Hypobaric hypoxia in 3000 m altitude leads to a significant decrease in circulating plasmacytoid dendritic cells in humans

Abstract

INTRODUCTION: Hypoxia is known to affect the immune system. It leads to an increase in pro-inflammatory cytokines such as interleukin-6 and influences the number of different inflammatory cells. This study investigates the effect of hypoxia on the number of different subsets of circulating human dendritic cells (DCs) as professional antigen-presenting cells.

METHODS: The number of circulating DCs was determined via Fluorescence activated cell sorting analysis in peripheral blood of 17 healthy volunteers (age 35.9±2.6 years) in normoxia (baseline, BL), hypoxia (altitude 3000 m, alpine passive escalation), and again normoxia (follow-up, FU).

RESULTS: Exposure to hypobaric hypoxia in high altitude, 3000 m, led to a significant decrease in the participants’ oxygen saturation, and an increase in the breathing frequency whereas blood pressure and heart rate were not significantly altered. FACS analysis revealed a significant hypoxia induced decrease in circulating plasmacytoid (p) DCs compared to baseline levels (BL: 0.10 [0.08–0.18] % of white blood cell count (WBC), 3000 m: 0.03 [0.02–0.06] % WBC, p < 0.001). During follow up, again a significant reconstitution of circulating pDCs was observed (FU: 0.16±[0.11–0.26] % WBC, p = 0.0013).

CONCLUSION: Hypobaric hypoxia caused by exposure to altitude results in a significant reduction in the number of circulating pDCs.

Keywords

Dendritic cells inflammation hypobaric hypoxia altitude

1 Introduction

Systemic hypoxia is known to induce alterations in many different physiologic functions in humans, such as respiration, metabolism, sympathetic nervous system, and immune system [11 , 29]. Hypoxia is a key trigger in many different diseases. Global hypoxia for example is the key factor in high altitude illness. But additionally, there are many diseases where tissue hypoxia is in the focus, such as in acute myocardial infarction, stroke, but also different types of shock. Moreover, besides its role in diseases, hypoxia also has a key role in training conditions such as altitude training for a growing number of athletes. In all of these diseases and conditions, inflammatory as well as well as hemorheological properties play important roles [2 , 21]. Whereas certain effects like the increase in erythropoiesis have already been well investigated, the effect of hypoxia on inflammation has only been scarcely understood so far. In patients with acute mountain sickness, serum levels of pro-inflammatory cytokines such as interleukin-6 have been shown to be increased. This effect is suggested to be responsible for the development of an increased capillary permeability that results in pulmonary and cerebral edema [6]. Multiple studies have also demonstrated elevated C-reactive protein, interleukin-6, and interleukin-8 levels as markers of inflammation and reduced interleukin-10 levels as an anti-inflammatory cytokine in healthy volunteers in high altitude [7 , 20]. Increased inflammatory cell infiltrations and elevated serum cytokine levels could also be observed in mouse-models after exposure to hypoxia [4, 26]. Moreover, hypoxia is also capable of affecting the cellular immunity. Low oxygen pressure results in an increase in the concentration of natural killer cells [9]. Moreover, hypoxia leads to a significant relative lymphopenia and neutrophilia [25]. In parallel to affecting the levels of immune cells in blood, hypoxia influences their cellular functions, too. Hypoxia was shown to cause an impaired T cell activity whereas B cell function remains unaffected [12].

Hypoxia also affects dendritic cells (DCs) [29]. DCs are important antigen-presenting cells linking both the innate as well as the adaptive immunity [16]. There are two major subpopulations of DCs: plasmacytoid (pDCs) and myeloid (mDCs) dendritic cells. PDCs are mainly involved in combating viral infections, mDCs are involved in bacterial and fungal infections [16]. After antigen-uptake DCs serve as activators of T cells. In vitro, low oxygen tension was shown to result in a different DC chemokine expression profile, which predicts a pro-inflammatory and pro-angiogenic environment through a change in the recruited immune cells [3]. Cell culture experiments with immature DCs could also demonstrate a hypoxia-induced increase in the cellular ability to activate T cells [17, 18]. So far, one publication by our group investigating the effect of hypoxia on the number of circulating dendritic cells and their subpopulations in vivo has been published [29]. This study described a reduction in the number of circulating plasmacytoid DCs during hypoxia. However, in this study hypoxia was achieved through the use of a hypoxic chamber replacing oxygen by nitrogen resembling an oxygen concentration equivalent to an altitude of 5500 m. This experimental setting resulted in normobaric hypoxia. In the present study, we aimed at investigating whether this effect could also be observed under “real-life” conditions of hypobaric hypoxia in an alpine setting. This additional study was performed to investigate whether only normobaric or also hypobaric hypoxia results in an increase in systemic inflammation which might be of relevance especially for athletes making a decision for either one of these training conditions in a hypoxic chamber versus “real-life” altitude training. Thus, in this study 17 healthy volunteers were exposed to hypobaric hypoxia in an altitude of 3000 m and the levels of circulating dendritic cells were analyzed and compared for hypoxia in high altitude and normoxia.

2 Materials and methods

2.1 Study protocol & subjects

This study exposed 17 apparently healthy volunteers (14 males, 3 females, age 35.9±2.6 years, height 178.3±1.9 cm, weight 77.6±3.9 kg) to hypobaric hypoxia for 3 days during a research excursion up to 3000 m altitude in Diavolezza, Switzerland. The study was approved by the local ethics committee of the Friedrich Schiller University, Jena, Germany. All study participants were recruited from the medical staff of the Friedrich Schiller University, Jena, Germany, and all gave their written informed consent. Inclusion criteria were absence of a chronic inflammatory disease or an acute infection during the past 4 weeks. Blood was withdrawn and vital parameters were recorded in normoxia at baseline (BL in Jena, 215 m), in hypoxia in 3000 m altitude (3000 m, blood withdrawal after 48 hours; exact location: Diavolezza, 7504 Pontresina, Switzerland, altitude 2978 m, GPS: N46.44089 E9.98350), and again in normoxia 48 hours after return to Jena (follow up, FU) (Table 1). No participant fulfilled the drop-out criteria (oxygen saturation <70%, symptoms of high altitude diseases). The study participants reached Diavolezza via the Diavolezza aerial cableway and without physical efforts. They did not consume alcohol during the first 48 hours.

2.2 Vital parameters

Peripheral oxygen saturation was measured by finger pulse oximetry (Masimo Radical-7, Masimo Corp., Calif. USA). Blood pressure, heart rate, and breathing frequency were determined manually.

2.3 Determination of the number of circulating DCs, DC subsets, IL-6, and other laboratory parameters

The number of circulating DCS and their subsets (mDCs, pDCs) were determined via fluorescence activated cell sorting (FACS) analysis using the Blood Dendritic Cell Enumeration kit (BDCA kit, Miltenyi Biotec) and BD Accuri C6 (BD Biosciences). This flow cytometer is a fully mobile device that has been set up at the mountain for these experiments in Diavolezza. The methods as well as the gating strategies have been formerly described [30] and are demonstrated in Fig. 1. Briefly, blood was obtained from the study’s participants and collected in EDTA tubes, cooled (–8°C) and processed within 12 hours. 300 μL of these blood samples were mixed with 20 μL of the anti-BDCA cocktail for DCP, another 300 μL with 20 μL of the control cocktail for isotype control respectively. Then incubation with 10 μL of a dead cell discriminator for 10 minutes under a 60 W light bulb followed before fixation process was initiated. Red blood cell lysis solution was used to remove erythrocytes. Using this method, the relative cell numbers of mDCs, and pDCs in relation to the number of white blood cells (% WBC) were detected. The absolute cell numbers were determined by multiplication of this result by the white blood cell count. For this reason, absolute cell numbers only exist for the baseline and follow-up measurements, because in hypoxia in the altitude of 3000 m no determination of the white blood cell number was possible.

Serum concentration of IL-6 was determined using the electrochemilumineszenz immunoassay (ECLIA) for IL-6 (ROCHE Diagnostics GmbH). BNP and hsCRP were measured using the immunoassays Architect BNP assay, and Multigent CRP Vario (Abbott Laboratories). The hematological determinations were performed in the automatic hematological analyzer XE5000 (Sysmex).

2.4 Statistical analysis

Shapiro Wilk test was used to test normal distribution of the variables. All results are presented as mean±SEM (in case of normal distribution) or median [25–75% CI] (for non-normally distributed data). Differences of parameters between three groups (normoxia baseline, hypoxia, normoxia follow-up) were determined using the One Way ANOVA test (for normally distributed data) or One Way ANOVA on Ranks (Kruskal-Wallis, for non-normally distributed data). Afterwards, Dunn’s Test was used for all pair wise comparison. Additionally, the non-parametric Friedman’s repeated measures analysis of variance on ranks followed by the Dunn’s method for all pair wise comparisons was tested. Differences of parameters between two groups (relevant for parameters only measured in normoxia at baseline as well as normoxia at follow-up) were tested by the paired students t-test (in case of normal distribution) or rank sum test.P-values <0.05 were considered statistically significant. SigmaPlot Software Version 12.0 (Systat Software Inc.) was used for these analyses.

3 Results

3.1 Baseline characteristics

The 17 volunteers that were enrolled in the present study were all healthy, of an age of 36±3 years, and in major of normal weight and height (178±2 cm, BMI 24.3±0.9 kg/m²). For all clinical baseline data see Table 1.

3.2 Impact of hypoxia on vital parameters and standard laboratory values

Hypoxic conditions such as in 3000 m altitude caused a decrease in the participants’ oxygen saturation from 98±0.4 to 93±0.5%, and a compensatory increase in the breathing frequency from 12±0.4 to 17±0.8/min. (p both <0.001). Interestingly, 36 hours after entering an altitude of 3000 m, there was no difference in heart rate or blood pressure visible compared to the baseline measurements.

Standard laboratory values (Table 2) were only obtained in baseline measurements in normoxia and follow-up measurements in normoxia due to procedural reasons. During the excursion in the altitude of 3000 m, a blood cell differential was not available. Compared to baseline data, 48 hours after the return to normoxia, only a slight increase in the number of circulating monocytes was observed (BL: 0.4±0.0×10⁹/l, FU: 0.5±0.0×10⁹/l, p = 0.017). There was no significant change in the other numbers of circulating cell populations, see Table 2.

3.3 Impact of hypoxia on the number of circulating DCs

During hypoxia in high altitude (3000 m), a significant reduction in pDCs was observed (Fig. 2A). In the baseline measurements during normoxia, pDCs accounted 0.10 [0.08–0.18] % of white blood cell count (% WBC). During hypoxia, pDCs accounted only 0.03 [0.02–0.06] % WBC which is a reduction to the 0.3-fold (p < 0.001). In the follow-up, pDCs reconstituted up to 0.16±[0.11–0.26] % WBC which represents a re- increase to the 5-fold compared to levels during hypoxia (p < 0.001). Comparing follow-up measurements with baseline data, a re-increase of pDCs to the 1.6-fold was observed (One Way ANOVA repeated measurements p = 0.0013, Bonferroni posthoc BL vs. hypoxia, BL vs. FU, hypoxia vs. FU p < 0.05, F: 3,160; Friedman Repeated Measures Analysis of Variance on Ranks: Chi-square = 12,00 with 2 degrees onf freedom, p < 0.001. P-value after Bonferroni correcture: 0.0033).

As mentioned above, for baseline and follow-up measurements, standard laboratory including the number of leukocytes reflecting WBC were obtained. Additionally to the relative number of pDCs (% WBC), an absolute number of pDCs can be calculated by multiplication of the relative pDC number (% WBC) and the number of leukocytes. The results of the calculated absolute number of pDCs showed also a re-increase to the 1.6-fold compared to the baseline value which supports the observed effect of the relative cell number.

For mDCs there were no obvious changes during hypoxia observed. Hypoxia only resulted in a very slight decrease in mDCs (BL: 0.18 [0.12–0.24] % WBC, 3000 m: 0.15 [0.07–0.18] % WBC, p = 0.17). In the follow-up measurement 48 hours after return to normoxia, a slight increase of mDCs was observed (FU: 0.20 [0.17–0.26] % WBC, p = 0.04) (Fig. 3). But these observations did not reach significance (One Way ANOVA repeated measurements p = 0.04, Bonferroni posthoc BL vs. hypoxia, BL vs. FU, hypoxia vs. FU p < 0.05, F: 3.160).

4 Discussion

Hypoxia plays a major role in different diseases. As systemic hypoxia it occurs e.g. in high altitude illness. As tissue hypoxia it is in the focus in diseases like acute myocardial infarction and stroke. But it also plays an important role in physiological situations caused by humans for the reasons of altitude training. In this kind of training, athletes train under hypoxic conditions to achieve higher performance improvements than during sea-level exercise. Besides the accelerating effect of hypoxia on erythropoiesis that leads to an increase in haemoglobin and a higher oxygen uptake which are favourable for endurance ability, its effects on systemic inflammation are rarely known so far [22]. Hypoxia has been shown to cause a shift towards inflammation which can be seen in in the cellular response as well as the cytokine profile in humans in high altitude with an increase in pro-inflammatory, and a decrease in anti-inflammatory cytokines [7, 20]. This activation of the immune system is supposed to protect the organism from possible pathogens. Inflammatory properties are often underestimated. For example, atherosclerosis was considered a degenerative disease for decades, thus now it is known as an inflammatory disease. This brings us to the key unanswered question whether hypoxia also triggers inflammatory responses that are not known yet.

Dendritic cells are professional antigen-presenting cells that have a pivotal role linking the innate and the adaptive immune system [16]. Besides their role in protecting the organism against pathogen-induced infections, they also seem to have an important function in other inflammatory processes and diseases. For instance, the number of circulating DCs has been changed in cardiovascular diseases like coronary heart disease including acute myocardial infarction [5, 30], and stroke [28]. Since hypoxia is an important pathomechanism that occurs in pathogen-induced infections and inflamed tissue but also cardiovascular diseases, investigation of the number of circulating dendritic cells has been the focus of the present study.

This study demonstrated a hypoxia-induced decline in circulating pDCs during hypobaric hypoxia. This result is similar to the observation of Yilmaz et al. [29] for normobaric hypoxia. Interestingly, our study was capable to show an additional hypoxia-induced slight decline in mDCs. This effect could not be seen during the experiment of Yilmaz et al. This might be due to different setting conditions like the occurrence of hypobaric hypoxia in our experiment and normobaric hypoxia in Yilmaz et al.’s experiment, but also because of the shorter exposure time of the study’s participants in the hypoxic chamber experiment of Yilmaz et al. In our experiment, participants were exposed to hypoxia for 48 hours before blood draw, in the hypoxic chamber experiment, participants were exposed to hypoxia for 7 hours.

Overall experimental setting of the present study in an altitude of 3000 m was able to demonstrate the same significant reduction of pDCs for hypobaric hypoxia caused by exposure to high altitude of 5500 m in normobaric hypoxia. Interestingly, 48 hours after return to normoxia, the number of pDCs re-increased and reached levels even higher than the initial values which might represent an over-compensatoryeffect.

It is not clear yet what causes the reduction in circulating plasmacytoid DCs during hypoxia. Yilmaz et al. [29] demonstrated in a cell culture experiment that hypoxia comparable to the oxygen levels in the hypoxic chamber does not induce a reduction in the number of pDCs via cell death. The most reasonable explanation might be the migration of pDCs into hypoxic tissues. Tissue hypoxia is known to occur in multiple diseases like pathogen infection, atherosclerosis, malignant tumors, and rheumatoid arthritis [1 , 27]. Antigen-presenting cells infiltrate these hypoxic tissues and operate as activators of T cells [24]. This effect suggests that under systemic hypoxia, dendritic cells might also infiltrate certain organs of lymphoid tissue attempting to mediate inflammation and combat possible pathogens. Similar effects of a decline in circulating DCs during inflammatory diseases like coronary heart disease [30] have been observed in the past, but these changes mainly affected mDCs. Interestingly, for mDCs an inhibitory effect of hypoxia on their migratory capacity has been demonstrated [19]. This might be a reason for the significantly less reduced number of circulating mDCs compared to pDCs during systemichypoxia.

Furthermore, the background of the increase in pDCs 48 hours after return to normoxia remains unknown. Based on the observation that the relative as well as the absolute number of pDCs in the follow-up measurement is increased, a possible effect of dilution or subpopulation shift that might play a role for results of relative cell numbers can be excluded. 24 hours of normoxia after exposure to hypoxia for 7 hours, Yilmaz et al. [29] could show that pDCs reached baseline values. In the present study, 48 hours of normoxia after exposure to hypoxia for 72 hours we were able to detect an even significantly higher number of pDCs compared to the baseline level. The reason for this effect might be an overwhelming production of pDCs in the bone marrow. Alternatively, re-emigration of pDCs that were present in local tissue back to the blood might be a possible mechanism.

5 Conclusion

In conclusion, the present study revealed a decrease in the number of circulating pDCs during hypobaric hypoxia as well as an excessive increase after return to normoxia reaching even higher levels of circulating pDCs compared to the baseline measurements. Further studies are required to investigate the mechanisms that are responsible for the hypoxia-induced pDC decline as well as its clinical relevance for of hypoxia-associated diseases and its possible effects on healthy subjects e.g. in altitude training.

Declaration of interests

Financial support by an unrestricted grant by Actelion Germany.

Footnotes

Acknowledgments

This study was supported by an unrestricted grant from Actelion Pharma Germany.

References

Bjornheden

, Levin

, Evaldsson

and Wiklund

, Evidence of hypoxic areas within the arterial wall in vivo. Arterioscler Thromb Vasc Biol19 (1999), 870–876.

Bor-Kucukatay

, Colak

, Erken

, Kilic-Toprak

and Kucukatay

, Altitude training induced alteration in erythrocyte rheological properties: A controlled comparison study in rats. Clin Hemorheol Microcirc58(4) (2014), 479–488.

Elia

A.R.

, Cappello

, Puppo

, Fraone

, Vanni

, Eva

, Musso

, Novelli

, Varesio

and Giovarelli

, Human dendritic cells differentiated in hypoxia down-modulate antigen uptake and change their chemokine expression profile. J Leukoc Biol84 (2008), 1472–1482.

Eltzschig

H.K.

and Carmeliet

, Hypoxia and inflammation. N Engl J Med364 (2011), 656–665.

Fukui

, Yasukawa

, Sugi

, Oba

, Nagata

, Kyogoku

, Futamata

, Yokoyama

, Kai

, Ueno

, Kage

and Imaizumi

, Transient reduction and activation of circulating dendritic cells in patients with acute myocardial infarction. Int J Cardiol160 (2012), 216–219.

Hackett

P.H.

and Roach

R.C.

, High-altitude illness. N Engl J Med345 (2001), 107–114.

Hartmann

, Tschop

, Fischer

, Bidlingmaier

, Riepl

, Tschop

, Hautmann

, Endres

and Toepfer

, High altitude increases circulating interleukin-6, interleukin-1 receptor antagonist and C-reactive protein. Cytokine12 (2000), 246–252.

Imtiyaz

H.Z.

and Simon

M.C.

, Hypoxia-inducible factors as essential regulators of inflammation. Curr Top Microbiol Immunol345 (2010), 105–120.

Klokker

, Kjaer

, Secher

N.H.

, Hanel

, Worm

, Kappel

and Pedersen

B.K.

, Natural killer cell response to exercise in humans: Effect of hypoxia and epidural anesthesia. J Appl Physiol78 (1995), 709–716.

10.

Lichtenauer

, Goebel

, Fritzenwanger

, Forster

, Betge

, Lauten

, Figulla

H.R.

and Jung

, Simulated temporary hypoxia triggers the release ofCD31+/Annexin+endothelialmicroparticles:Aprospective pilot study in humans. Clin Hemorheol Microcirc2014 [Epub ahead of print].

11.

Mazzeo

R.S.

, Brooks

G.A.

, Butterfield

G.E.

, Podolin

D.A.

, Wolfel

E.E.

and Reeves

J.T.

, Acclimatization to high altitude increase muscle sympathetic activity both at rest and during exercise.R. Am J Physiol269 (1995), 201–207.

12.

Meehan

, Duncan

, Neale

, Taylor

, Muchmore

, Scott

, Ramsey

, Smith

, Rock

, Goldblum

, et al., Operation Everest II: Alterations in the immune system at high altitudes. J Clin Immunol8 (1988), 397–406.

13.

Miao

, Reference range of hematocrit in the elderly with respect to altitude. Clin Hemorheol Microcirc29(1), (2003), 25–31.

14.

Miao

, Zhiyuan

, Qingsheng

and Haiyan

, The relationship between reference value of old people’s erythrocyte sedimentation rate and altitude. Cin Hemorheol Microcirc24(3), (2001), 155–159.

15.

Nantais

, Dumbarton

T.C.

, Farah

, Maxan

, Zhou

, Minor

and Lehmann

, Impact of methylene blue in addition to norepinephrine on the intestinal microcirculation in experimental septic shock. Clin Hemorheol Microcirc58(1), (2014), 97–105.

16.

Niessner

and Weyand

C.M.

, Dendritic cells in atherosclerotic disease. Clin Immunol134 (2010), 25–32.

17.

Ogino

, Onishi

, Suzuki

, Morisaki

, Tanaka

and Katano

, Inclusive estimation of complex antigen presentation functions of monocyte-derived dendritic cells differentiated under normoxia and hypoxia conditions. Cancer Immunol Immunother61 (2011), 409–424.

18.

Pierobon

, Bosco

M.C.

, Blengio

, Raggi

, Eva

, Filippi

, Musso

, Novelli

, Cappello

, Varesio

and Giovarelli

, Chronic hypoxia reprograms human immature dendritic cells by inducing a proinflammatory phenotype and TREM-1 expression. Eur J Immunol43 (2013), 949–966.

19.

, Yang

M.X.

, Kong

B.H.

, Qi

, Lam

Q.L.

, Yan

, Li

, Zhang

and Lu

, Hypoxia inhibits the migratory capacity of human monocyte-derived dendritic cells. Immunol Cell Biol83 (2005), 668–673.

20.

Rohm

, Ratka

, Pistulli

, Goebel

, Gecks

, Figulla

, Yilmaz

and Jung

, Impact of Systemic Normobaric Short-term Hypoxia on Pro-inflammatory and Anti-inflammatory Cytokines in Healthy Volunteers. Clinical Laboratory2015 [in print].

21.

Sallisalmi

, Tenhunen

, Kultti

, Tammi

and Pettilä

, Plasma hyaluronan and hemorheology in patients with septic shock: A clinical and experimental study. Clin Hemorheol Microcirc56(2), (2014), 133–144.

22.

Saunders

P.J.

, Garvican-Lewis

L.A.

, Schmidt

W.F.

and Gore

C.J.

, Relationship between changes in haemoglobin mass and maximal oxygen uptake after hypoxic exposure. Br J Sports Med47(1), (2013), i26–i30.

23.

Sivakumar

, Akhavani

M.A.

, Winlove

C.P.

, Taylor

P.C.

and Paleolog

E.M.

, and Kang, Synovial hypoxia as a cause of tendon rupture in rheumatoid arthritis. J Hand Surg Am33 (2008), 49–58.

24.

Strehl

, Fangradt

, Fearon

, Gaber

, Buttgereit

and Veale

D.J.

, Hypoxia: How does the monocyte-macrophage system respond to changes in oxygen availability?. J Leukoc Biol95 (2014), 233–241.

25.

Thake

C.D.

, Mian

, Garnham

A.W.

and Mian

, Leukocyte counts and neutrophil activity during 4h of hypocapnic hypoxia equivalent to m. Aviat Space Environ Med75 (2004), 811–817.

26.

Thompson

L.F.

, Eltzschig

H.K.

, Ibla

J.C.

, Van De Wiele

C.J.

, Resta

, Morote-Garcia

J.C.

and Colgan

S.P.

, Crucial role for ecto-5’-nucleotidase (CD73) in vascular leakage during hypoxia. J Exp Med200 (2004), 1395–1405.

27.

Vaupel

, Kelleher

D.K.

and Hockel

, Oxygen status of malignant tumors: Pathogenesis of hypoxia and significance for tumor therapy. Semin Oncol28 (2001), 29–35.

28.

Yilmaz

, Fuchs

, Dietel

, Altendorf

, Cicha

, Stumpf

, Schellinger

P.D.

, Blumcke

, Schwab

, Daniel

W.G.

, Garlichs

C.D.

and Kollmar

, Transient decrease in circulating dendritic cell precursors after acute stroke: Potential recruitment into the brain. Clin Sci (Lond)118 (2010), 147–157.

29.

Yilmaz

, Ratka

, Rohm

, Pistulli

, Goebel

, Asadi

, Petri

, Kiehntopf

, Figulla

H.R.

and Jung

, Decrease in circulating plasmacytoid dendritic cells during short-term systemic normobaric hypoxia. Eur J Clin Invest2015 [Epub ahead of print].

30.

Yilmaz

, Weber

, Cicha

, Stumpf

, Klein

, Raithel

, Daniel

W.G.

and Garlichs

C.D.

, Decrease in circulating myeloid dendritic cell precursors in coronary artery disease. J Am Coll Cardiol48 (2006), 70–80.