Characterization of Buffy Coat-Derived Granulocytes for Clinical

UvA-DARE (Digital Academic Repository)

Neutrophil defects and deficiencies van de Geer, A.

Publication date 2020 Document Version Other version License Other Link to publication

Citation for published version (APA): van de Geer, A. (2020). Neutrophil defects and deficiencies.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:01 Oct 2021 Chapter 6 1

Characterization of buffy-coat-derived granulocytes for clinical use: a 2 comparison with granulocyte-colony stimulating factor/dexamethasone- pretreated donor-derived products. 3

Annemarie van de Geer1, Roel P. Gazendam1, Anton T.J. Tool1, John L. van Hamme1, Dirk de Korte1,2, Timo K. van den Berg1, Sacha S. Zeerleder3,4 and Taco W. Kuijpers1,5. 4 1. Dept. of Blood Cell Research, Sanquin Research, Amsterdam, The Netherlands 2. Dept. of Product and Process Development, Sanquin Blood Bank, Amsterdam, The Netherlands 3. Dept. of Immunopathology, Sanquin Research, Amsterdam, The Netherlands 4. Dept. of Hematology, Academic Medical Center, University of Amsterdam, The Netherlands 5. Dept. of Pediatric Hematology, Immunology & Infectious disease, Emma Children’s Hospital, Academic Medical Center, University of 5 Amsterdam, The Netherlands

Vox Sang. 2017 Feb;112(2):173-182

A Chapter 6

Abstract

Background and objectives Buffy coat-derived granulocytes have been described as an alternative to the apheresis product from donors pretreated with dexamethasone and granulocyte-colony-stimulating factor (G-CSF). The latter is – dependent on the local and national settings - obtained following a demanding and time-consuming procedure, which is undesirable in critically ill septic patients. In contrast, buffy coat-derived products have a large volume and are often heavily contaminated with red cells and platelets. We developed a new pooled buffy coat-derived product with high purity and small volume, and performed a comprehensive functional characterization of these granulocytes.

Materials and methods We pooled ten buffy coats following the production of platelet concentrates. Saline 0.9% was added to decrease the viscosity and the product was split into plasma, red cells and a “super” buffy coat. Functional data of the granulocytes were compared to those obtained with granulocytes from healthy controls and G-CSF/dexamethasone-pretreated donors.

Results Buffy coat-derived granulocytes showed adhesion, chemotaxis, reactive oxygen species production, degranulation, NETosis and in-vitro killing ofStaphylococcus aureus, Escherichia coli, and Aspergillus species comparable to control and G-CSF/dexamethasone-derived granulocytes. Candida killing was superior compared to G-CSF/dexamethasone-derived granulocytes. Immunophenotyping was normal, especially no signs of activation in the buffy coat-derived granulocytes were seen. Viability was reduced. Buffy coats are readily available in the regular blood production process and would take away the concerns around the apheresis product.

Conclusion The product described appears a promising alternative for transfusion purposes.

Key words: neutrophil, granulocyte, transfusion, buffy coat, G-CSF, dexamethasone

142 Buffy coat – derived granulocytes for clinical use

Introduction

Life-threatening bacterial and fungal infections are a major problem in patients using medi- cations with immunosuppressive effects, like chemotherapy. Reasons are intensified chemotherapy regimens leading to prolonged severe neutropenia and rising resistance against antimicrobial agents. This results in mortality rates up to 90% in patients with invasive fungal infections.1-3 Administration of donor granulocytes in addition to antimicrobial drugsto patients with refractory infections may be a last resort, but data from large randomized controlled trials demonstrating its benefits are lacking. Nevertheless, there are data suggesting less infection-related mortality after granulocyte transfusions4-8. To obtain enough granulocytes for donation, donors are pretreated with granulocyte-colony-stimulating factor (G-CSF) and dexamethasone, or with steroids alone9 for granulocyte mobilization from the bone marrow prior to leukapheresis (from the G-CSF ánd dexamethasone pretreated donors: “GTX”). We have previously shown that these granulocytes function well10, although there is a significant partial in-vitro Candida killing defect11. Also, donor recruitment may be demanding, entails the risk of too few available donors and can be considered unethical in countries where patients are responsible for their own donor recruitment. 6 Dependent on the local and national settings, donors undergo extensive medical investiga- tion before pretreatment and are allowed to donate up to 4-5 times, whereas the recipients usually need more transfusions. To improve separation of blood components during leukapheresis low amounts of Hydroxyethyl Starch (HES) are used in many countries entailing the risk of fluid retention, coagulopathy and allergic reactions12,13. Much higher volumes of HES have been proven to cause higher morbidity and mortality rates in critically ill patients and are therefore not used in a clinical setting anymore14,15. To bypass any of these concerns, granulocytes from buffy coats (BCN) might be an alternative source for transfusion. BCN are readily available in the blood production process, and there is already some evidence that BCN are an appropriate alternative to GTX16. This study provides a new BCN product, with high purity and small volume. It builds upon work of Bashir et al16 to use granulocytes from pooled buffy coats for transfusion and by our knowledge, it is the first to offer an extensive comparison between the functionality of BCN, GTX and control granulocytes. We found that BCN functions are equivalent to those of GTX and controls.

Materials and methods

Pooled buffy coat preparation, GTX donor pretreatment Thrombocyte concentrates were produced from overnight stored whole blood (room temperature) derived from healthy voluntary donors at the Dutch national blood bank, Sanquin.

143 Chapter 6

After this, buffy coat pools from 5 different donors were left. Two of these poolswere merged (according to the methods as described by others17) and 250 ml Saline 0.9% was added to decrease the viscosity and improve blood component separation. The suspension was centrifuged (700g, room temperature, 4 minutes and 15 seconds) and separated in red cells, a ‘super’ buffy coat (our transfusion product) and a supernatant, with a Compomat G5 system (Fresenius Kabi, Bad Homburg, Germany). Granulocytes isolated from the ‘super’ buffy concentrates were used for the experiments. Heparinized blood samples from GTX donors were obtained from different donors by venipuncture at the moment of donation. These donors had received 5 µg/kg G-CSF sub- cutaneously and 8 mg dexamethasone orally 12-16 hours prior to donation. Heparinized control blood samples were collected by venipuncture from healthy volunteers at Sanquin. Buffy coats and samples from GTX donors were obtained from volunteer donors at Sanquin after written informed consent was obtained. Control samples were obtained via an internal system at Sanquin after informed consent was given and after consulting the Medical Ethical Committee from the Academical Medical Center Amsterdam. All procedures are conducted in accordance with the 1975 declaration of Helsinki as revised in 2013.

Granulocyte isolation and cell counts We have isolated granulocytes from the ‘super’ buffy coat concentrates and from the heparinized blood samples from GTX donors at time of donation and controls. These cells were used for all experiments and will be called BCN-derived, GTX-derived, and control granulocytes. Our previous studies have shown that there is no difference in results as to whether we isolate granulocytes from the GTX product bag or after a separate venipuncture (data not shown). Granulocyte isolation was performed with a Percoll gradient of 1.076 g/ml for the GTX and controls and 1.074 g/ml for the BCN concentrates, as this last cell suspension is 18 hours old. Cells were centrifuged (800g, 21°C, 20 minutes) and erythrocytes in the pellet were lysed twice with isotonic NH4Cl-KHCO3-EDTA solution at 4°C. Thereafter, the solution was centrifuged (500g, 4°C, 5 minutes) and cells were resuspended in incubation medium (132 mM

NaCl, 6 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 1.2 mM potassium phosphate, 20 mM HEPES, 5.5 mM glucose and 0.5% (w/v) human serum albumin, pH 7.4) on room temperature. Purity of the collected granulocytes was >95%, as judged by cytospin. Granulocyte counts were determined with an automated cell counter (Sysmex XT-2000iV, Kobe, Japan). Samples were taken before and after pooling, after irradiation and after component separation for the determination of cell recovery.

Granulocyte immunophenotyping Expression of granulocyte surface markers was analyzed by direct flow cytometry (FACS) in which fluorescein isothiocyanate (FITC)- or phycoerythrin (PE)- labelled monoclonal antibod-

144 Buffy coat – derived granulocytes for clinical use ies were used according to the manufacturer’s instructions. CD9, CD11b, CD18, CD55, CD59 and CD66b were from Sanquin (Sanquin Reagents, Amsterdam, Netherlands); EMR3, CD32, CD64 and CD177 were from AbD Serotec (Serotec, Oxford, UK); L-selectin and CD16 were from BectonDickinson (BD PharMingen, San Diego, CA). Cells were used at a concentration of 5x106/ml. Per sample, 10,000 gated events were collected, and cells were gated based on their forward and side scatter.

Functional characteristics of isolated granulocytes To assess granulocyte adhesion, granulocytes (2x106/ml) were labeled with calcein-AM (1 μM) (Molecular Probes, Leiden, Netherlands) and incubated with different stimuli in an uncoated 96-well Maxisorp plate (Thermo Fisher, Roskilde, Denmark) for 30 min at 37°C. After washing with PBS, Triton X-100 (0.5% v/v) was added, and cell adhesion was assessed by measuring fluorescence with the Infinite F200-pro plate reader (Tecan, Männedorf, Switzerland). Chemotaxis was measured with Fluoroblock inserts in a Falcon 24-well plate at a granulocyte concentration of 5x106/ml (Corning Inc., Corning, NY, USA), as described earlier18. Granulocyte NADPH oxidase activity after stimulation with zymosan (1 mg/ml), serum- 6 treated zymosan (STZ, 1 mg/ml), phorbol-myristate acetate (PMA, 100 ng/ml) and Platelet Activating Factor (PAF, 1 uM) followed by formyl-methionyl-leucyl-phenylalanine (fMLP, 1 uM) (all from Sigma-Aldrich, St Louis, MO) was assessed by measuring hydrogen peroxide 6 (H2O2) release with an Amplex Red kit at a granulocyte concentration of 1x10 /ml (Life Tech- nologies Carlsbad, CA), as described before18. Protease release after degranulation was measured by degradation of fluorescent DQ- green-BSA (bovine serum albumin) (Life Technologies). Granulocytes (2x106/ml) in incubation medium were incubated with DQ-BSA and 1 µM PAF or 5 ug/ml cytochalasin B for 5 min at 37°C. Then, cells were stimulated with 1 µM fMLP or 100 ng/ml PMA (all from Sigma- Aldrich, St. Louis, MO). An unstimulated control value and a 100% content value with Triton X-100 (1% w/v) were determined. Degranulation was assessed with the Infinite F200-pro plate reader (Tecan).

NET formation 250.000 Granulocytes were incubated on cover glasses (12 mm, Braunschweig, Germany) in a sterile 24-well plate (Nuclon Delta Surface, Thermo Fischer) at 37°C, 5% CO2 for 10 minutes. Then PMA (100 ng/ml) was added, cells were incubated for 3 hours (37°C, 5%

CO2) and fixed with 3.7% (w/v) paraformaldehyde (PFA) in PBS. After overnight storage (4°C) cells were incubated with mouse anti-myeloperoxidase (MPO) (Abcam, Cambridge, UK) and rabbit anti-elastase (neutrophil elastase, NE) (Sanquin) 1:500 in PBS/BSA for 30 minutes (room temperature). After that, NET components MPO, NE and DNA were stained red, green and blue, respectively, with secondary antibodies anti-mouse (Goat anti-Mouse 633

145 Chapter 6

RD, 926-68070, Li-Cor, Bad Homburg, Germany), anti-rabbit (Goat anti-rabbit IgG, 488, Life technologies) 1:500, and Hoechst (Sigma-Aldrich) 1:40,000. NET formation was quantified by confocal microscopy (TCS SP8, Leica, Wetzlar, Germany) as described by others19. In short, for all groups, four fields of 2922x2922 µm were captured using a 40x objective lens. The same contrast was used for all images. Images from individual color channels (red for MPO, green for NE and blue for DNA) were exported to ImageJ (NIH). Contrast was adjusted to minimize background autofluorescence and the same fluorescent threshold was applied in all experiments. MPO and NE stained areas were measured as the percentage of image area covered by positive fluorescence staining per field with ImageJ.

Microbial killing Bacterial killing was determined as described by others20. In short, Eschericia coli (strain ML-35) and Staphylococcus aureus (strain Oxford) were grown aerobically at 37°C in Luria

Bertani (LB) Broth for three hours and brought to an OD600 of 1.0 in PBS. The bacteria were opsonized with 10% (v/v) human pooled serum (37°C, 15 minutes) before incubation with granulocytes (5x106/ml) at a 5:1 ratio. At different timepoints, 50-μl samples were added to

2.5 ml of H2O/NaOH pH11, after which the samples were applied on LB Agar plates. After overnight incubation (37°C, 5% CO2) CFU’s were counted. Granulocyte killing of Candida albicans (strain SC5314) conidia was analyzed as described earlier11. Aspergillus fumigatus conidia (clinical isolate) were incubated overnight (37°C, 5%

CO2 ) in medium (RPMI-1640 without phenol red (Life Technologies), with 10% (v/v) FCS and 2 mM glutamine) for hyphae outgrowth. The next day, hyphae were opsonized with 10% (v/v) human pooled serum (15 minutes, 37°C) and granulocytes (0-1x105 cells) were incubated on the hyphae for 1 hour at 37°C. Granulocytes were lysed with H2O/NaOH pH11 and incubated with MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; thiazolyl blue) overnight at room temperature. Acidic isopropanol (isopropanol + 50mM HCl) was added and a plate reader (Infinite F200-pro, Tecan) was used to measure the optical density at 570 nm.

Granulocyte viability

In a sterile 96-well plate, 100.000 cells were incubated overnight (37°C, 5% CO2) with either incubation medium or G-CSF (10 ng/ml). At desired timepoints, cells were washed and resuspended in incubation medium with extra Calcium (2.5 mM). The percentage of viable cells was determined by AnnexinV-Fitc (1:300) (BD Pharmingen, Franklin Lakes, NJ) and Hoechst (1:30.000) (Sigma-Aldrich) staining on flow cytometry.

Statistics First, an outlier analysis was performed. After removing outliers, all data were normally distributed. Hence, a one-way ANOVA with post-hoc Tukey unpaired t-test was performed,

146 Buffy coat – derived granulocytes for clinical use comparing at all times two groups (BCN vs GTX, control vs BCN, control vs GTX-derived granulocytes). Data were considered significant when p<0.05. Statistical analysis was -per formed with SPSS 23 (IBM, Armonk NY).

Results

Product characteristics. Granulocyte counts in single buffy coats were 1.0 ± 0.2 x 109 (N=9). One remainder from five pooled buffy coats contained 3.8 ± 1.5 x 109 granulocytes (N=20). After pooling two remain- ders and generation of a new ‘super’ buffy coat, i.e. our transfusion product, the granulocyte yield was 5.6 ± 1.5 x 109 granulocytes in 56.1 ± 6 ml (N=8; all counts are in mean ± SD).

Normal expression of surface markers, adhesion and chemotaxis by buffy coat granulocytes. Surface markers were analyzed by flow cytometry on BCN-, control- and GTX-derived granulocytes. Immunophenotyping demonstrated normal expression of maturation marker EMR3, 6 adhesion molecule L-selectin (CD62L), integrin CR3 (CD11b/CD18) and the Fcɣ-receptors in BCN-derived granulocytes compared to control granulocytes. On GTX-derived granulocytes EMR3, L-selectin and Fcɣ-receptor IIIb (CD16) were reduced. DAF (CD55), CD66b, CD63 and CD64 were elevated in GTX- compared to BCN-derived- and control granulocytes. No elevation of activation markers was seen on the cell membrane of BCN-derived granulocytes (Figure 1). In line with unaltered expression of CR3 and Fcɣ-receptors, BCN normally adhered to various stimuli (Figure 2A). Chemotaxis of BCN-derived granulocytes was only slightly reduced towards IL-8, whereas the unstimulated BCN- and GTX-derived granulocytes already demonstrated some activity (Figure 2B).

Intact NADPH oxidase activity and degranulation in BCN-derived granulocytes Granulocytes possess two mechanisms for microbial killing. Oxidative killing is carried out by the NADPH oxidase system producing Reactive Oxygen Species (ROS), including hydrogen peroxide (H2O2). Non-oxidative killing largely depends on the proteolytic content of cytoplas- mic granules. A third, not irrefutably proven killing mechanism is Neutrophil Extracellular

Traps (NETs) formation. Examination of H2O2 production showed priming of fMLP-induced NADPH-oxidase activity by BCN-derived granulocytes (Figure 2C). Granulocyte protease release was comparable in BCN-, control- and GTX-derived granulocytes (Figure 2D). PMA- induced NET formation was also comparable in all groups (Figure 3).

147

Chapter 6

8 0 0 0 0 * * B C N C o n tro ls 6 0 0 0 0 G T X * * 4 0 0 0 0 * * *

* I 2 0 0 0 0 * F

M 1 0 0 0 0 * * * 8 0 0 0 * * * * * * 6 0 0 0 * * 4 0 0 0 * * * * * 2 0 0 0 0

o 3 7 5 8 9 b b in 6 2 4 5 3 a C E c R 7 5 1 D 1 6 t 1 3 6 D 6 1 T P n 1 D D 1 6 c D D D 7 D 1 I a M C e F tr l E D C C D D l C C C C D C C C e C r n B t o -s n o c L c

Figure 1. Granulocyte immunophenotyping. Expression of various surface antigens on BCN was comparedto surface antigen expression on control- and GTX-derived neutrophils. Results are presented as MFI (mean ± SEM), significant differences are indicated by * when p<0.05, ** when p<0.01 and *** when p<0.001. N=5-10.

A . B . B C N B C N 1 0 0 C o n tro l 1 0 0 0 0 C o n tro l G T X

** G T X t * 8 0 n

e 8 0 0 0 t

n x o i e s 6 0 ,

e 6 0 0 0 s i h x d a A 4 0 * t 4 0 0 0 ** o % m

2 0 e h 2 0 0 0 **

** C **

0 - 0 F T S s F P a F A T y 5 S P A L N M - a F 8 C D L C P M C T P - / 3 f P 5 L A IL G P M C M P B A f L P C . B C N D . B C N 2 .0 C o n tro l 1 5 0 0 C o n tro ls s ' G T X G T X

N M

P 1 .5 6 0 . 1 0 0 0 1 n . i n i m

** /

m 1 .0 U /

2 ** F O R 2 5 0 0 H

l 0 .5 o m n

0 .0 0 - n Z P P - P P A A 0 A 0 a T L L L L M M s S M -1 P M M fM fM P P o f /f / / / / X m F F B F B T y A o A o z A t P t P P y y C C Figure 2. Granulocyte functions. A) Granulocyte adhesion is presented as % adhesion towards different stimuli.

B) Chemotaxis is presented as the maximal slope of cell migration towards different stimuli. C)2 H O2 production in nmol/106 granulocytes/min. D) Granulocyte protease release after degranulation is shown in relative fluorescence units (RFU)/min. N=8. Results are presented as mean ± SEM, significant differences are indicated by * when p<0.05, ** when p<0.01 and *** when p<0.001.

148

Buffy coat – derived granulocytes for clinical use

A . ) ) a a e 4 0 4 0 r e r a a f f o o % ( 3 0 % 3 0 ( a e a r e a r a y t y i 2 0 2 0 t s i s n e n e d d T 1 0 1 0 T E E N N O E P

0 N

M 0

s X N l l s X o T N T C r C o t G r G B B t n n B. o o C C

Figure 3. NET formation in response to PMA. A) Quantification of NETosis using fluorescently conjugated MPO and NE antibodies. N=3, with 4 fields/experiment analyzed. Results are in mean ± SEM. B) Fluorescence imaging of NETs represented by MPO (red), NE (green) and DNA (blue) and a merge, are shown in BCN (left), control (middle) and GTX (right).

Candida killing is superior to that of GTX granulocytes Granulocyte-mediated killing of Gram-positive S. aureus, Gram-negative, E. coli, and the fungi C. albicans and A. fumigatus was tested. BCN-derived granulocytes showed normal killing capacity of both bacteria and fungi (Figure 4A-D), including non-opsonized and serum- opsonized Candida conidia (Figure 4D). We recently found that the GTX granulocytes showed a significant defect of approximately 50% in the killing of Candida conidia11.

Survival of BCN-derived granulocytes as compared to GTX-derived granulocytes Upon release from the bone marrow, granulocytes have a half-life in the circulation of about 8-10 hours. Previous data showed considerably higher survival rates of GTX-derived granulo- cytes10. The viability of BCN-derived granulocytes is >95% after granulocyte isolation, which is comparable to the control- and GTX-derived granulocytes. BCN-derived granulocytes showed impaired overnight survival compared to control granulocytes after G-CSF incubation, whereas without G-CSF cell survival was comparable to control granulocytes (Figure 5A). Survival pattern analysis showed higher survival rates in buffy coat-derived granulocytes with G-CSF over control granulocytes without G-CSF incubation (Figure 5B).

149 Chapter 6

A . B .

1 0 0 B C N 1 0 0 B C N C o n tro l

y C o n tro l t

i l y i t G T X i 7 5 7 5 l b G T X i a i b a v i v s i u l 5 0 5 0 e o

r c u . a E . S % 2 5 2 5 % **

0 0 0 1 0 2 0 3 0 0 2 0 4 0 6 0 T im e (m in .) T im e (m in .)

C . D . 1 0 0 1 0 0 B C N y t i y l t i

C o n tro l i l b 8 0 i * * * a b i 7 5

G T X a v i v s * * * u

6 0 s t n a a

g 5 0

c n i i b

4 0 l m a u . f . C 2 5

A 2 0 %

0 0

0 .0 2 .5 5 .0 7 .5 1 0 .0 A O A O A O C C C C C C l l N N o X X r o 4 C C t tr T T [G ra n u lo c y te s , x 1 0 ] B B n n G G o o C C

Figure 4. Microbial killing. A) E. coli (N=4-27) and B) S. aureus (N=4-40) killing, shown by % of CFU’s at different timepoints. C) OD’s of viable Aspergillus fumigatus hyphae, after incubation with different amounts of granulocytes. N= 5-10. D) Unopsonized (CA) and serum-opsonized (CO) Candida albicans conidia killing, shown by % of CFU’s after 2 hour coculturing with granulocytes. N=6-19. Data are shown as mean ± SEM. Significant differences are indicated by * when p<0.05, ** when p<0.01 and *** when p<0.001.

A . B . 8 0 * 1 0 0 * W ith G -C S F C o n tro l w ith o u t W ith o u t G -C S F G -C S F 8 0 C o n tro l w ith 6 0 s

l G -C S F

l y t e i l

c B C N w ith o u t i 6 0 b e l

a G -C S F

i 4 0 b V a B C N w ith i 4 0 V % G -C S F % 2 0 2 0

0 0 l 0 5 1 0 1 5 2 0 N o X tr T C G B n o T im e (h r) C

Figure 5. Granulocyte viability A) With (black bars) or without (grey bars) overnight incubation with G-CSF. Data indicated as mean ± SEM, N=5-9. B) Apoptosis in time of BCN and controls with and without G-CSF. N=2.

150 Buffy coat – derived granulocytes for clinical use

Discussion

To the best of our knowledge, this is the first study that directly investigates the feasibil- ity and function of BCN concentrates as an alternative for GTX under the same conditions tested. We have extended the work of Bashir et al. who investigated the buffy-coat derived product used in the UK16, where GTX-derived granulocytes are less frequently used due to legal limitations in two ways. First, the product is different. Bashir et al. use a product made from pooled buffy coats from 10 whole blood donations, Platelet Additive Solution (PAS) and plasma, resulting in 1x1010 granulocytes in 200-250 ml21. This product is big in volume, high in platelets and relatively high in red blood cells. Our product is low in platelets, which is a benefit on itself when taking into account a recent publication on the safety of platelet transfusions22, low in red blood cells and has a volume of 56 ml. Second, the granulocytes are characterized in more detail, adding adhesion, degranulation, immunophenotyping, NET formation and microbial killing. Especially microbial killing is of importance in a product used in neutropenic patients. Our study demonstrates that BCN products may be a promising alternative for GTX in clinical settings. Although GTX granulocytes have been previously shown to localize at the site of a presumed infection23, studies on in-vivo survival and functionality 6 of granulocytes from BCN products have not yet been performed. The amount of transfused granulocytes is vital for the effect of the transfusion. BCN concentrates contain approximately 5.6 x 109 granulocytes, GTX concentrates approximately 1-5 x 1010 granulocytes10. Previous (uncontrolled) studies concluded that at least 1 x 1010 granulocytes are necessary to reduce infection-related mortality in adult patients with neutropenia or granulocyte dysfunction4,8,16,24 and 1.4-3 x 108 granulocytes/kg are recom- mended for children25. A more recent study (RING: A Randomized Controlled Trial of G-CSF- stimulated Granulocytes in Granulocytopenic Patients)26 suggested in a post-hoc analysis that patients receiving ≥ 0.6 x 109 granulocytes/kg/transfusion tended to have better clinical outcomes compared to patients who received less. In that case, BCN products would contain insufficient numbers of granulocytes for the use in adult patients. However, some consid- erations should be taken into account before applying these estimates to the use of BCN products. First, the RING study was inconclusive because patient enrollment was half of the number intended after power calculation. Also, the number of GTX granulocytes that may extravasate could be different from the BCN-derived granulocytes which have not been pre-exposed to G-CSF and corticosteroids. Next to that, the post-hoc analysis may be only applicable to patients selected for GTX and enrolled in the study. Hence, a direct comparison is not possible and the results of the RING study may be of limited value to our BCN product. A large part of the granulocyte loss is due to the leukocyte reduction filter (standard leukocyte reduction filter, Fresenius Kabi) used during the production of the thrombocyte concentrate, which contained 3.4 ± 0.7 x 109 granulocytes (N=6, mean ± SD). Whereas it is easy to flow these granulocytes back into the residual buffy coat, tests pointed out that

151 Chapter 6 these granulocytes possibly have an activated phenotype shown by elevated unstimulated adhesion and chemotaxis. A method to flow back only non-activated granulocytes from the filter is now being tested at our lab. For now we recommend transfusion of two bags of BCN over flowing back the “filter granulocytes”. Doing this does not exceed the amount of donors and thus possible HLA-antibody production compared to the Bashir product. This product showed similar HLA-alloimmunization rates compared to other studies27. Ideally, BCN concentrates are used as a bridging product in the absence of GTX. We showed that granulocyte basic functions are well preserved in-vitro in BCN-derived granulocytes. IL-8-induced chemotaxis of BCN-derived granulocytes was found to be (non- significantly) reduced, which may be a result of overnight storage and the production of thrombocyte concentrates prior to BCN concentrate generation. Nevertheless, the impact of this chemotactic ability cannot be directly translated to the in-vivo situation. fMLP-induced NADPH-oxidase activity seems to be elevated, which can be considered as a priming- ef fect28. However, background NADPH oxidase activity in resting BCN-derived granulocytes is as low as resting control granulocytes, suggesting a well-preserved quiescent state. Flow cytometry showed no signs of BCN-derived granulocyte activation, as the activation markers CD11a, CD11b, CD16, CD18, L-selectin, CD63, CD64 and CD177 were comparable to control granulocytes. BCN-derived granulocytes had a normal EMR3 expression, which is reduced in GTX-derived granulocytes suggesting that these cells have not reached full maturation yet upon release from the bone marrow. The same reasoning may explain the reduced L-selectin and CD16 expression and modest elevations of CD55, CD66b and CD64 on the surface of GTX-derived granulocytes. Killing mechanisms of BCN-derived granulocytes are comparable to control- and GTX- derived granulocytes. However, killing of Candida albicans conidia is significantly impaired in GTX-derived granulocytes11, while this pathogen is an important cause of neutropenic sepsis with mortality rates up to 40-60%29,30. This might be a result of the immaturity of GTX granulocytes and consequently differences in granular content31,32. More specific, we have shown that several antimicrobial granule proteins were reduced in GTX-derived granulocytes, especially Major Basic Protein (MBP). MBP has candidacidal activity, which was confirmed by impaired Candida killing of neutrophil granulocyte-like Crisp-Cas9 NB4-KO-MBP cells. Aspergillus hyphae, on the other hand, were normally killed by GTX- and BCN-derived granu- locytes11. Thus, it appears that BCN-derived granulocytes kill bacterial and fungal pathogens as efficiently as control granulocytes do, whereas the GTX-derived granulocytes showed a defect in the killing of Candida albicans tested under similar conditions. Survival pattern analysis showed acceptable survival of BCN-derived granulocytes after ex-vivo G-CSF incubation, which we hypothesize to mimic the situation after transfusion, as neutropenic patients are usually having increased G-CSF levels and/or receive daily G-CSF treatment. The combination of a lower granulocyte yield and a shorter lifespan is a possible disadvantage of BCN concentrates. Nevertheless, this can be resolved by transfusing more

152 Buffy coat – derived granulocytes for clinical use frequently, and/or more buffy pools. As residual buffy coats are readily available in the blood bank process, and the volume of the transfusion product is low, it is no problem to transfuse two or more of these products. In conclusion, buffy coat-derived granulocytes show good functional performance, can be made readily available as a transfusion product during the regular blood cell production process, and would take away any concern to use growth factors, corticosteroids and HES in healthy donors and the logistic issues around GTX production.

Acknowledgements The authors would like to thank prof. Dirk Roos for his cooperation during the study design and critical reading of the manuscript, dr. Johan Lagerberg for supplying the buffy coats, dr. Hans Vrielink, Fikreta Danovic and Aafke Smienk for the help with GTX donor material and dr. Benjamin Nota for his help with the statistical analysis.

153 Chapter 6

Reference List

1. Denning DW, Bromley MJ. Infectious Disease. How to bolster the antifungal pipeline. Science 2015 Mar 27;347(6229):1414-6. 2. Perlin DS, Shor E, Zhao Y. Update on Antifungal Drug Resistance. Curr.Clin.Microbiol.Rep. 2015 Jun 1;2(2):84-95. 3. World Health Organization. ANTIMICROBIAL RESISTANCE Global Report on Surveillance. 2014. 4. van de Wetering MD, Weggelaar N, Offringa M, Caron HN, Kuijpers TW. Granulocyte transfusions in neutropaenic children: a systematic review of the literature. Eur.J.Cancer 2007 Sep;43(14):2082-92. 5. Pammi M, Brocklehurst P. Granulocyte transfusions for neonates with confirmed or suspected sepsis and neutropenia. Cochrane.Database.Syst.Rev. 2011;(10):CD003956. 6. Sachs UJ, Reiter A, Walter T, Bein G, Woessmann W. Safety and efficacy of therapeutic early onset granulocyte transfusions in pediatric patients with neutropenia and severe infections. Transfusion 2006 Nov;46(11):1909-14. 7. Price TH, Bowden RA, Boeckh M, Bux J, Nelson K, Liles WC, Dale DC. Phase I/II trial of neutrophil transfusions from donors stimulated with G-CSF and dexamethasone for treatment of patients with infections in hematopoietic stem cell transplantation. Blood 2000 Jun 1;95(11):3302-9. 8. Estcourt LJ, Stanworth S, Doree C, Blanco P, Hopewell S, Trivella M, Massey E. Granulocyte transfusions for preventing infections in people with neutropenia or neutrophil dysfunction. Cochrane.Database. Syst.Rev. 2015;6:CD005341. 9. Kolovratova V, List J, Worel N, Jilma P, Leitner G. Administration of prednisolone intravenously or dexamethason orally in random donors reveals equal results in granulocyte collection: a single centre experience. Vox Sang. 2012 Jul;103(1):75-8. 10. Drewniak A, Boelens JJ, Vrielink H, Tool AT, Bruin MC, Heuvel-Eibrink M, Ball L, van de Wetering MD, Roos D, Kuijpers TW. Granulocyte concentrates: prolonged functional capacity during storage in the presence of phenotypic changes. Haematologica 2008 Jul;93(7):1058-67. 11. Gazendam RP, van de Geer A, van Hamme JL, Tool AT. Impaired killing of Candida albicans by granulocytes mobilized for transfusion purposes: a role for granule components. Haematologica 2016 Jan 22;Epub ahead of print. 12. US Food and Drug Administration adds black boxed warning to Hydroxyethyl starch (HES) American Society for Apheresis. 2013 Sep 12. 13. Therapeutic Apheresis Services; Donor Information Granulocyte Collection NHS Blood and Transplant. 2014 Jul 16. 14. Myburgh JA, Finfer S, Bellomo R, Billot L, Cass A, Gattas D, Glass P, Lipman J, Liu B, McArthur C, et al. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N.Engl.J Med. 2012 Nov 15;367(20):1901-11. 15. Perner A, Haase N, Guttormsen AB, Tenhunen J, Klemenzson G, Aneman A, Madsen KR, Moller MH, Elkjaer JM, Poulsen LM, et al. Hydroxyethyl starch 130/0.42 versus Ringer’s acetate in severe sepsis. N.Engl.J Med. 2012 Jul 12;367(2):124-34. 16. Bashir S, Stanworth S, Massey E, Goddard F, Cardigan R. Neutrophil function is preserved in a pooled granulocyte component prepared from whole blood donations. Br.J.Haematol. 2008 Mar;140(6):701- 11. 17. Bontekoe IJ, van der Meer PF, Mast G, de KD. Separation of centrifuged whole blood and pooled buffy coats using the new CompoMat G5: 3 years experience. Vox Sang. 2014 Aug 1;140-7.

154 Buffy coat – derived granulocytes for clinical use

18. Kuijpers TW, Alders M, Tool AT, Mellink C, Roos D, Hennekam RC. Hematologic abnormalities in Shwa- chman Diamond syndrome: lack of genotype-phenotype relationship. Blood 2005 Jul 1;106(1):356- 61. 19. McDonald B, Urrutia R, Yipp BG, Jenne CN, Kubes P. Intravascular neutrophil extracellular traps cap- ture bacteria from the bloodstream during sepsis. Cell Host.Microbe 2012 Sep 13;12(3):324-33. 20. Decleva E, Menegazzi R, Busetto S, Patriarca P, Dri P. Common methodology is inadequate for studies on the microbicidal activity of neutrophils. J.Leukoc.Biol. 2006 Jan;79(1):87-94. 21. NHS Blood and Transplant. Clinical guidelines for the use of granulocyte transfusions. Massey, E. 2012. 22. Hong H, Xiao W, Lazarus HM, Good CE, Maitta RW, Jacobs MR. Detection of septic transfusion reactions to platelet transfusions by active and passive surveillance. Blood 2016 Jan 28. 23. Adkins D, Goodgold H, Hendershott L, Johnston M, Cravens D, Spitzer G. Indium-labeled white blood cells apheresed from donors receiving G-CSF localize to sites of inflammation when infused into al- logeneic bone marrow transplant recipients. Bone Marrow Transplant 1997 Apr;19(8):809-12. 24. Stanworth SJ, Massey E, Hyde C, Brunskill S, Lucas G, Navarrete C, Marks DI. Granulocyte transfusions for treating infections in patients with neutropenia or neutrophil dysfunction. Cochrane.Database. Syst.Rev. 2005 Jul 1;(3). 25. Seidel MG, Minkov M, Witt V, Matthes-Martin S, Potschger U, Worel N, Leitner G, Stary J, Gadner H, Peters C. Granulocyte transfusions in children and young adults: does the dose matter? J.Pediatr. Hematol.Oncol. 2009 Mar;31(3):166-72. 26. Price TH, Boeckh M, Harrison RW, McCullough J, Ness PM, Strauss RG, Nichols WG, Hamza TH, Cush- 6 ing MM, King KE, et al. Efficacy of transfusion with granulocytes from G-CSF/dexamethasone-treated donors in neutropenic patients with infection. Blood 2015 Oct 29;126(18):2153-61. 27. Massey E, Harding K, Kahan BC, Llewelyn C, Wynn R, Moppett J, Robinson SP, Green A, Lucas G, Sadani D, et al. The granulocytes in neutropenia 1 (GIN 1) study: a safety study of granulocytes collected from whole blood and stored in additive solution and plasma. Transfus.Med. 2012 Aug;22(4):277-84. 28. DeLeo FR, Renee J, McCormick S, Nakamura M, Apicella M, Weiss JP, Nauseef WM. Neutrophils exposed to bacterial lipopolysaccharide upregulate NADPH oxidase assembly. J.Clin.Invest 1998 Jan 15;101(2):455-63. 29. DiNubile MJ, Hille D, Sable CA, Kartsonis NA. Invasive candidiasis in cancer patients: observations from a randomized clinical trial. J.Infect. 2005 Jun;50(5):443-9. 30. Barton CD, Waugh LK, Nielsen MJ, Paulus S. Febrile neutropenia in children treated for malignancy. J.Infect. 2015 Jun;71 Suppl 1:S27-S35. 31. Borregaard N. Neutrophils, from marrow to microbes. Immunity. 2010 Nov 24;33(5):657-70. 32. Clemmensen SN, Udby L, Borregaard N. Subcellular fractionation of human neutrophils and analysis of subcellular markers. Methods Mol.Biol. 2014;1124:53-76.

155