Lymphocyte apoptosis is resistant to erythropoietin in porcine endotoxemia

 2010 The Authors APMIS  2010 APMIS DOI 10.1111/j.1600-0463.2010.02704.x

APMIS 119: 143–154

Lymphocyte apoptosis is resistant to erythropoietin in porcine endotoxemia CHRISTOFFER SØLLING,1,2 UFFE NYGAARD,1,2 ANTON T. CHRISTENSEN,1,2 LISE WOGENSEN,2,3 JAN KROG1,2 and ELSE K. TØNNESEN1,2 1

Department of Anesthesiology and Intensive Care Medicine, Aarhus University Hospital; 2Institute of Clinical Medicine, Aarhus University; and 3The Research Laboratory for Biochemical Pathology, Aarhus University Hospital, Aarhus, Denmark

Sølling C, Nygaard U, Christensen AT, Wogensen L, Krog J, Tønnesen EK. Lymphocyte apoptosis is resistant to erythropoietin in porcine endotoxemia. APMIS 2011; 119: 143–54. Sepsis-induced lymphocyte apoptosis plays an important role in the development of immune suppression in septic patients. Erythropoietin (EPO) is a multifunctional cytokine with antiapoptotic properties. We hypothesized that EPO could mitigate mononuclear cell (MNC) apoptosis and modify the dynamic changes of MNCs during endotoxemia. Twenty-six pigs were randomized into three groups: (i) lipopolysaccharides (LPS), (ii) EPO (epoetin-a, 5000 IU ⁄ kg) administered 60 min prior to LPS, and (iii) sham. At 120 min of endotoxemia, the animals were fluid resuscitated and the LPS infusion was reduced. MNCs were isolated at 0, 60, 240, and 540 min of endotoxemia, and apoptosis was assessed by flow cytometry. Apoptosis in splenic biopsies was quantified by immunohistochemistry. Endotoxemia increased the number of apoptotic MNCs in the blood (p £ 0.01) and the spleen (p = 0.03), and EPO did not modify this increase. The number of T-helper and cytotoxic T cells declined during endotoxemia. The dynamic changes of the MNC subsets were not modified by treatment with EPO. In conclusion, EPO did not modify the LPS-induced changes of MNC subsets or mitigate the levels of apoptosis of MNCs in the blood or in the spleen. This study does not support that EPO confers protection against lymphocyte apoptosis. Key words: Erythropoietin; apoptosis; mononuclear cells; endotoxemia. Christoffer Sølling, Department of Anesthesiology and Intensive Care Medicine, Aarhus University Hospital, Aarhus Sygehus, Norrebrogade 44, DK-8000 Aarhus C, Denmark. e-mail: [email protected]

Sepsis is a systemic disorder, with a protean clinical picture and a complex pathogenesis characterized by both proinflammatory and anti-inflammatory elements. While the organ dysfunction and failure that develop in the early phases of severe sepsis are believed to result from an excessive inflammatory response (1), the later phases of sepsis are often characterized by immunosuppression (2). During the later phase, patients demonstrate higher susceptibility to new infections and often succumb to nosocomial infections Received 20 August 2010. Accepted 20 October 2010

caused by low virulence pathogens (2–4). The risk of developing secondary infections correlates with the degree of leukopenia, which is a prominent occurrence in severe sepsis (5–7). Leukopenia may be a result of leukocyte apoptosis and, interestingly, Hotchkiss et al. (8) have demonstrated that patients who died of severe sepsis had higher numbers of apoptotic immune cells in splenic biopsies than patients who died from trauma or non-septic critical illness. Consistent with this report, Weber et al. (9) have demonstrated high numbers of apoptotic circulating lymphocytes in septic patients. 143

SØLLING et al.

While lymphocyte apoptosis reduces the number of functional immune cells, phagocytosis of the apoptotic lymphocytes may also induce a state of immunosuppression by mechanisms that are still unknown (10, 11). In animal models of experimental sepsis, inhibition of lymphocyte apoptosis by caspase inhibitors has significantly improved survival (12, 13), and therefore, lymphocyte apoptosis may be crucial to the development of sepsis-induced immunosuppression. Erythropoietin (EPO) is a well-known pharmaceutical agent that has been used for several decades in the treatment of certain anemias. In the bone marrow, EPO inhibits apoptosis of erythrocyte progenitor cells and thereby increases formation of new red blood cells. In addition to its hematopoietic effects, EPO is now recognized as a multifunctional cytokine with anti-inflammatory, antioxidative, and antiapoptotic properties (14–16). In models of renal ischemia–reperfusion injuries, EPO attenuates tubular apoptosis and improves renal function (14, 17); and in models of endotoxemia, EPO improves survival and decreases apoptosis in the lungs and liver (18). EPO receptors are present in normal T and B cells (19), and a study has demonstrated that EPO inhibits cytokine production in peripheral blood mononuclear cells (PBMCs) after in vitro stimulation (20).

We hypothesized that EPO, as a strong antiapoptotic mediator, could mitigate mononuclear cell (MNC) apoptosis during lipopolysaccharide (LPS)-induced endotoxemia. The aims of the study were, therefore, to assess the antiapoptotic effects of EPO in a porcine model of endotoxic shock by quantifying apoptosis in PBMC and in splenic lymphocytes. As secondary outcomes, we evaluated whether EPO could modify the dynamic changes in the distribution of monocytes, T-helper cells, cytotoxic lymphocytes, and B cells during an endotoxic challenge.

MATERIALS AND METHODS This study was approved by the National Committee on Animal Research Ethics (Experimental Animal Inspectorate, Copenhagen, Denmark) and conducted in accordance with the ‘Principles of Laboratory Animal Care’ (NIH publication No. 86-23; revised 1996). Animal preparation Twenty-six female crossbred Landrace ⁄ Yorkshire ⁄ Duroc pigs (31–35 kg) were fasted overnight, but allowed free access to water. The animals were pre-medicated with intramuscular injection of midazolam (0.5 mg ⁄ kg), s-ketamine (5 mg ⁄ kg), and atropine (0.5 mg). Anesthesia was induced with midazolam (0.5 mg ⁄ kg) and s-ketamine (5 mg ⁄ kg) intravenously and maintained with a continuous infusion

Intubation , surgical preparation, and sixty minutes rest.

± Epoietin-α (5000 IU/kg) Volume loading HES + NaCl Groups:

Fluid resuscitation to MAP > 65 mmHg

LPS 20 µg/kg/h

LPS ± EPO

LPS 2.5 µg/kg/h

Basal fluid substitution: Ringer´s acetate 10 mL/kg/h + 2.5% glucose

Volume loading HES + NaCl Sham:

Fluid resuscitation to MAP > 65 mmHg

Basal fluid substitution: Ringer´s acetate 10 mL/kg/h + 2.5% glucose – 60

0

30

60

120

240

540

600 min

PBMC:

Fig. 1. Schematic presentation of the experimental protocol. The animals were randomly allocated to: (i) lipopolysaccharide (LPS; n = 9), (ii) LPS + erythropoietin (EPO; n = 9), and (iii) sham, anesthetized, and operated only (n = 8). LPS infusion was initiated at 0 min in the endotoxic groups. After 120 min, the LPS infusion was reduced and the animals were fluid resuscitated. Blood samples for isolation of peripheral blood mononuclear cells were obtained at 0, 60, 240, and 540 min of endotoxemia. Samples from the spleen were obtained at the end of the experiment. HES, hydroxyethyl starch; MAP, mean arterial pressure. See text for details.

144

 2010 The Authors APMIS  2010 APMIS

EPO ON PBMC APOPTOSIS DURING ENDOTOXEMIA

of fentanyl (60 lg ⁄ kg ⁄ h) and midazolam (6 mg ⁄ kg ⁄ h). No neuromuscular blocking agent was used. The animals were endotracheally intubated (6.5 mm, Portex Tube; Smiths Medical International, Kent, UK) and volume-controlled ventilated (S ⁄ 5 Avance; Datex Ohmeda, Madison, WI, USA) using a tidal volume: 10 mL ⁄ kg; positive end-expiratory pressure (PEEP): 5 cm H2O; and fraction of inspiratory oxygen (FiO2): 0.4. The ventilation rate was adjusted to maintain end-tidal CO2 between 5.0 kPa and 6.0 kPa. A catheter was inserted in the right external jugular vein for fluid and drug management and in the right carotid artery for blood pressure monitoring and blood sampling. All animals were laparotomized, and the incision was closed in two layers. After the preparations, the animals were left to rest for 2 h prior to infusion of endotoxin. The body temperature was regulated by air warming (WarmTouch; Tyco Healthcare, Hamshire, UK) and maintained between 37 C and 38 C in the sham group and above 37 C in the two endotoxic groups. All animals received a bolus of saline before the surgical preparations (15 mL ⁄ kg), and Ringer’s acetate supplemented with 2.5% glucose (10 mL ⁄ kg ⁄ h) was infused throughout the experiment. Experimental protocol The animals were randomized, in a blinded manner, to receive either 5000 IU ⁄ kg of EPO (epoetin-a, Eprex; Janssen–Cilag, Birkeroed, Denmark) intravenously 1 h prior to endotoxemia (LPS + EPO group; n = 9) or 5 mL saline (LPS group; n = 9). A group of control animals were only anesthetized and operated on without receiving EPO or LPS (sham group; n = 8) (Fig. 1). All investigators were naive to the randomization of the LPS groups until the completion of the laboratory analyses. Endotoxemia was induced by an infusion of Escherichia coli LPS (0111: B4; Sigma, St. Louis, MO, USA) starting at 2.5 lg ⁄ kg ⁄ h and by increasing stepwise to 20 lg ⁄ kg ⁄ h during 30 min. If mean arterial pressure (MAP) decreased to the level of mean pulmonary artery

pressure (MPAP) during the first hour of endotoxemia, intravenous epinephrine (0.1 lg ⁄ kg) was given to avoid circulatory collapse and death. After 2 h of endotoxemia, the animals were fluid resuscitated, and the LPS infusion was reduced to a maintenance dose of 2.5 lg ⁄ kg ⁄ h for the remaining 8 h. Fluid resuscitation comprised of 15 mL ⁄ kg of hydroxyethyl starch 130 ⁄ 0.4 (Voluven; Fresenius Kabi, Bad Homburg, Germany) and 30 mL ⁄ kg of saline. The need for additional fluid was evaluated hourly, and saline (500–000 mL) was administered to keep MAP above 65 mmHg. Hypoglycemia (blood glucose < 3.5 mmol ⁄ L) was treated with refractory doses of glucose (500 mg ⁄ L) until normoglycemia was achieved. Severe shivering was treated with intravenous pethidine (25 mg). After 600 min of endotoxemia, the animals were killed with an overdose of intravenous pentobarbital. Isolation of PBMCs Venous peripheral blood samples were collected into lithium-heparinized tubes (Greiner Bio-One GmbH, Frickenhausen, Germany) at 0, 60, 240, and 540 min of endotoxemia. Blood samples were diluted 1:1 in RPMI-1640 and the PBMCs were isolated by density gradient centrifugation using Lympholyte-Mammal (Cederlane Laboratories Ltd., Hornby, ON, Canada). PBMCs harvested from the interface were washed twice in phosphate-buffered saline supplemented with 2% heat-inactivated fetal calf serum (FCS; Gibco, Paisley, UK) and stored in RPMI-1640 supplemented with 10% FCS in polypropylene tubes (Greiner BioOne GmbH) at 4 C until analysis. Flow cytometry The PBMCs were adjusted to 2 · 106 cells ⁄ mL in annexin-binding buffer (cat. no.: 556454; BD Biosciences, Broendby, Denmark) and 100 lL of cell suspension was added to each test tube containing antibody combinations specific for various surface

Table 1. Antibodies and apoptosis-detecting fluorescent markers. The antibodies conjugates and apoptosisdetecting markers were combined to include one lineage marker (CD45, CD4, CD8, or CD21) and both annexin V and 7AAD Specificity Clone Isotype Form Company Cat. no. Concentration (lL) CD45 Pig MIL13 IgG1 FITC AbD Serotec MCA1218F 5 CD4 Pig MIL17 IgG2b FITC AbD Serotec MCA1749F 10 CD8 Pig MIL12 IgG2a FITC AbD Serotec MCA1223F 5 CD14 Pig MIL2 IgG2b FITC AbD Serotec MCA1218F 5 CD21 Human1 Bly-4 IgG1J PE BD 555422 10 Annexin V APC BD 550475 5 7AAD BD 559925 40 FITC, fluorescein isothiocyanate; PE, phycoerythrin; APC, allophycocyanin; 7AAD, 7-aminoactinomycin D; IgG, immunoglobulin G. 1 Porcine cross-reactive antibody. Recognizes porcine B lymphocytes (41).  2010 The Authors APMIS  2010 APMIS

145

SØLLING et al.

markers. The antibody panel is shown in Table 1. Antibody concentrations were determined by prior titration experiments to optimize the discrimination between positive and negative populations in single color tubes (data not shown). Apoptosis was assessed using the fluorescent DNA dye 7-aminoactinomycin D (7AAD; BD Pharmingen, Broendby, Denmark) (21) and annexin V (Annexin V–APC; BD Pharmingen). After gentle vortexing, the cells were incubated for 15 min in the dark at room temperature. After incubation, the samples were washed once in annexinbinding buffer (BD Pharmingen; 200 g in 5 min) and resuspended in annexin-binding buffer supplemented with the non-fluorescent actinomycin D (cat. no.: 01815; Sigma-Aldrich, Broendby, Denmark; 20 lg ⁄ mL) to avoid release of 7AAD (22). All tubes were gently vortexed and analyzed within 2 h using a FASC Canto multicolor flow cytometer (BD Biosciences, Broendby, Denmark). For each sample, 10 000 PBMCs (defined as CD45+ cells) were acquired and the lymphoid and myeloid cell populations were identified according to their forward scatter (FSC) and side scatter (SSC) characteristics (Fig. 2). Three-color flow cytometry analysis was performed using one marker to define cell type, CD4+ (T-helper cells), CD8+ (cytotoxic T cells), CD21+ (B cells), CD14+ (monocytes), and two markers to define apoptosis, 7AAD and annexin V. Staining with 7AAD results in three distinct populations. A population with a high 7AAD fluorescence (7AADbright) was considered to represent dead cells; a population with a low 7AAD fluorescence (7AADdim) was considered to represent apoptotic cells; and a population not stained by 7AAD (7AAD)) was considered to represent living cells

Sham LPS LPS + EPO

Leukocytes (×109 L–1)

35 30 25 20 15

Total leukocyte count The total number of leukocytes was counted manually using whole blood mixed with methyl violet-acetic acid in a Bu¨rker–Tu¨rk counting chamber. Immunohistochemistry At 600 min of endotoxemia, the spleen was removed, and randomly sampled tissue blocks of approximately 0.5 · 0.5 · 0.5 mm were fixed overnight in formaldehyde (4%, pH 7.4) and embedded in paraffin. The samples were cut into 3-lm sections. Caspase-dependent apoptosis was quantified using immunohistochemistry for activated caspase3. Deparaffinized and rehydrated sections were boiled in citrate buffer, and endogenous peroxidase was inhibited by incubation in 3% H2O2 for 10 min. Sections were blocked (5% normal goat serum in TBS, 0.1% Tween-20) and incubated with the primary antibody (anticaspase-3; Abcam Plc, Cambridge, UK). The ABC kit standard peroxidase system (Vector Laboratories, Burlingame, CA, USA) was used with a biotinylated goat antirabbit secondary antibody (1:200) and 3.3¢-diaminobenzidine tetrahydrochloride (DAB; Sigma) as the chromogen. Slides were counterstained with Mayer’s hematoxylin for 3 min. As a negative control, the primary antibody was omitted. Apoptotic MNCs were quantified using light microscopy. Ten random vision fields (·200) per section were viewed, giving a total of 20 vision fields per spleen. Statistics

#

10

*

5 0 0

120

240

360

480

600

Time (min)

Fig. 2. Total leukocyte counts during endotoxemia in the sham and endotoxemic groups. Endotoxemia significantly reduced the number of circulating leukocytes. Values are mean and SD; # indicates p = 0.0001 (MANOVA) and * indicates p = 0.0009 (MANOVA) compared with the sham group.

146

(21). In this study, only apoptotic (7AADdim) cells were included in the analysis. The results from the 7AAD analysis were confirmed with annexin V staining. Annexin V binds to early apoptotic, late apoptotic, and dead cells. To exclude dead cells, only annexin V-positive cells that were 7AADbright-negative were included. Data from the flow cytometry were analyzed using FlowJo software, version 7.5.5 (Tree Star, Inc., Ashland, OR, USA).

All flow cytometry and histologic data were logarithmically transformed to ensure normal distribution. The differences between baseline values were analyzed using one-way multivariate analysis of variance (MANOVA), while time dependency and differences between groups were analyzed using two-way MANOVA for repeated measurements. If there was no time ⁄ group interaction, the difference in mean levels was analyzed using Student’s t-test. Data are presented as mean ± SD unless otherwise stated. Two-tailed p-values less than 0.05 were considered statistically significant. All analyses were performed

 2010 The Authors APMIS  2010 APMIS

EPO ON PBMC APOPTOSIS DURING ENDOTOXEMIA

using the Stata software package 10.0 (StataCorp LP, College Station, TX, USA).

RESULTS Experimental design and control parameters

One animal from each of the LPS groups died during endotoxemia. Thus, complete data were available for 24 animals, i.e., for eight animals in each of the following groups: LPS, LPS + EPO, and sham. There were no differences between the three groups in weight or baseline hemodynamics (MAP, heart rate, cardiac index, central venous pressure, or MPAP; all p > 0.05, one-way MANOVA) and no differences in the

hemodynamics during the experiment between the two LPS groups (all p > 0.05, two-way MANOVA). Total leukocyte count declined during endotoxemia

The leukocyte levels are shown in Fig. 2. As demonstrated previously, induction of endotoxemia leads to reduced levels of circulating leukocytes (23). The lowest number was observed at 240 min of endotoxemia, followed by a slight increase from 240 min to 540 min. There was no difference in the levels of circulating leukocytes between the animals treated with placebo or EPO. In the sham group, the leukocyte count was constant throughout the experiment.

Table 2. Percentages of peripheral blood mononuclear cell (PBMC) subpopulations during endotoxemia. The percentages of lymphocytes, lymphocyte subsets, monocytes, and cells of an emerging myeloid subpopulation during the experiment are shown. The myeloid subpopulation presumably consists of band neutrophils (see text for details) PBMC subpopulation Baseline 60 min 240 min 540 min Lymphocytes (% of PBMC) Sham 94.1 ± 1.9 93.9 ± 2.9 93.8 ± 2.6 93.3 ± 3.2 92.7 ± 2.0 91.5 ± 5.4 86.8 ± 5.13 61.3 ± 12.23 LPS1,2 LPS + EPO1,2 91.9 ± 4.4 93.3 ± 5.8 84.0 ± 9.2 56.7 ± 19.03 CD4+ (% of lymphocytes) Sham 20.7 ± 3.8 20.9 ± 3.1 20.6 ± 4.2 20.2 ± 2.5 25.7 ± 5.8 20.6 ± 8.1 21.1 ± 4.7 19.3 ± 7.23 LPS1,2 1,2 LPS + EPO 24.3 ± 6.2 20.8 ± 8.0 22.6 ± 6.3 15.9 ± 5.33 CD8+ (% of lymphocytes) 19.8 ± 3.2 21.0 ± 4.0 19.9 ± 4.2 20.7 ± 4.7 Sham2 22.7 ± 4.5 18.2 ± 5.33 12.0 ± 4.33 12.4 ± 4.83 LPS1,2 LPS + EPO1,2 23.1 ± 7.5 18.4 ± 3.3 11.7 ± 1.73 13.5 ± 6.53 CD4 ⁄ CD8 Sham 1.1 ± 0.2 1.0 ± 0.1 1.1 ± 0.2 1.0 ± 0.2 1.2 ± 0.3 1.2 ± 0.4 1.8 ± 0.33 1.6 ± 0.43 LPS1,2 1,2 3 LPS + EPO 1.1 ± 0.4 1.1 ± 0.5 1.9 ± 0.5 1.3 ± 0.4 CD21+ (% of lymphocytes) 12.0 ± 5.5 13.9 ± 6.33 13.0 ± 6.8 10.6 ± 5.8 Sham2 1,2 LPS 12.3 ± 2.1 9.6 ± 3.33 16.3 ± 2.93 11.7 ± 3.0 LPS + EPO1,2 10.6 ± 4.3 6.9 ± 3.53 12.5 ± 4.5 9.4 ± 2.93 Monocytes (% of PBMC) Sham 4.3 ± 2.0 4.8 ± 2.7 5.0 ± 2.3 5.2 ± 3.2 4.8 ± 1.5 2.1 ± 1.13 3.3 ± 1.03 4.3 ± 1.6 LPS1,2 5.1 ± 3.8 1.8 ± 0.73 3.8 ± 1.8 3.3 ± 1.1 LPS + EPO1,2 Myeloid subpopulation (% of PBMC) Sham 1.7 ± 0.5 1.3 ± 0.7 1.2 ± 0.6 1.5 ± 1.3 2.5 ± 1.0 6.3 ± 4.5 9.9 ± 5.03 34.5 ± 12.33 LPS1,2 LPS + EPO1,2 2.9 ± 1.3 4.9 ± 5.4 12.2 ± 8.33 40.0 ± 19.43 LPS, lipopolysaccharides; EPO, erythropoietin. 1 Significant time ⁄ group interaction compared with sham (two-way MANOVA). 2 Significant change over time (two-way MANOVA). 3 Significant compared with baseline (paired t-test).  2010 The Authors APMIS  2010 APMIS

147

SØLLING et al.

Fig. 3. Flow cytometry gating of peripheral blood mononuclear cells (PBMC) according to their forward and side scatter properties. PBMCs were isolated at 0 (A, E), 60 (B, F), 240 (C, G), and 540 min (D, H). The lipopolysaccharide (LPS; A–D) plots are representative of both endotoxemic groups. A gate was set around lymphocytes (L), monocytes (M1), and an LPS-induced myeloid subpopulation (M2) that presumably consists of band neutrophils. M2 was only observed in the endotoxic groups and emerged at 240 min of endotoxemia.

Dynamic changes in lymphocyte subsets and myeloid populations in response to endotoxemia

The percentage of CD8-positive lymphocytes significantly declined during the endotoxic challenge with no differences between the LPS and LPS + EPO groups (p = 0.55, two-way MANOVA; Table 2). The number of CD4-positive cells also declined, but to a lesser extent than the CD8-positive cells, thereby resulting in a higher CD4 ⁄ CD8 ratio during endotoxemia. There were no differences between the changes in CD4 percentages (p = 0.56, two-way MANOVA) or CD4 ⁄ CD8 ratio (p = 0.24, two-way MANOVA) between the two LPS groups. The percentages of B cells, as identified by their CD21 receptor, changed in a biphasic pattern, with an instant decline during the first 60 min followed by an increase. The LPS and the LPS + EPO groups displayed similar dynamic changes of B cells over time (p = 0.89, two-way MANOVA). The number of monocytes was constant in the sham group during the experiment, but declined significantly in both LPS groups without any 148

effects from EPO intervention (Table 2). Endotoxemia dramatically induced a subpopulation of myeloid cells with a more granular appearance (higher SSC reflection; Fig. 3). This new population comprised 86.5% ± 12.7 CD14positive cells. Endotoxemia increased PBMC and splenic lymphocyte apoptosis

A representative example of the data from the apoptosis quantification using 7AAD and annexin V staining is shown in Fig. 4. Annexin V staining was used to validate the results of the 7AAD staining, and the two methods were consistent except in the analysis of monocyte apoptosis. Using 7AAD staining, there was only a trend toward an increase in the frequency of apoptotic monocytes during endotoxemia (p = 0.17, two-way MANOVA), whereas this increase was significant using annexin V staining (p = 0.01, two-way MANOVA). The number of apoptotic PBMCs increased significantly during high-dose endotoxemia, but  2010 The Authors APMIS  2010 APMIS

EPO ON PBMC APOPTOSIS DURING ENDOTOXEMIA

A

B

C

E

F

D

G

Fig. 4. Flow cytometry detection of apoptosis using 7AADdim and annexin V. Representative examples of flow cytometry detection of apoptosis in CD45-positive peripheral blood mononuclear cells (PBMCs) and T-helper cells (CD4+ lymphocytes). Lymphocytes, monocytes, and the lipopolysaccharide (LPS)-induced myeloid subpopulation (M2) were identified by their size and granularity in a forward scatter (FSC) side scatter (SSC) plot (A). The CD45+ cells were identified (B) and the frequencies of 7AADbright, 7AADdim, and 7AADneg of CD45+ PBMCs are shown (C). Apoptosis was identified as 7AADdim cells, whereas 7AADbright were regarded as dead cells. The frequencies of apoptotic cells were confirmed by annexin V staining; cells were considered apoptotic if they were positive for annexin V but negative for 7AADbright (D). The analysis of T-helper cells is shown (E) as an example of the quantification of apoptosis in lymphocyte subsets. Apoptotic T-helper cells were identified as CD4+ ⁄ 7AADdim (F) and CD4+ ⁄ annexin V+ ⁄ 7AADbright-neg (G). 7AAD indicates 7-aminoactinomycin D.

was restored after fluid resuscitation and reduction in LPS dose (Fig. 5). EPO had no impact on the mean numbers of apoptotic PBMCs (p = 0.23, t-test). The increase in apoptotic PBMCs was not reflected by an increase in apoptosis of the subsets of circulating lymphocytes (Table 3). Among CD4+ lymphocytes, there were lower frequencies of apoptotic cells during endotoxemia, and the numbers of apoptotic CD8+ and apoptotic CD21+ lymphocytes did not increase significantly compared with the sham group (Table 3). The frequency of apoptosis in the emerging endotoxin-induced myeloid subpopulation was only analyzed at 240 and 540 min, and only in the LPS groups, as the population could not be detected at earlier time points and the population was not detectable in the sham group. The frequency of apoptotic cells in the  2010 The Authors APMIS  2010 APMIS

myeloid subpopulation was not significantly different compared with those of the lymphocyte or monocyte populations at the corresponding time points (all p > 0.05, paired t-tests). The immunohistochemical quantification of apoptotic lymphocytes in splenic biopsies is shown in Fig. 6. At 600 min of endotoxemia, the number of apoptotic lymphocytes was higher in the endotoxemic groups compared with that in the sham group (p = 0.03; Fig. 7), but there was no difference between EPO- and placebo-treated LPS groups.

DISCUSSION In this study, we tested whether EPO, a multifunctional and antiapoptotic cytokine, could 149

7AADdim positive cells (indexed to baseline)

SØLLING et al.

2.5

Sham LPS LPS + EPO

2.0

1.5

*

1.0

# 0.5

0.0 0

60

120

180

240

300

360

420

480

540

Time (min)

Fig. 5. Endotoxemia induced apoptosis in peripheral blood mononuclear cells (PBMCs). The figure depicts the changes in frequency of apoptotic (7AADdim) PBMCs during the experiment. The changes over time were significantly different between the lipopolysaccharide (LPS) group and the sham group (p = 0.004, MANOVA), the LPS + erythropoietin (EPO) group and the sham group (p = 0.01, MANOVA), but not between the two LPS groups (p = 0.74, MANOVA). There was no difference between the endotoxic groups in the mean level of apoptosis (p = 0.23, t-test); , sham group; d, LPS; s, EPO group. 7AAD indicates 7-aminoactinomycin D.

inhibit the apoptosis of lymphocytes and monocytes during endotoxemic shock. The endotoxic challenge resulted in elevated numbers of apoptotic MNCs in both blood samples and in biopsies from the spleen. This increase in apoptotic MNCs is in accordance with previous reports demonstrating apoptosis of lymphocytes in the spleen, thymus, and lymph nodes (24–27) and apoptosis of monocytes in peripheral blood (23) in response to endotoxemia. The frequencies of T-helper cells (CD4+) and cytotoxic T cells (CD8+) declined in peripheral blood during endotoxemia, but we were unable to determine whether this was because of apoptosis. The percentages of apoptotic CD4+ or CD8+ T cells did not increase in spite of the decline in the numbers of these cells. This could be the result of a rapid clearance of circulating apoptotic cells by the reticuloendothelial system. The spleen is an important part of the reticuloendothelial system, and the number of apoptotic cells was significantly increased in spleen samples in the endotoxic groups; however, whether this increase is because of the capture of circulating apoptotic cells or in situ 150

apoptotic splenic lymphocytes cannot be determined. Endotoxemia dramatically induced an emerging population of cells with higher granularity and myeloid characteristics. This cell population has been described previously, and it consists of band neutrophils with a high level of apoptosis (23). In the present study, the degree of apoptosis in this cell population did not differ significantly from that of the lymphocyte or monocyte populations. These divergent results may be explained by the use of fresh cells in our study. Freezing and thawing procedures can induce apoptosis and may be responsible for the high number of apoptotic cells observed previously (23, 28). EPO did not reduce apoptosis of MNCs, either in peripheral blood or in splenic biopsies. Furthermore, EPO had no effect on the dynamic changes in various lymphocyte subsets or on the decline in total leukocyte counts. The number of apoptotic lymphocytes induced by the endotoxic shock was low, and although PBMC apoptosis was analyzed by two different methods, and we found a good correlation between the methods, we may still not have completely identified all apoptotic cells. The methods used may not be sufficiently sensitive to detect small differences in the number of apoptotic cells. In addition, the circulating apoptotic lymphocytes may have been cleared from circulation and therefore not detected by flow cytometry of PBMCs. To detect apoptotic cells cleared from the circulation, apoptosis in splenic biopsies was quantified; however, apoptotic lymphocytes are readily phagocytosed and digested and therefore the numbers of apoptotic cells detected in the spleen only represent a fraction of the total number of lymphocytes that have undergone apoptosis during the 600 min of endotoxemia. The degree of apoptosis may therefore be underestimated. Splenic biopsies from more time points may be needed to demonstrate an effect; however, this was not included because of the risk of severe bleeding. In addition to the aforementioned limitations, this study is based on an endotoxemic model that does not completely reflect sepsis. In particular, the time course of endotoxemia, which is characterized by a rapid onset, is different from human sepsis. Restricted time of observation is a major limitation when comparing the model  2010 The Authors APMIS  2010 APMIS

EPO ON PBMC APOPTOSIS DURING ENDOTOXEMIA

Caspase-3 positive cells (cells per mm2)

Table 3. The percentages of 7AADdim-positive peripheral blood mononuclear cell (PBMC) subpopulations during endotoxemia. The percentages of apoptotic (7AADdim) lymphocytes, lymphocyte subsets, monocytes, and cells of an emerging myeloid subpopulation that presumably consist of band neutrophils are shown PBMC subpopulations and groups Baseline 60 min 240 min 540 min 7AADdim (% of lymphocytes) Sham 1.2 ± 1.7 1.0 ± 1.6 1.1 ± 1.7 1.6 ± 1.7 LPS 1.6 ± 1.4 1.2 ± 1.4 2.4 ± 1.4 2.0 ± 2.2 LPS + EPO 1.4 ± 1.5 1.1 ± 1.7 2.2 ± 1.4 1.4 ± 1.9 7AADdim (% of CD4+ lymphocytes) 2.0 ± 2.0 1.8 ± 1.9 2.3 ± 1.9 3.3 ± 1.73 Sham1 LPS 2.3 ± 1.9 2.2 ± 1.9 1.8 ± 1.5 1.7 ± 2.1 2.2 ± 1.5 2.1 ± 1.5 1.5 ± 1.43 2.0 ± 2.0 LPS + EPO1,2 7AADdim (% of CD8+ lymphocytes) Sham 2.5 ± 2.0 2.1 ± 2.2 3.1 ± 1.8 4.3 ± 1.6 LPS 2.3 ± 2.0 2.8 ± 1.7 2.6 ± 1.5 2.9 ± 2.6 LPS + EPO 2.6 ± 1.5 2.2 ± 1.6 2.3 ± 1.6 2.5 ± 2.1 7AADdim (% of CD21+ lymphocytes) Sham 0.9 ± 1.6 0.9 ± 1.5 1.0 ± 1.7 1.1 ± 2.0 LPS 0.6 ± 2.4 0.3 ± 1.9 0.8 ± 1.6 0.9 ± 1.8 0.4 ± 1.8 0.3 ± 2.6 0.7 ± 1.5 0.4 ± 2.1 LPS + EPO2 7AADdim (% of monocytes) Sham 1.1 ± 3.0 0.8 ± 2.2 0.8 ± 2.5 0.9 ± 1.7 LPS1 1.3 ± 2.1 2.2 ± 2.5 2.4 ± 2.13 1.8 ± 2.13 1 3 3 LPS + EPO 0.9 ± 2.4 1.6 ± 2.1 2.3 ± 1.8 1.8 ± 1.9 7AADdim (% of myeloid subpopulation) Sham – – – – LPS – – 4.7 ± 2.6 1.9 ± 4.2 LPS + EPO – – 3.7 ± 3.2 0.8 ± 2.8 7AAD, 7-aminoactinomycin D; LPS, lipopolysaccharides; EPO, erythropoietin; ‘–’ indicates that the myeloid subpopulation was not detected in the sham group or in the endotoxemic group before 240 min of endotoxemia. 1 Significant change over time (two-way MANOVA). 2 Significant time ⁄ group interaction compared with sham (two-way MANOVA). 3 Significant compared with baseline (paired t-test).

80

p = 0.03*

60

40

20

0 Sham

LPS

LPS + EPO

Fig. 6. Immunohistochemical detection of activated caspase-3 in splenic biopsies obtained at the end of the experiment. Twenty random vision fields from each biopsy were examined using light microscope (·200). Significantly higher numbers of caspase-3positive lymphocytes were found in the endotoxemic groups compared with the sham group.  2010 The Authors APMIS  2010 APMIS

with human sepsis. However, the cytokine responses, leukopenia, and the increased number of apoptotic lymphocytes in the spleen mimic some aspects of the responses observed in septic patients. The use of recombinant human EPO (rHEPO) instead of porcine EPO is another limitation; however, EPO is highly conserved among mammals and porcine EPO and porcine EPO receptor show a high degree of homology with the human counterparts (29, 30). Importantly, rhEPO stimulates porcine hematopoietic cells (31), and organ-protective effects of rhEPO have been demonstrated in porcine models of ischemia–reperfusion injuries (32–34), demonstrating that rhEPO is biologically active in pigs. Little is known about the significance of lymphocyte apoptosis and whether apoptosis of immune cells plays a physiologic role, for example, in counterbalancing a preponderant proinflammatory response. Mitigating lympho151

SØLLING et al.

A

B

C

Fig. 7. Representative examples of the immunohistochemical detection of activated caspase-3 in biopsies from the spleen obtained at the end of the experiment. Caspase-3-positive cells stain brown. Apoptotic lymphocytes are normally present in the spleen, as demonstrated in biopsies from the sham group (A). At the end of the experiment, the number of caspase-3-positive cells was significantly increased in the endotoxemic groups (B) with no effect of erythropoietin. As a negative control, the primary antibody was omitted (C).

cyte apoptosis to improve the immune system during the immunosuppressive phase of severe sepsis is an intriguing potential therapy, but it is not without concerns. Reducing lymphocyte apoptosis during a predominant proinflammatory response may, in fact, further aggravate the inflammation. In addition, recent reports have raised questions about the safety of rhEPO administration. In patients with acute stroke, treatment with rhEPO resulted in increased mortality (35), and two recent metaanalyses have demonstrated that correcting anemia with rhEPO in patients with cancer may worsen survival (36, 37). rhEPO increases the risk of thrombosis, and caution should be taken when used in patients who are already at an increased risk of thrombosis, such as the critically ill patients. Introduction of rhEPO as an antiapoptotic agent in the treatment of sepsis should, therefore, be followed by careful monitoring of the immune response and adverse effects. The organ-protective and antiapoptotic properties of EPO are supposedly mediated through EPO receptors present on non-hematopoietic cells. However, whether these receptors are actually functionally expressed in the cells has been questioned recently (38–40). The exact physiologic role of EPO outside the bone marrow and whether rhEPO treatment may be beneficial in septic patients remain to be clarified. However, the present study does not support that rhEPO has a role in protection against sepsis-induced lymphocyte apoptosis.

The authors acknowledge the skilled technical assistance of Lene Vestergaard (Department of

152

Anesthesiology and Intensive Care Medicine, University hospital of Aarhus) and Henrik Sørensen (Institute of Clinical medicine, Aarhus University). This study was supported by the A. P. Moeller Foundation for the Advancement of Medical Science (Copenhagen, Denmark), and Holger–Ruth Hesses Foundation (Fredericia, Denmark).

REFERENCES 1. Wang H, Ma S. The cytokine storm and factors determining the sequence and severity of organ dysfunction in multiple organ dysfunction syndrome. Am J Emerg Med 2008;26:711–15. 2. Kox WJ, Volk T, Kox SN, Volk HD. Immunomodulatory therapies in sepsis. Intensive Care Med 2000;26(Suppl 1):124–8. 3. Hartemink KJ, Paul MA, Spijkstra JJ, Girbes AR, Polderman KH. Immunoparalysis as a cause for invasive aspergillosis? Intensive Care Med 2003;29:2068–71. 4. Volk HD, Reinke P, Docke WD. Clinical aspects: from systemic inflammation to ‘immunoparalysis’. Chem Immunol 2000;74:162–77. 5. Rajan G, Sleigh JW. Lymphocyte counts and the development of nosocomial sepsis. Intensive Care Med 1997;23:1187. 6. Felmet KA, Hall MW, Clark RS, Jaffe R, Carcillo JA. Prolonged lymphopenia, lymphoid depletion, and hypoprolactinemia in children with nosocomial sepsis and multiple organ failure. J Immunol 2005;174:3765–72. 7. Toti P, De FC, Occhini R, Schuerfeld K, Stumpo M, Epistolato MC, et al. Spleen depletion in neonatal sepsis and chorioamnionitis. Am J Clin Pathol 2004;122:765–71. 8. Hotchkiss RS, Tinsley KW, Swanson PE, Schmieg RE Jr, Hui JJ, Chang KC, et al. Sepsisinduced apoptosis causes progressive profound depletion of B and CD4+ T lymphocytes in humans. J Immunol 2001;166:6952–63.

 2010 The Authors APMIS  2010 APMIS

EPO ON PBMC APOPTOSIS DURING ENDOTOXEMIA

9. Weber SU, Schewe JC, Lehmann LE, Muller S, Book M, Klaschik S, et al. Induction of Bim and Bid gene expression during accelerated apoptosis in severe sepsis. Crit Care 2008;12:R128. 10. Freire-de-Lima CG, Nascimento DO, Soares MB, Bozza PT, Castro-Faria-Neto HC, De Mello FG, et al. Uptake of apoptotic cells drives the growth of a pathogenic trypanosome in macrophages. Nature 2000;403:199–203. 11. Hotchkiss RS, Chang KC, Grayson MH, Tinsley KW, Dunne BS, Davis CG, et al. Adoptive transfer of apoptotic splenocytes worsens survival, whereas adoptive transfer of necrotic splenocytes improves survival in sepsis. Proc Natl Acad Sci USA 2003;100:6724–9. 12. Hotchkiss RS, Chang KC, Swanson PE, Tinsley KW, Hui JJ, Klender P, et al. Caspase inhibitors improve survival in sepsis: a critical role of the lymphocyte. Nat Immunol 2000;1:496–501. 13. Wesche-Soldato DE, Chung CS, Lomas-Neira J, Doughty LA, Gregory SH, Ayala A. In vivo delivery of caspase-8 or Fas siRNA improves the survival of septic mice. Blood 2005;106:2295– 301. 14. Sharples EJ, Patel N, Brown P, Stewart K, Mota-Philipe H, Sheaff M, et al. Erythropoietin protects the kidney against the injury and dysfunction caused by ischemia-reperfusion. J Am Soc Nephrol 2004;15:2115–24. 15. Ates E, Yalcin AU, Yilmaz S, Koken T, Tokyol C. Protective effect of erythropoietin on renal ischemia and reperfusion injury. ANZ J Surg 2005;75:1100–5. 16. Cuzzocrea S, Di PR, Mazzon E, Patel NS, Genovese T, Muia C, et al. Erythropoietin reduces the development of nonseptic shock induced by zymosan in mice. Crit Care Med 2006;34:1168–77. 17. Patel NS, Sharples EJ, Cuzzocrea S, Chatterjee PK, Britti D, Yaqoob MM, et al. Pretreatment with EPO reduces the injury and dysfunction caused by ischemia ⁄ reperfusion in the mouse kidney in vivo. Kidney Int 2004;66:983–9. 18. Aoshiba K, Onizawa S, Tsuji T, Nagai A. Therapeutic effects of erythropoietin in murine models of endotoxin shock. Crit Care Med 2009;37:889– 98. 19. Kokhaei P, Abdalla AO, Hansson L, Mikaelsson E, Kubbies M, Haselbeck A, et al. Expression of erythropoietin receptor and in vitro functional effects of epoetins in B-cell malignancies. Clin Cancer Res 2007;13:3536–44. 20. Strunk T, Hartel C, Temming P, Matzke N, Zimmer J, Schultz C. Erythropoietin inhibits cytokine production of neonatal and adult leukocytes. Acta Paediatr 2008;97:16–20. 21. Schmid I, Uittenbogaart CH, Keld B, Giorgi JV. A rapid method for measuring apoptosis and dual-color immunofluorescence by single laser  2010 The Authors APMIS  2010 APMIS

22.

23.

24.

25.

26.

27.

28.

29.

30.

31. 32.

33.

flow cytometry. J Immunol Methods 1994;170: 145–57. Lecoeur H, Gougeon ML. Comparative analysis of flow cytometric methods for apoptosis quantitation in murine thymocytes and human peripheral lymphocytes from controls and HIV-infected persons. Evidence for interference by granulocytes and erythrocytes. J Immunol Methods 1996;198:87–99. Ebdrup L, Krog J, Granfeldt A, Tonnesen E, Hokland M. Dynamic expression of the signal regulatory protein alpha and CD18 on porcine PBMC during acute endotoxaemia. Scand J Immunol 2008;68:430–7. Zhang YH, Takahashi K, Jiang GZ, Kawai M, Fukada M, Yokochi T. In vivo induction of apoptosis (programmed cell death) in mouse thymus by administration of lipopolysaccharide. Infect Immun 1993;61:5044–8. Norimatsu M, Ono T, Aoki A, Ohishi K, Takahashi T, Watanabe G, et al. Lipopolysaccharideinduced apoptosis in swine lymphocytes in vivo. Infect Immun 1995;63:1122–6. Haendeler J, Messmer UK, Brune B, Neugebauer E, Dimmeler S. Endotoxic shock leads to apoptosis in vivo and reduces Bcl-2. Shock 1996; 6:405–9. Nielsen JS, Larsson A, Brix-Christensen V, Nyengaard JR, Ledet T, Tonnesen E. Endotoxemiainduced lymphocyte apoptosis is augmented by a hyperinsulinemic-euglycemic clamp. Anesthesiology 2005;102:768–73. Sarkar S, Kalia V, Montelaro RC. Caspase-mediated apoptosis and cell death of rhesus macaque CD4+ T-cells due to cryopreservation of peripheral blood mononuclear cells can be rescued by cytokine treatment after thawing. Cryobiology 2003;47:44–58. Wen D, Boissel JP, Tracy TE, Gruninger RH, Mulcahy LS, Czelusniak J, et al. Erythropoietin structure–function relationships: high degree of sequence homology among mammals. Blood 1993;82:1507–16. Pearson PL, Smith TP, Sonstegard TS, Klemcke HG, Christenson RK, Vallet JL. Porcine erythropoietin receptor: molecular cloning and expression in embryonic and fetal liver. Domest Anim Endocrinol 2000;19:25–38. Kovacevic M, Bozic T, Ivanovic Z. Erythroid progenitor cells from pig bone marrow and peripheral blood. Vet J 1999;158:196–203. Forman CJ, Johnson DW, Nicol DL. Erythropoietin administration protects against functional impairment and cell death after ischaemic renal injury in pigs. BJU Int 2007;99:162–5. Simon F, Scheuerle A, Calzia E, Bassi G, Oter S, Duy CN, et al. Erythropoietin during porcine aortic balloon occlusion-induced ischemia ⁄ reperfusion injury. Crit Care Med 2008;36:2143–50.

153

SØLLING et al.

34. Shimoda M, Sawada T, Iwasaki Y, Mori S, Kijima H, Okada T, et al. Erythropoietin strongly protects the liver from ischemia-reperfusion injury in a pig model. Hepatogastroenterology 2009;56:470–5. 35. Ehrenreich H, Weissenborn K, Prange H, Schneider D, Weimar C, Wartenberg K, et al. Recombinant human erythropoietin in the treatment of acute ischemic stroke. Stroke 2009;40:e647–e656. 36. Tonelli M, Hemmelgarn B, Reiman T, Manns B, Reaume MN, Lloyd A, et al. Benefits and harms of erythropoiesis-stimulating agents for anemia related to cancer: a meta-analysis. CMAJ 2009; 180:E62–E71. 37. Bohlius J, Schmidlin K, Brillant C, Schwarzer G, Trelle S, Seidenfeld J, et al. Recombinant human erythropoiesis-stimulating agents and mortality in patients with cancer: a meta-analysis of randomised trials. Lancet 2009;373:1532–42.

154

38. Sinclair AM, Coxon A, McCaffery I, Kaufman S, Paweletz K, Liu L, et al. Functional erythropoietin receptor is undetectable in endothelial, cardiac, neuronal, and renal cells. Blood 2010;115: 4264–72. 39. Elliott S, Busse L, Bass MB, Lu H, Sarosi I, Sinclair AM, et al. Anti-Epo receptor antibodies do not predict Epo receptor expression. Blood 2006;107:1892–5. 40. Laugsch M, Metzen E, Svensson T, Depping R, Jelkmann W. Lack of functional erythropoietin receptors of cancer cell lines. Int J Cancer 2008;122:1005–11. 41. Haverson K, Saalmuller A, Alvarez B, Alonso F, Bailey M, Bianchi AT, et al. Overview of the third international workshop on swine leukocyte differentiation antigens. Vet Immunol Immunopathol 2001;80:5–23.

 2010 The Authors APMIS  2010 APMIS