

1,25(OH)2D to regulate vitamin D–responsive intracellular widely associated with disease processes affecting several vita-

356 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 353–361, 2010 www.jasn.org CLINICAL RESEARCH min D–responsive tissues in patients with normal renal func- still exhibited an activation of traditional degradation path- tion, the impact in patients with ESRD, who exhibit a dimin- ways when compared with baseline measurements (Figure ished capacity for renal 1,25(OH)2D production yet an intact 1A). The decline in the CYP27B1-to-CYP24 ratio is consistent 14,25–27 capacity for local production, remains ill-defined. With with the known effect of 1,25(OH)2D to elicit an increase in the 1,35 this study, we administered cholecalciferol (D3) to patients local production of CYP24, although the exact contribution with ESRD and 25(OH)D deficiency and evaluated in vivo of this enzyme to the local inactivation of vitamin D metabo- 36 changes in 1,25(OH)2D-responsive monocyte protein expres- lites in monocytes remains uncertain. The rise in VDR ex- sion. Whereas previous in vivo investigations of the nonclassi- pression after the administration of cholecalciferol suggests cal effects of vitamin D therapy in patients on hemodialysis that nutritional vitamin D therapy may indeed regulate VDR- have focused on defining monocyte responses to 1,25(OH)2D mediated cellular pathways. In addition, the trend toward a administration,28,29 to our knowledge, this is the first prospec- further rise in VDR expression with paricalcitol therapy after tive, in vivo investigation of the nonclassical, biologic effects of 25(OH)D repletion suggests additive effects of these therapies nutritional vitamin D (D2 or D3) therapy in patients with beyond that of each individual therapy alone. Because VDR ESRD. activation has been associated with a suppression of monocyte Because we examined patients who had ESRD and lacked TLR2 and an increase in cathelicidin expression in vitro,10,21,37 significant residual renal function, our a priori assumption was we examined whether cholecalciferol and paricalcitol therapy that this unique population would afford us the opportunity to led to similar alterations in these proteins in vivo. Increments study the effects of locally generated 1,25(OH)2D from mono- in VDR were not associated with changes in overall monocyte cytes in response to 25(OH)D repletion; however, we observed TLR2 expression, and cathelicidin expression was paradoxi- an unexpected rise in serum 1,25(OH)2D levels with 25(OH)D cally slightly decreased (Figure 1B). Thus, the in vivo response repletion in our patients (Table 1), a response previously de- of monocytes to vitamin D differs from the reported in vitro scribed by others.30–32 As a result, we are unable to distinguish responses of isolated monocytes, at least in patients with im- between changes in monocyte protein expression that are due paired renal function. As suggested by others, concurrent stim- to locally generated 1,25(OH)2D from possible effects of the ulation of TLR2 may be needed to stimulate cathelicidin in circulating hormone. The observed local upregulation of vivo.10,38 Consistent with our findings, a recent investigation of CYP24 in monocytes suggests that the peripheral conversion of patients with intact renal function found that circulating

25(OH)D to 1,25(OH)2D may not contribute to circulating 25(OH)D levels did not correlate with monocyte cathelicidin calcitriol levels. Alternatively, it is possible that the kidneys mRNA expression.38 possess residual CYP27B1 activity and, in the face of adequate The next significant observation in this investigation was an 25(OH)D substrate, are the source of the increased circulating effect of cholecalciferol therapy to affect preferentially specific

1,25(OH)2D in our patients with ESRD. Although such a the- monocyte subpopulations. In this regard, we found that ory may explain why a previous study by other investigators 25(OH)D repletion resulted in the emergence of two discrete found that nephrectomized rats lack peripheral conversion of monocyte subsets, one exhibiting very high levels of VDR and

25(OH)D to 1,25(OH)2D after the administration of radiola- the other demonstrating lower levels of VDR expression (Fig- beled 25(OH)D,33 it would not explain why serum ure 2A). Further characterization of these subpopulations re- ϩ 1,25(OH)2D levels rise in anephric patients after the adminis- vealed the high-VDR subset to be exclusively CD16 , whereas ϩ Ϫ tration of nutritional vitamin D.30 It is possible that the accu- the low-VDR subset consisted of both CD16 and CD16 rate measurement of serum 1,25(OH)2D levels in response to monocytes (Figure 2B). In addition, we observed both TLR2 cholecalciferol administration may be confounded by a lack of and cathelicidin expression to be significantly higher in the specificity of available clinical assays to assess biologically ac- high-VDR monocyte subset compared with the monocyte tive 1,25(OH)2D, which could explain why FGF-23 and PTH population expressing lower VDR levels (Figure 3). Because an levels were unchanged by 25(OH)D repletion in these patients. evaluation of monocyte CD16 expression was added to our Because calcitriol administration has been shown to increase protocol only after observing the emergence of the new high- FGF-23 production by bone in animal models,34 it was not VDR monocyte population that resulted from cholecalciferol unexpected that we observed a rise in mean FGF-23 levels in treatment, we were unable to determine whether activation of patients after paricalcitol administration (P ϭ 0.08). the monocyte VDR resulted in a differentiation of CD16Ϫ Our second important finding is that nutritional vitamin D monocytes into a CD16ϩ phenotype or existing CD16ϩ mono-

(D3) alters the expression of vitamin D–regulatory proteins in cytes simply demonstrated a greater degree of VDR upregula- circulating monocytes of patients with ESRD. In this regard, tion after 25(OH)D repletion. Because patients with ESRD are 25(OH)D repletion was accompanied by increments in CYP24 known to exhibit higher levels of circulating CD16ϩ mono- (Figure 1A) and VDR expression (Figure 1B) but resulted in a cytes,39,40 which have been hypothesized to contribute to the suppression of CYP27B1 expression (Figure 1A). Interestingly, pathogenesis of cardiovascular disease and other inflammatory CYP24 expression decreased back to baseline after the discon- comorbidities in this setting,41–45 it will be important to define tinuation of cholecalciferol therapy, yet the ratio of CYP27B1 further the downstream functional changes that result from to CYP24 expression remained low, suggesting that monocytes the observed VDR upregulation in the CD16ϩ monocyte sub-

J Am Soc Nephrol 21: 353–361, 2010 Vitamin D and Monocytes 357 CLINICAL RESEARCH www.jasn.org population with cholecalciferol therapy. The future character- Taken together, our findings provide in vivo evidence that ization of changes in VDR-dependent pathways with this ther- 25(OH)D repletion exerts biologic effects on circulating apy may allow us to understand better the emerging monocytes and may alter the inflammatory phenotype of pa- association between 25(OH)D deficiency and adverse cardio- tients with ESRD. In addition, our data suggest that specific vascular outcomes in patients with ESRD.46,47 monocyte subpopulations may exhibit unique responses to Finally, 25(OH)D repletion caused a decline in serum in- this therapy. Our study also provides evidence for a separate flammatory cytokine levels, including IL-8, IL-6, and TNF-␣ and possibly additive effect of cholecalciferol and paricalcitol (Figure 4). We observed an approximate 55, 30, and 60% re- therapy on VDR-responsive cellular pathways, thereby poten- duction in serum IL-8, IL-6, and TNF-␣ levels, respectively, tially allowing reduced administration of more expensive cal- after cholecalciferol therapy. Both IL-8 and TNF-␣ levels re- citriol analogues to activate nonclassical VDR-dependent mained suppressed after discontinuation of cholecalciferol functions. Future comparative testing of the separate in vivo and restart of paricalcitol, whereas IL-6 levels seemed to in- responses of monocyte subsets to cholecalciferol versus calcit- crease back to the baseline levels. Alterations in monocytes and riol analogue therapy, however, is needed to justify the incor- monocyte-derived cytokines have been implicated in the in- poration of 25(OH)D repletion into treatment recommenda- flammatory pathology that exists in ESRD.39,41,44,48,49 IL-8, tions for patients with ESRD. As our therapeutic options IL-6, and TNF-␣ are possible direct mediators in the patho- advance for the treatment of secondary hyperparathyroidism genesis of atherosclerosis.50–58 In addition, there seems to be a in patients with ESRD, it is imperative to develop studies that significant correlation between the level of these various cyto- compare the nonclassical, biologic effects of nutritional vita- kines and patient outcomes, with multiple studies suggesting min D to calcitriol analogue therapy for the treatment of vita- that elevations in IL-8, IL-6, and TNF-␣ levels are associated min D deficiency associated with progressive kidney disease. with increased morbidity and mortality in patients with ESRD.49,59–62 Regardless, the observed reductions in serum cytokine profiles that occurred with 25(OH)D repletion in these CONCISE METHODS patients further support our hypothesis that the correction of nutritional vitamin D deficiency may improve the inflamma- Study Design tory phenotype of patients with ESRD through nonclassical In a prospective study involving seven patients who had nutritional effects on circulating monocytes and possibly other tissues. vitamin D insufficiency and ESRD and were on hemodialysis, we in- We acknowledge that there are several limitations to this vestigated changes in the expression of several vitamin D–responsive investigation. First, the small number of patients and short- proteins within circulating monocytes after 25(OH)D repletion with term nature of this study make it difficult to draw definitive cholecalciferol (Biotech Pharmacal, Fayetteville, AR) 50,000 U twice conclusions regarding benefit or harm of this therapy. It is per week. After initial screening of medical history and current health, possible that 25(OH)D deficiency in the setting of ESRD rep- patients were enrolled according to predetermined inclusion and ex- resents an adaptive response to prevent unwanted physiologic clusion criteria. Inclusion criteria were long-term hemodialysis treat- changes that occur with failing kidneys and that repletion of ment at University of Kansas Medical Center outpatient dialysis unit 25(OH)D levels could have detrimental effects for these pa- three times per week, dialysis vintage Ͼ6 mo, and 25(OH)D level Ͻ25 tients. Prospective, long-term studies focusing on the func- ng/ml. Exclusion criteria were active infection, acute illness within last tional changes in monocytes, progression of comorbidities, month, intact PTH Ͼ600 pg/ml, history of chronic inflammatory and outcomes measures are needed for further assessment of disease (e.g., inflammatory bowel disease, lupus, rheumatoid arthri- the potential impact of this therapy. Likewise, we acknowledge tis), cinacalcet therapy, allergy to cholecalciferol or paricalcitol, pre- that acute changes in cell function may not translate to chronic vious parathyroidectomy, previous renal transplantation, active treat- alterations within these cell populations. It is quite possible ment with immunosuppressant medications, and noncompliance that the changes in monocyte protein expression and serum with hemodialysis prescription. After consent was obtained, patients cytokine levels that were witnessed in this study were a re- underwent a washout period from paricalcitol for 4 wk. After wash- sponse to the acute administration of cholecalciferol and pari- out, patients had the first of three blood draws collected for monocyte calcitol, which may differ substantially from the long-term ad- isolation and subsequent measurement of monocyte VDR, CYP27B1, ministration of vitamin D derivatives. Furthermore, it remains CYP24, cathelicidin, and TLR2 expression by flow cytometry. In ad- unclear whether 25(OH)D repletion has any salutary effects at dition to the monocyte analysis, patients had serum samples collected the local cellular level that would extend beyond the effects of for measurement of cytokine levels and markers of mineral metabo- systemic repletion of 1,25(OH)2D with calcitriol analogue lism, including 25(OH)D, 1,25(OH)2D, phosphorus, calcium, FGF- therapies. Prospective analyses comparing long-term func- 23, and PTH levels before the initiation of cholecalciferol therapy. tional changes in monocyte subpopulations in response to nu- Patients were subsequently started on cholecalciferol 50,000 U twice tritional vitamin D versus calcitriol analogue therapy or com- weekly. 25(OH)D levels were monitored at 3 and 6 wk of therapy, and bination therapy with these two agents would provide dosing adjustments were made according to these levels. At the 3-wk important information to help guide future therapeutic strat- time point, patients who experienced a slow correction of 25(OH)D egies for vitamin D replacement in the setting of ESRD. levels (as defined by levels Ͻ30 ng/ml) had their cholecalciferol dos-

358 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 353–361, 2010 www.jasn.org CLINICAL RESEARCH ing increased to 50,000 U thrice weekly, whereas patients who dem- sc-66851, and sc-21578; Santa Cruz Biotechnology, Santa Cruz, CA) onstrated a more rapid correction (Ͼ45 ng/ml at 3 wk) had their and TLR2 (cat. no. 56-9922-73; eBioscience, San Diego, CA). The cholecalciferol dosing decreased to 50,000 U once weekly. At the 6-wk VDR, CYP27b1, CYP24, and cathelicidin antibodies were labeled ac- time point, 25(OH)D levels were again reevaluated and cholecalcif- cording to package directions using Zenon antibody labeling kit (cat. erol dosing was adjusted on the basis of the following parameters: nos. Z25113, Z25608, Z25341, and Z25602; Invitrogen, Carlsbad, Patients who had levels Ͻ45 ng/ml at 6 wk had their dosing increased CA). Labeled antibodies were applied immediately to the monocytes to 50,000 U thrice weekly, whereas patients exhibiting a more rapid and read by flow cytometry (LSRII; BD Biosciences, Diva6 software). correction (levels Ͼ60 ng/ml) had their cholecalciferol dosing de- CD16 staining was performed on a separate aliquot of cells using 20 ␮l creased to 50,000 U once weekly. After a total of 8 wk of cholecalciferol of FITC-labeled CD16 antibody (cat. no. 55407; BD Biosciences) therapy, the second blood draw was taken for monocyte isolation and added after the CD14 labeling step outlined already. processing, as well as the identical serum measurements that were performed before the beginning of treatment. At this point, cholecal- Flow Cytometric Analysis ciferol therapy was stopped and patients were placed back on their Monocytes labeled with individual antibodies were used for initial original paricalcitol dosing regimen. A third and final monocyte draw LSRII setup, and fluorescence minus one protocol was used to estab- was obtained 2 wk after the restart of paricalcitol therapy, and serum lish gates for monocyte analysis.63 Gating strategy, voltages, and com- measurements were once again performed. This study was approved pensation remained constant throughout the experiment. Zenon kits by the institutional review board of the University of Kansas Medical demonstrated consistent results in consecutive labeling for the exper- Center, and all participants provided written consent before study iment duration. enrollment. Serum Analysis Monocyte Isolation For serum samples, 4 ml of whole blood was collected in serum iso- For monocyte isolation, 30 ml of whole blood was collected in phle- lation tubes and centrifuged at 3700 rpm for 7 min. Serum calcium botomy tubes containing EDTA anticoagulant and subsequently pro- and phosphorus were measured by a Unicel DxC 800 (Beckman cessed using Ficoll-Paque (Stem Cell Technologies, Vancouver, Brit- Coulter). Intact serum PTH was measured by human intact PTH ish Columbia, Canada) centrifugation to obtain the lymphocyte ELISA kit (Calbiotech, Spring Valley, CA). 25(OH)D and fraction. Ficoll-Paque centrifugation was performed by first mixing 1,25(OH)2D levels were measured by RIA per kit instructions (Dia- 15 ml of whole blood with 15 ml of 1ϫ PBS supplemented with 2% sorin, Saluggia, Italy). Serum FGF-23 levels were measured by an in- FBS, then layering this mixture over 15 ml of Ficoll-Paque solution in tact FGF-23 ELISA (Kainos, Tokyo, Japan). IL-8, IL-6, and TNF-␣ a 50-ml conical bottom tube. Next, this mixture was centrifuged at levels were measured in duplicate by cytokine bead array per package 400 ϫ g for 35 min at 20°C. After centrifugation, the middle lympho- instructions (cat. no. 551811; BD Biosciences). cyte layer was transferred to a new conical bottom tube. This layer was diluted with PBS supplemented with FBS, vortexed to remove any cell Statistical Analysis aggregates, and centrifuged at 300 ϫ g for 10 min. The supernatant We evaluated differences between groups by one-way ANOVA for was removed, and cells were suspended in PBS with FBS and centri- multiple group comparisons and t test for two-group comparisons. fuged at 200 ϫ g for 10 min. This step was repeated one additional All values are expressed as means Ϯ SD. P Ͻ 0.05 was considered time to help remove platelets and any remaining Ficoll-Paque solu- statistically significant. All computations were performed using the tion from the lymphocyte sample. The final supernatant was re- GraphPad Prism 5 software (GraphPad Software, San Diego, CA). moved, and cells were resuspended in plain 1ϫ PBS. Isolated cells were further processed by negative selection for the CD14 cell surface marker using a labeled magnetic bead kit (Stem Cell Technologies) to ACKNOWLEDGMENTS isolate the monocyte fraction. This work was supported by the University of Kansas Medical Center Antibody Labeling of Target Proteins GCRC grant M01 RR 023940 National Center for Research Resourc- After an unstained fraction of monocytes was reserved in a separate es/National Institutes of Health and the University of Kansas Medical tube, 10 ␮l of PECy7-labeled CD14 antibody (cat. no. 557907; BD Center Max Allen Clinical Scholar Award. Biosciences, San Jose, CA) was added to the remaining monocytes and incubated for 15 min in the dark at 4°C. Cells were washed with 1ϫ PBS and centrifuged at 1000 rpm for 3 min. The supernatant was DISCLOSURES discarded, and 500 ␮l of Cytofix/Cytoperm solution (BD Biosciences) None. was added to the cells and incubated for 20 min at 4°C. After incuba- tion, cells were washed twice with 1 ml of Perm/Wash buffer (BD Biosciences). After final centrifugation, the supernatant was again dis- REFERENCES carded and the final pellet was resuspended in 100 ␮lof1ϫ PBS. The monocytes were then subdivided for staining with antibodies specific 1. Jones G, Strugnell SA, DeLuca HF: Current understanding of the for VDR, CYP27b1, CYP24, cathelicidin (cat. nos. sc-13133, sc-49642, molecular actions of vitamin D. Physiol Rev 78: 1193–1231, 1998

J Am Soc Nephrol 21: 353–361, 2010 Vitamin D and Monocytes 359 CLINICAL RESEARCH www.jasn.org

2. Zisman AL, Hristova M, Ho LT, Sprague SM: Impact of ergocalciferol 20. Zlatkina TI, Podgorodnikova NA: Evaluation of vitamin D deficiency in treatment of vitamin D deficiency on serum parathyroid hormone pulmonary tuberculosis [in Russian]. Probl Tuberk 51: 46–49, 1973 concentrations in chronic kidney disease. Am J Nephrol 27: 36–43, 21. Sadeghi K, Wessner B, Laggner U, Ploder M, Tamandl D, Friedl J, 2007 Zugel U, Steinmeyer A, Pollak A, Roth E, Boltz-Nitulescu G, Spittler A: 3. Al-Aly Z, Qazi RA, Gonzalez EA, Zeringue A, Martin KJ: Changes in Vitamin D3 down-regulates monocyte TLR expression and triggers serum 25-hydroxyvitamin D and plasma intact PTH levels following hyporesponsiveness to pathogen-associated molecular patterns. Eur treatment with ergocalciferol in patients with CKD. Am J Kidney Dis J Immunol 36: 361–370, 2006 50: 59–68, 2007 22. Kreutz M, Andreesen R, Krause SW, Szabo A, Ritz E, Reichel H: 4. Saab G, Young DO, Gincherman Y, Giles K, Norwood K, Coyne DW: 1,25-Dihydroxyvitamin D3 production and vitamin D3 receptor ex- Prevalence of vitamin D deficiency and the safety and effectiveness of pression are developmentally regulated during differentiation of hu- monthly ergocalciferol in hemodialysis patients. Nephron Clin Pract man monocytes into macrophages. Blood 82: 1300–1307, 1993 105: c132–c138, 2007 23. Kreutz M, Andreesen R: Induction of human monocyte to macrophage 5. Briese S, Wiesner S, Will JC, Lembcke A, Opgen-Rhein B, Nissel R, maturation in vitro by 1,25-dihydroxyvitamin D3. Blood 76: 2457– Wernecke KD, Andreae J, Haffner D, Querfeld U: Arterial and cardiac 2461, 1990 disease in young adults with childhood-onset end-stage renal disease: 24. Hewison M, Freeman L, Hughes SV, Evans KN, Bland R, Eliopoulos Impact of calcium and vitamin D therapy. Nephrol Dial Transplant 21: AG, Kilby MD, Moss PA, Chakraverty R: Differential regulation of 1906–1914, 2006 vitamin D receptor and its ligand in human monocyte-derived den- 6. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart dritic cells. J Immunol 170: 5382–5390, 2003 PM, Hewison M: Extrarenal expression of 25-hydroxyvitamin d(3)-1 25. Hypponen E, Laara E, Reunanen A, Jarvelin MR, Virtanen SM: Intake of alpha-hydroxylase. J Clin Endocrinol Metab 86: 888–894, 2001 vitamin D and risk of type 1 diabetes: A birth-cohort study. Lancet 358: 7. Dusso A, Brown A, Slatopolsky E: Extrarenal production of calcitriol. 1500–1503, 2001 Semin Nephrol 14: 144–155, 1994 26. Kristal-Boneh E, Froom P, Harari G, Ribak J: Association of calcitriol 8. Gorham ED, Garland CF, Garland FC, Grant WB, Mohr SB, Lipkin M, and blood pressure in normotensive men. Hypertension 30: 1289– Newmark HL, Giovannucci E, Wei M, Holick MF: Vitamin D and pre- 1294, 1997 vention of colorectal cancer. J Steroid Biochem Mol Biol 97: 179–194, 27. Holick MF: Calcium plus vitamin D and the risk of colorectal cancer. 2005 N Engl J Med 354: 2287–2288, author reply 2287–2288, 2006 9. Stead WW, Senner JW, Reddick WT, Lofgren JP: Racial differences in 28. Hubel E, Kiefer T, Weber J, Mettang T, Kuhlmann U: In vivo effect of susceptibility to infection by Mycobacterium tuberculosis. N Engl 1,25-dihydroxyvitamin D3 on phagocyte function in hemodialysis pa- J Med 322: 422–427, 1990 tients. Kidney Int 40: 927–933, 1991 10. Liu PT, Stenger S, Li H, Wenzel L, Tan BH, Krutzik SR, Ochoa MT, 29. Tokuda N, Kano M, Meiri H, Nomoto K, Naito S: Calcitriol therapy Schauber J, Wu K, Meinken C, Kamen DL, Wagner M, Bals R, Stein- modulates the cellular immune responses in hemodialysis patients. meyer A, Zugel U, Gallo RL, Eisenberg D, Hewison M, Hollis BW, Am J Nephrol 20: 129–137, 2000 Adams JS, Bloom BR, Modlin RL: Toll-like receptor triggering of a 30. Lambert PW, Stern PH, Avioli RC, Brackett NC, Turner RT, Greene A, vitamin D-mediated human antimicrobial response. Science 311: Fu IY, Bell NH: Evidence for extrarenal production of 1 alpha,25- 1770–1773, 2006 dihydroxyvitamin D in man. J Clin Invest 69: 722–725, 1982 11. Boonstra A, Barrat FJ, Crain C, Heath VL, Savelkoul HF, O’Garra A: 31. Dusso A, Lopez-Hilker S, Rapp N, Slatopolsky E: Extra-renal produc- 1alpha,25-Dihydroxyvitamin d3 has a direct effect on naive CD4(ϩ)T tion of calcitriol in chronic renal failure. Kidney Int 34: 368–375, 1988 cells to enhance the development of Th2 cells. J Immunol 167: 4974– 32. Lund B, Clausen E, Friedberg M, Lund B, Moszkowicz M, Nielsen SP, 4980, 2001 Sorensen OH: Serum 1,25-dihydroxycholecalciferol in anephric, hae- 12. Garland FC, Garland CF, Gorham ED, Young JF: Geographic variation modialyzed and kidney-transplanted patients: Effect of vitamin D3 in breast cancer mortality in the United States: A hypothesis involving supplement. Nephron 25: 30–33, 1980 exposure to solar radiation. Prev Med 19: 614–622, 1990 33. Shultz TD, Fox J, Heath H 3rd, Kumar R: Do tissues other than the 13. Giulietti A, van Etten E, Overbergh L, Stoffels K, Bouillon R, Mathieu kidney produce 1,25-dihydroxyvitamin D3 in vivo? A reexamination. C: Monocytes from type 2 diabetic patients have a pro-inflammatory Proc Natl Acad Sci U S A 80: 1746–1750, 1983 profile: 1,25-Dihydroxyvitamin D(3) works as anti-inflammatory. Dia- 34. Liu S, Tang W, Zhou J, Stubbs JR, Luo Q, Pi M, Quarles LD: Fibroblast betes Res Clin Pract 77: 47–57, 2007 growth factor 23 is a counter-regulatory phosphaturic hormone for 14. Giovannucci E, Liu Y, Rimm EB, Hollis BW, Fuchs CS, Stampfer MJ, vitamin D. J Am Soc Nephrol 17: 1305–1315, 2006 Willett WC: Prospective study of predictors of vitamin D status and 35. Monkawa T, Yoshida T, Hayashi M, Saruta T: Identification of 25- cancer incidence and mortality in men. J Natl Cancer Inst 98: 451– hydroxyvitamin D3 1alpha-hydroxylase gene expression in macro- 459, 2006 phages. Kidney Int 58: 559–568, 2000 15. Zittermann A, Schleithoff SS, Tenderich G, Berthold HK, Korfer R, 36. Ren S, Nguyen L, Wu S, Encinas C, Adams JS, Hewison M: Alternative Stehle P: Low vitamin D status: A contributing factor in the patho- splicing of vitamin D-24-hydroxylase: A novel mechanism for the reg- genesis of congestive heart failure? J Am Coll Cardiol 41: 105–112, ulation of extrarenal 1,25-dihydroxyvitamin D synthesis. J Biol Chem 2003 280: 20604–20611, 2005 16. Davies PD, Brown RC, Woodhead JS: Serum concentrations of vitamin 37. Do JE, Kwon SY, Park S, Lee ES: Effects of vitamin D on expression of D metabolites in untreated tuberculosis. Thorax 40: 187–190, 1985 Toll-like receptors of monocytes from patients with Behcet’s disease. 17. Pramanik R, Asplin JR, Lindeman C, Favus MJ, Bai S, Coe FL: Lipo- Rheumatology (Oxford) 47: 840–848, 2008 polysaccharide negatively modulates vitamin D action by down-reg- 38. Adams JS, Ren S, Liu PT, Chun RF, Lagishetty V, Gombart AF, Borre- ulating expression of vitamin D-induced VDR in human monocytic gaard N, Modlin RL, Hewison M: Vitamin d-directed rheostatic regu- THP-1 cells. Cell Immunol 232: 137–143, 2004 lation of monocyte antibacterial responses. J Immunol 182: 4289– 18. Narang NK, Gupta RC, Jain MK: Role of vitamin D in pulmonary 4295, 2009 tuberculosis. J Assoc Physicians India 32: 185–188, 1984 39. Heine GH, Ulrich C, Seibert E, Seiler S, Marell J, Reichart B, Krause M, 19. Fritsche J, Mondal K, Ehrnsperger A, Andreesen R, Kreutz M: Regu- Schlitt A, Kohler H, Girndt M: CD14(ϩϩ)CD16ϩ monocytes but not lation of 25-hydroxyvitamin D3–1 alpha-hydroxylase and production total monocyte numbers predict cardiovascular events in dialysis pa- of 1 alpha,25-dihydroxyvitamin D3 by human dendritic cells. Blood tients. Kidney Int 73: 622–629, 2008 102: 3314–3316, 2003 40. Brauner A, Lu Y, Hallden G, Hylander B, Lundahl J: Difference in the

360 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 353–361, 2010 www.jasn.org CLINICAL RESEARCH

blood monocyte phenotype between uremic patients and healthy 52. Okada M, Matsumori A, Ono K, Furukawa Y, Shioi T, Iwasaki A, controls: Its relation to monocyte differentiation into macrophages in Matsushima K, Sasayama S: Cyclic stretch upregulates production of the peritoneal cavity. Inflammation 22: 55–66, 1998 interleukin-8 and monocyte chemotactic and activating factor/mono- 41. Merino A, Nogueras S, Buendia P, Ojeda R, Carracedo J, Ramirez- cyte chemoattractant protein-1 in human endothelial cells. Arterio- Chamond R, Martin-Malo A, Aljama P: Microinflammation and endo- scler Thromb Vasc Biol 18: 894–901, 1998 thelial damage in hemodialysis. Contrib Nephrol 161: 83–88, 2008 53. Simonini A, Moscucci M, Muller DW, Bates ER, Pagani FD, Burdick 42. Nockher WA, Scherberich JE: Expanded CD14ϩ CD16ϩ monocyte MD, Strieter RM: IL-8 is an angiogenic factor in human coronary subpopulation in patients with acute and chronic infections undergo- atherectomy tissue. Circulation 101: 1519–1526, 2000 ing hemodialysis. Infect Immun 66: 2782–2790, 1998 54. Huber SA, Sakkinen P, Conze D, Hardin N, Tracy R: Interleukin-6 43. Ziegler-Heitbrock L: The CD14ϩ CD16ϩ blood monocytes: Their role exacerbates early atherosclerosis in mice. Arterioscler Thromb Vasc in infection and inflammation. J Leukoc Biol 81: 584–592, 2007 Biol 19: 2364–2367, 1999 44. Ramirez R, Carracedo J, Berdud I, Carretero D, Merino A, Rodriguez 55. Venugopal SK, Devaraj S, Yuhanna I, Shaul P, Jialal I: Demonstration M, Tetta C, Martin-Malo A, Aljama P: Microinflammation in hemodi- that C-reactive protein decreases eNOS expression and bioactivity in alysis is related to a preactivated subset of monocytes. Hemodial Int human aortic endothelial cells. Circulation 106: 1439–1441, 2002 10[Suppl 1]: S24–S27, 2006 56. Shioi A, Katagi M, Okuno Y, Mori K, Jono S, Koyama H, Nishizawa Y: 45. Ulrich C, Heine GH, Gerhart MK, Kohler H, Girndt M: Proinflammatory Induction of bone-type alkaline phosphatase in human vascular CD14ϩCD16ϩ monocytes are associated with subclinical atheroscle- smooth muscle cells: Roles of tumor necrosis factor-alpha and on- rosis in renal transplant patients. Am J Transplant 8: 103–110, 2008 costatin M derived from macrophages. Circ Res 91: 9–16, 2002 46. Matias PJ, Ferreira C, Jorge C, Borges M, Aires I, Amaral T, Gil C, 57. Tintut Y, Patel J, Parhami F, Demer LL: Tumor necrosis factor-alpha Cortez J, Ferreira A: 25-Hydroxyvitamin D3, arterial calcifications and promotes in vitro calcification of vascular cells via the cAMP pathway. cardiovascular risk markers in haemodialysis patients. Nephrol Dial Circulation 102: 2636–2642, 2000 Transplant 24: 611–618, 2009 58. Al-Aly Z, Shao JS, Lai CF, Huang E, Cai J, Behrmann A, Cheng SL, 47. London GM, Guerin AP, Verbeke FH, Pannier B, Boutouyrie P, Mar- Towler DA: Aortic Msx2-Wnt calcification cascade is regulated by chais SJ, Metivier F: Mineral metabolism and arterial functions in TNF-alpha-dependent signals in diabetic LdlrϪ/Ϫ mice. Arterioscler end-stage renal disease: Potential role of 25-hydroxyvitamin D defi- Thromb Vasc Biol 27: 2589–2596, 2007 ciency. J Am Soc Nephrol 18: 613–620, 2007 59. Panichi V, Maggiore U, Taccola D, Migliori M, Rizza GM, Consani C, 48. Stenvinkel P, Ketteler M, Johnson RJ, Lindholm B, Pecoits-Filho R, Bertini A, Sposini S, Perez-Garcia R, Rindi P, Palla R, Tetta C: Interleu- Riella M, Heimburger O, Cederholm T, Girndt M: IL-10, IL-6, and kin-6 is a stronger predictor of total and cardiovascular mortality than TNF-alpha: Central factors in the altered cytokine network of uremia: C-reactive protein in haemodialysis patients. Nephrol Dial Transplant The good, the bad, and the ugly. Kidney Int 67: 1216–1233, 2005 19: 1154–1160, 2004 49. Panichi V, Taccola D, Rizza GM, Consani C, Ghiadoni L, Filippi C, 60. Pecoits-Filho R, Barany P, Lindholm B, Heimburger O, Stenvinkel P: Cristofani R, Panicucci E, Migliori M, Sidoti A, Biagioli M, Boracelli D, Interleukin-6 is an independent predictor of mortality in patients start- Barsotti G, Tetta C: Interleukin-8 is a powerful prognostic predictor of ing dialysis treatment. Nephrol Dial Transplant 17: 1684–1688, 2002 all-cause and cardiovascular mortality in dialytic patients. Nephron 61. Kimmel PL, Phillips TM, Simmens SJ, Peterson RA, Weihs KL, Alleyne Clin Pract 102: c51–c58, 2006 S, Cruz I, Yanovski JA, Veis JH: Immunologic function and survival in 50. Wang N, Tabas I, Winchester R, Ravalli S, Rabbani LE, Tall A: Inter- hemodialysis patients. Kidney Int 54: 236–244, 1998 leukin 8 is induced by cholesterol loading of macrophages and ex- 62. Kalantar-Zadeh K, Kopple JD, Humphreys MH, Block G: Comparing pressed by macrophage foam cells in human atheroma. J Biol Chem outcome predictability of markers of malnutrition-inflammation com- 271: 8837–8842, 1996 plex syndrome in haemodialysis patients. Nephrol Dial Transplant 19: 51. Wang JM, Sica A, Peri G, Walter S, Padura IM, Libby P, Ceska M, 1507–1519, 2004 Lindley I, Colotta F, Mantovani A: Expression of monocyte chemotac- 63. Perfetto SP, Chattopadhyay PK, Roederer M: Seventeen-colour flow tic protein and interleukin-8 by cytokine-activated human vascular cytometry: Unravelling the immune system. Nat Rev Immunol 4: 648– smooth muscle cells. Arterioscler Thromb 11: 1166–1174, 1991 655, 2004

J Am Soc Nephrol 21: 353–361, 2010 Vitamin D and Monocytes 361