Inhibition of lymphocyte blastogenic response in healthy donors treated with recombinant human granulocyte colony-stimulating factor (rhG-CSF): possible role of lactoferrin and interleukin-1 receptor antagonist

Inhibition of lymphocyte blastogenic response in healthy donors treated with recombinant human granulocyte colony-stimulating factor (rhG-CSF): possible role of lactoferrin and interleukin-1 receptor antagonist
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Bone Marrow Transplantation, (1997) 20 , 355–364 ©  1997 Stockton Press All rights reserved 0268–3369/97 $12.00 Inhibition of lymphocyte blastogenic response in healthy donorstreated with recombinant human granulocyte colony-stimulating factor(rhG-CSF): possible role of lactoferrin and interleukin-1 receptorantagonist S Rutella 1,2 , C Rumi 1,2 , U Testa 3 , S Sica 2 , L Teofili 2 , R Martucci 3 , C Peschle 3,4 and G Leone 2 1 Center for the Flow Cytometric Study of Blood Cells and   2  Department of Hematology, Catholic University, Rome;  3  Department of  Hematology and Oncology, Istituto Superiore di Sanita` , Rome, Italy; and   4 Thomas Jefferson Cancer Institute, Thomas JeffersonUniversity, Philadelphia, PA, USA Summary:  The administration of recombinant human granulocyte col-ony-stimulating factor (rhG-CSF) to normal individualsenhances hemopoietic progenitors circulating into periph- The effects of rhG-CSF on lymphocyte blastogenesiswere evaluated in six healthy donors, submitted to pro-  eral blood for subsequent collection and use for syngeneicand allogeneic transplantation (APBPCT) or for graft rejec- genitor cell mobilization for allogeneic transplantation.Neutrophil, monocyte and lymphocyte count increased  tion and relapse after T cell non-depleted marrow trans-plants. 1,2 In a recent survey, 3 although 1–2 logs more T 6.7-fold, 5.3-fold and 2.0-fold on day  + 4 of rhG-CSF ascompared with baseline. The DNA stimulation index  lymphocytes were routinely infused with unmanipulatedperipheral blood, the incidence and severity of acute graft- (DNA SI) of 72 h phytohemagglutinin (PHA)-treatedcultures decreased from 20% (15–35.5) prior to rhG-  versus-host disease (aGVHD) were comparable to marrowinfusion, with approximately half of the patients developing CSF to 6.7% (1.5–11.9;  P  0.0026), 8% (4–12; P  0.0091) and 15% (9–22;  P  0.0091) on days  + 2,  + 4  grade II or greater and 17% grade III–IV aGVHD. Russell et al 4 found a 37% incidence of grade II–IV aGVHD in and   6; similarily, reactivity to concanavalin Adecreased from 18% (12–20) to 1.8% (0.5–7;  P  0.01),  26 patients receiving unmanipulated allogeneic PBPC fromHLA-matched siblings and a similar incidence of chronic 3% (2–8;  P  0.01) and 5% (2–11;  P  0.009). Nochanges of lymphocyte response to pokeweed mitogen  GVHD at 1 year; Rosenfeld  et al 5 and Urbano-Ispizua  et al 6 reported an actuarial incidence of 37% for grade II–IV were observed. DNA SI of PHA-treated culturesinversely correlated with neutrophil and monocyte  aGVHD following progenitor cell mobilization with rhG-CSF. It has also been suggested that the cryopreservation count. IL-1 receptor antagonist (IL-1ra) and lactoferrin(Lf) plasma levels sharply increased and correlated with  of the graft may selectively deplete or induce anergy in Tlymphocytes, thus accounting for the absence of life- neutrophil and monocyte count. IL-10 increased five-fold on day   2, returned to pretreatment values there-  threatening aGVHD. 7 A recent retrospective analysis revealed a greater inci- after and did not show any correlation with DNA SI,suggesting that it was not responsible for the observed  dence of chronic GVHD and elevated TNF-   plasma levelsin patients not receiving rhG-CSF after allogeneic marrow phenomena. Interestingly, DNA SI of PHA-treated cul-tures inversely correlated with IL-1ra and Lf levels.  infusion (80% as compared with 18% in recipients givenrhG-CSF); 8 moreover,  in vitro  studies showed a down- CD3  and CD19  lymphocyte activation status, ieCD23, CD25, CD30 and HLA-DR coexpression, was not  modulation of allogeneic immune responses by G-CSF viainhibition of TNF-   production. 9 affected by rhG-CSF administration. Pharmacologicaldoses of rhG-CSF in healthy donors inhibit lymphocyte  The  in vivo  effects of rhG-CSF on peripheral blood lym-phocytes remain controversial. 10,11 We recently reported a blastogenesis via an increased production and/or releaseof immunoregulatory soluble mediators, ie IL-1ra and  significant elevation of lymphocyte and monocyte count inhealthy subjects receiving 16   g/kg/day rhG-CSF. 12,13 Lf, by primed neutrophils and monocytes.Keywords:  G-CSF; lymphocyte blastogenesis; flow Although a suppressed response to PHA due to aphereticprocedures was observed in peripheral stem cell collectionscytometryof patients with non-Hodgkin lymphoma, 14,15 no report, toour knowledge, focused on lymphocyte reactivity to mito-gens in healthy subjects treated with rhG-CSF for allo-geneic progenitor cell transplantation.In the present study, lymphocyte blastogenic responses Correspondence: Dr S Rutella, Center for the Flow Cytometric Study of  to phytohemagglutinin (PHA), concanavalin A (Con A) and Blood Cells, Department of Hematology, Istituto di Semeiotica Medica, pokeweed mitogen (PWM) were evaluated by flow cytome- Catholic University Medical School, Largo A Gemelli 8, 00168 Rome, try and propidium iodide staining in six normal donors; 16 ItalyReceived 27 December 1996; accepted 11 May 1997  results were correlated with plasma levels of immunoregul-  G-CSF inhibits lymphocyte blastogenesis in healthy donors S Rutella  et al  356 atory soluble mediators, ie IL-1 receptor antagonist (IL- tration and on days  + 2,  + 4,  + 6 and  + 30. Maximum lympho-cyte recruitment and proliferation in response to lectins1ra), lactoferrin (Lf) and interleukin-10 (IL-10). In addition,interleukin-8 (IL-8) plasma levels were measured; recent occur after prolonged stimulation; 20 consequently, 250  lof unseparated heparin-anticoagulated blood were culturedinvestigations showed an increase of endogenous IL-8 fol-lowing granulocyte–macrophage (GM)-CSF administration for 72 h at 37 ° Cin 5% CO 2  atmosphere in mitogen-contain-ing lyophilized culture medium, reconstituted in sterile dis-to healthy volunteers, responsible for sustained neutrophilactivation, prolonged intravascular neutrophil circulation tilled water (Blastest; Ylem, Rome, Italy). Combined cellsurface antigen and DNA staining was performed as fol-time and release of Lf and elastase. 17 lows: 21 culture supernatants were removed by centrifug-ation at 300  g  for 5 min in PBS-1%BSA and cells wereincubated for 30 min at 4 ° C with a proper dilution of FITC- Materials and methods conjugated anti-CD3 mAb. After erythrocyte lysis, cellswere fixed for 1 h in cold PBS containing 0.25% paraform-  Donors’ characteristics aldehyde (final concentration). After centrifugation, cellsSix HLA-matched healthy donors (three males, threewere permeabilized for 15 min at 37 ° C in 1 ml of 0.2%females, median age 38 years, range 30–50) were submittedTween 20 in PBS and stained for 1 h on ice with 1 ml of to a 5-day course of subcutaneous rhG-CSF (16  g/kg/day,PBS containing 10  g/ml propidium iodide (PI), 11.25Granocyte; Rhoˆne-Poulenc Rorer, Milan, Italy) to mobilizeKunitz units RNAse and 0.1% sodium azide (Sigma-Ald-hemopoietic progenitors for APBPCT or to recruit donorrich, Milan, Italy).leukocytes for leukemia in relapse after allogeneic marrowtransplantation. All donors gave written informed consent Flow cytometry and the investigations were approved by the InstitutionalHuman Research Committee. Dose and schedule of rhG-All samples were run through a FACScan flow cytometerCSF administration were elected according to previously(Becton Dickinson) equipped with an argon laser emittingpublished studies. 3,18 No donor experienced serious side-at 488 nm. FITC and PE signals were measured at 530 nmeffects except for moderate back pain.and 575 nm respectively; spectral overlap was minimizedby electronic compensation with calibrite beads (BectonDickinson) before each determination series.  Lymphocyte phenotype and activation status For surface lymphocyte phenotyping, a minimum of Peripheral blood samples were drawn 30 min before rhG- 3500 CD45 +++ events was acquired in list mode usingCSF daily administration for 6 consecutive days. Absolute CellQuest software (Becton Dickinson); forward and sideand differential cell counts were measured using an auto- scatter were collected as linear signals and all fluorescentmated hematologic analyzer. emissions were collected on a four-decade logarithmicFor flow cytometric determinations, 100  l of EDTA- scale.anticoagulated sample were incubated for 30 min at 4 ° C For DNA analysis, a minimum of 10 000 CD3 + eventswith proper dilutions of the following fluorescein isothy- was acquired, setting the phycoerythrin emission channelocyanate (FITC) or phycoerythrin (PE) conjugated mono- to a linear scale for PI fluorescence and excluding debrisclonal antibodies (mAb): CD45 (2D1 clone, IgG 1 ), CD14 and aggregates. Calculations of cell cycle compartments(MøP9 clone, IgG 2b ) CD4 (SK3 clone, IgG 1 ), CD3 (SK7 were performed on a 1024-channel histogram with ModFitclone, IgG 1 ), CD8, (SK1 clone, IgG 1 ) CD19 (SJ25C1 clone, LT DNA content analysis software (Becton Dickinson).IgG 1 ), CD16 (NKP15 clone, IgG 2b ), CD56 (MY31 clone, The mean coefficients of variation (CVs) for DNA euploidIgG 1 ), CD23 (EBVCS-5 clone, IgG 1 ), CD25 (2A3 histograms, a reliable measure of S-phase accuracy, wereclone, IgG 1 ), HLA-DR (L243 clone, IgG 2a ), CD11a (G25.2 always   8%, as recently recommended. 22 S-phase dimen-clone, IgG 2a ) (Becton Dickinson, CA, USA) and CD30 sion was expressed as DNA stimulation index (SI: % S-(Ber-H2 clone, IgG 1 , Dako, Glostrup, Denmark). Negative phase in stimulated cultures  −  % S-phase in unstimulatedcontrols consisted of cells incubated under the same experi- samples).mental conditions with isotype-matched FITC- and PE-con- jugated irrelevant mAb. After erythrocyte lysis, cells were G-CSF, IL-1ra, Lf, IL-10 and IL-8 determination washed in PBS supplemented with 0.1 m M  EDTA and0.02% NaN 2  (PBS-EDTA) and kept on ice until flow cyto- Plasma samples were obtained before rhG-CSF adminis-metric examination. tration and daily for 6 consecutive days. Blood was drawnCD4 + T cell clones with established profiles of cytokine in EDTA and plasma was separated by centrifugation (15secretion (helper type-1 (Th1) and type-2 (Th2)), generous min at 4000 r.p.m.) shortly after collection, aliquoted andgift of G Del Prete, Florence, Italy), were used as negative stored at  − 80 ° C until used.and positive controls respectively for CD30 staining. 19 Plasma concentrations of G-CSF, IL-1ra, Lf, IL-8 andIL-10 were evaluated by sensitive and specific immuno-assays (R&D System, British Biotechnology, Cowley, UK).  Lymphocyte blastogenesis The detection thresholds were 10 pg/ml for IL-1ra and IL-8, 20 ng/ml for Lf and 5 pg/ml for IL-10 and G-CSF. EachFor the evaluation of lymphocyte reactivity to mitogens(PHA 5  g/ml, ConA 5  g/ml, PWM 5  g/ml), peripheral cytokine level represents the mean value observed in threeseparate determinations. The intra-assay variability for theblood samples were obtained prior to rhG-CSF adminis-  G-CSF inhibits lymphocyte blastogenesis in healthy donors S Rutella  et al  357 various cytokine determinations was 5 to 10%. Details of   Lymphocyte phenotype and activation status plasma cytokine evaluation were extensively reportedA statistically significant increase of CD3 + and CD19 + lym-elsewhere. 23 phocytes occurred after rhG-CSF administration peakingon day  + 4. Similarily, CD4 + , CD8 + and NK cells Statistical methods  (CD3 − CD16 + CD56 + ) increased in all donors; consequently,the CD4 +  /CD8 + ratio was not affected (Table 1).To test the approximation of population distribution to nor-Activated B and T lymphocytes, measured by their co-mality, y (g) statistics (for symmetry and kurtosis testing)expression of HLA-DR, CD25 and CD23 antigens, showedas well as probability plots were used. Results were asym-no significant modification following rhG-CSF treatment.metrically distributed and, consequently, were presented asA minor subset (2–3%) of CD4 + T lymphocytes co-median values and interquartile ranges. Correlations wereexpressed HLA-DR activation antigen prior to rhG-CSFexplored by Spearman rank analysis (Rs) and all compari-sons were performed with Wilcoxon W test for paireddeterminations. The criterion for statistical significance was Table 1  Flow cytometric analysis of lymphocyte subsets in healthy defined as  P  0.05. donors treated with rhG-CSF 16  g/kg/day  Baseline Peak P value Results CD3 + 1.43 (0.78–1.78) 2.24 (1.14–4.58)   0.0001CD19 + 0.2 (0.1–0.32) 0.56 (0.22–1.39)   0.0001 Changes in leukocyte subsets CD4 + 0.75 (0.47–0.96) 1.32 (0.6–2.52)   0.0001CD8 + 0.63 (0.25–1.09) 1.12 (0.54–2.13)   0.0001 White blood cell, neutrophil, monocyte and lymphocyte CD4/CD8 ratio 1.07 (0.66–2.28) 1.19 (0.78–2.09) NS count progressively increased in all donors and showed a CD3 − CD16 + CD56 + 0.29 (0.18–0.7) 0.60 (0.32–0.86)   0.0001 peak on day  + 4 of rhG-CSF treatment, corresponding toa 5.4-fold (range 2.9–9.5,  P  =  0.0001), 6.7-fold (3.4–10.2, As compared with baseline values, all lymphocyte subsets showed a stat- P  =  0.002), 5.3-fold (2.5–5.9,  P  =  0.0075) and 2.0-fold istically significant increase, reaching a peak value on day  + 4 of rhG- (1.1–2.3,  P  =  0.0014) elevation respectively, as compared  CSF treatment. Median values and ranges; Wilcoxon W test for paireddeterminations; NS  =  not significant. with baseline levels (Figure 1). **** ****** ****************** ****** **** **** ***** 4030201002. 6543210 Days DaysLymphocytesNeutrophilsMonocytesWBC    C  e   l   l  s    ×     1   0     3     /   l      µ    C  e   l   l  s    ×     1   0     3     /   l      µ Figure 1  White blood cell, neutrophil, monocyte and lymphocyte counts in normal donors receiving rhG-CSF 16  g/kg/day. All results were presentedas median values. * P  0.01 and ** P  0.001 as compared with baseline values (Wilcoxon W test for paired determinations).  G-CSF inhibits lymphocyte blastogenesis in healthy donors S Rutella  et al  358 administration and this percentage did not increase during cytes increased from 58% (55–77) to 61% (59–79) on day + 4 of rhG-CSF treatment ( P =  NS; Figure 2a), indicatingthe treatment course (Figure 2a). CD4 + CD25 + cellsincreased from 36% (26–54) to 40% (38–54) on day  + 4 that  ex vivo  lymphocyte activation status was not modifiedas compared with pretreatment levels.( P  =  NS); similarily, CD19 + CD23 + activated B lympho- 10080604020010080604020065432100 1 2 3 4 5 6 Days CD30 FITC    %    C   D   4           +    /   H   L   A  -   D   R           +    %    C   D   4           +    /   C   D   2   5           +    %    C   D   1   9           +    /   C   D   2   3           +    C   D   4   P   E 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 10 4 10 3 10 2 10 1 10 0 Day 0Day 2Day 4Day 6 a b Figure 2  ( a ) Expression of activation markers CD23, CD25 and HLA-DR on CD4 + and CD19 + lymphocytes; median values and interquartile ranges.No statistically significant differences were detected following rhG-CSF administration as compared with baseline values. ( b ) Expression of CD30 antigenon CD4 + lymphocytes from a representative subject; a negligible fraction of CD4 + CD30 + cells was detected both prior to rhG-CSF administration andduring treatment course.  G-CSF inhibits lymphocyte blastogenesis in healthy donors S Rutella  et al  359 Type-2 cytokine-producing (Th2) CD30 + CD4 + lympho-cytes increased from 0.16% (0–0.3) to 0.2% (0–0.4) on day + 4 of rhG-CSF administration (Figure 2b;  P  =  NS), sug-gesting that rhG-CSF did not shift peripheral lymphocyteto a Th2-like cytokine profile, at least as measured by thecoexpression of CD30 antigen.More than 98% of CD4 and CD8 T lymphocytes co-expressed CD11a leukocyte integrin prior to rhG-CSFadministration and no modifications of percent CD11a + cells or CD11a staining intensity occurred during the rhG-CSF course.  Lymphocyte blastogenesis Lymphocyte transformation in response to lectins can beevaluated in a standardized and reproducible manner bymeasuring the relative DNA content of propidium iodide-stained nuclei. 16 In our study, median CVs, a reliable meas-ure of S-phase accuracy, were always  3% in unstimulatedcultures and varied from 3.9 (3.8–4.3) prior to rhG-CSF to3.5 (2.7–4.8), 3.4 (3.2–3.7) and 3.4 (2.8–4.1) on days  + 2, + 4 and  + 6, respectively.DNA SI of PHA-treated cultures markedly decreasedfrom 20% (15–35.5) to 6.7% (1.5–11.9,  P  =  0.0026), 8%(4–12,  P  =  0.0091) and 15% (9–22,  P  = 0.0091) respect-ively on days  + 2,  + 4 and  + 6 of rhG-CSF treatment. TheDNA SI of PHA-treated lymphocytes from a representativesubject are shown in Figure 3. However, the addition of rhG-CSF to PHA-stimulated cultures of healthy subjectsnot receiving exogenous rhG-CSF did not modify theblastogenic response as evaluated in terms of percent S-phase (data not shown), suggesting that the observed  in vivo inhibition of lymphocyte proliferation is not directlymediated by rhG-CSF.Responsiveness to ConA decreased from 18% (12–20)prior to rhG-CSF to 1.8% (0.5–7;  P  0.01), 3% (2–8; P  0.01) and 5% (2–11;  P  =  0.009) on days  + 2,  + 4 and + 6 respectively (data not shown). No significant changesof cell cycle distribution were observed in PWM-stimulatedcultures, suggesting that the B lymphocyte compartmentwas not affected by rhG-CSF treatment (data not shown).All modifications were completely abrogated 3 weeks afterrhG-CSF discontinuation.The DNA SI of PHA-stimulated lymphocytes showed an 50040030020010010008006004002000 10008006004002000 40032024016010008006004002000 30024018012010008006004002000 BaselineDay +2Day +4Day +6DNA SI = 36%DNA SI = 2%DNA SI = 3%DNA SI = 6%    R  e   l  a   t   i  v  e  c  e   l   l  n  u  m   b  e  r Relative DNA content 18015012090603000800600 inverse correlation with neutrophil (Rs  = − 0.75, Figure 3  Histogram analysis of relative DNA content (propidium iodide P  =  0.0008) and monocyte counts (Rs  = − 0.64;  P  =  0.007;  staining) of a representative subject given rhG-CSF; S-phase percentagesof gated peripheral blood CD3 + lymphocytes markedly declined in all Figures 4a and b), indicating that the inhibition of lympho- donors following rhG-CSF administration. DNA stimulation index  =  S- cyte blastogenesis was maximal in coincidence with the phase of PHA-stimulated cultures  −  S-phase of unstimulated cultures. maximal expansion of the neutrophil and monocytecompartments, induced by rhG-CSF.130 pg/ml to 420 pg/ml ( P =  0.022), 350 pg/ml ( P =  0.009)and 420 pg/ml ( P  =  0.0013; Figure 5). IL-10 levels G-CSF, IL-1ra, Lf, IL-10 and IL-8 levels increased from 5 pg/ml to 25 pg/ml on day  + 2 ( P  =  NS),returned to pretreatment values thereafter and showed noG-CSF plasma levels consistently increased followingexogenous rhG-CSF administration and returned to pre- significant correlation with DNA SI (Rs  = − 0.22).IL-ra and Lf plasma levels positively correlated withtreatment values 24 h after rhG-CSF discontinuation(Figure 5). neutrophil count (Rs =  0.75,  P  =  0.0001 and Rs  =  0.67, P  =  0.0003 respectively, Figure 6a and b); similarly, IL-1raIL-1ra plasma concentration increased from 198.6 pg/mlto 1220 pg/ml ( P  =  0.0023), 2871 pg/ml ( P  = 0.0020) and plasma levels showed a strong correlation with monocytecount (Rs  =  0.65,  P  =  0.0003; Figure 6c).2937.8 pg/ml ( P  =  0.0049) on days  + 2,  + 4 and  + 6 of rhG-CSF treatment, respectively; similarly, Lf increased from Interestingly, IL-1ra and Lf inversely correlated with
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks