- Email: [email protected]
Microbicidal activity measured by flow cytometry: Optimization and standardization for detection of primary and functional deficiencies
Microbicidal activity measured by flow cytometry: optimization and standardization for detection of primary and functional deficiencies M Jeraiby, K Sidi Yahya, Depince-Berger, C Lambert PII: DOI: Reference:
S0022-1759(16)30220-4 doi:10.1016/j.jim.2016.09.010 JIM 12225
To appear in:
Journal of Immunological Methods
Received date: Revised date: Accepted date:
6 June 2016 11 August 2016 26 September 2016
Please cite this article as: Jeraiby, M., Sidi Yahya, K., Depince-Berger, Lambert, C., Microbicidal activity measured by flow cytometry: optimization and standardization for detection of primary and functional deficiencies, Journal of Immunological Methods (2016), doi:10.1016/j.jim.2016.09.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
IP
T
Microbicidal activity measured by flow cytometry: optimization and standardization for detection of primary and functional deficiencies
SC R
JERAIBY M1, SIDI YAHYA K, DEPINCE-BERGER, LAMBERT C1
MA
NU
Immunology Laboratory, CNRS UMR5307 Labo Georges Friedel (LGF), Pole de Biologie-Pathologie, University Hospital. St Etienne, France
D
Corresponding author: Dr. Claude Lambert, Immunology Laboratory, Pole de Biologie-Pathologie, UnivHospital, CHU St Etienne, F 42055 ST ETIENNE Cedex 2, France
TE
E-mail : [email protected].
CE P
Phone number: +33 4 77 12 05 13
AC
Fax number: + 33 4 77 12 05 52
Abbreviations:
HLA-DR: human leucocyte antigen D-related; CRP: C-reactive protein; PMN: Polymorphonuclear neutrophil; ROS: reactive oxygen species; E coli: Escherichia coli; DHR123: Dihydrorhodamine 123; R123: Rhodamine 123; EDTA: Ethylene Diamine Tetra Acetic Acid; BSA: Bovine serum albumin; PMA: Phorbol 12 Myristate 13Acetate; CV: Coefficient of variation; NADPH: Nicotin amide adenine dinucleotide phosphate; PBS: Phosphate-Buffered Saline; MdFI: Median florescence intensity; dim: diminished; FSC: Forward Scatter; SSC: Side Scatter; FITC: Fluorescein isothiocyanate; PE-Cy7 Phycoerythrine Cyanin 7; APC: Allophycocyanin; NS : non significant; LPS: lipopolysaccharide ; IL :
ACCEPTED MANUSCRIPT interleukine ; ICU : Intensive care unit;
IP
T
Abstract
Microbicidal activity is related to the production of reactive oxygen species (ROS)
SC R
that can be measured by Flow-Cytometry using Rhodamine 123 (R123). Few assays have been proposed to measure ROS production, usually on heparinized
NU
samples but none of them is standardized. Here we propose to improve the test by selecting polymorphonuclears (PMN) and monocytes, labelled and activated in
MA
one step to keep the test short, and to standardize the process even between different systems (i.e. Navios and FACSCanto) using Fluorescence Intensity
TE
D
Target Setting (“FITS”). We applied this test on 15 patients without inflammation, 19 patients from an intensive care unit (ICU) and 11 healthy volunteers. Results:
CE P
provided calcium restitution, we show that the test can be performed on EDTA that is a better sample preservative. The results were highly correlated between
AC
instruments (r² = 0,898). PMN CD16 (and not CD14) expression was altered under stimulation with E. coli (MdFI = 239.3+93.5) or PMA (139.7+76.8) as compared
to
resting
sample
(307.6+145.1).
RH123
was
strongly
and
homogeneously induced by PMA (14.2+6.6) and more heterogeneously by E coli (MdFI 21.9+23.4) as compared to unstimulated PNN (0.9+1.3, p<0.0001). The test useful for genetic disorders but is also for secondary deficiencies as observed in ICU (E coli RH123 MFI = 10.5+11.1 patients vs 30.1+26.5 in healthy donors). In ICU, CD16 expression was already altered on unstimulated samples (MdFI = 197.4+131.2 vs 418, 2+81.3 in healthy donors; p=<0.0001). Bacterial stimulation
ACCEPTED MANUSCRIPT was dependent of the complement that partly explains deficiency to bacterial
IP
T
stimulus in ICU patients.
Keywords: Microbicidal activity, oxidative burst, immunodeficiency,
NU
SC R
Rhodamine, flow cytometry, standardization, granulocytes.
Phagocytosis
of
MA
1. Introduction micro-organisms
or
cell
debris
is
the
main
task
of
D
polymorphonuclear leukocytes (PMNs). It is facilitated by plasma products such as complement components, antibodies and scavengers (1). This mechanism is
TE
named opsonisation. Several steps in opsonisation, including complement activation and adhesion, require cations such as Calcium and Magnesium.
CE P
Phagocytes contain a large arsenal of enzymes for proteolysis and production of reactive oxidative species (ROS) (2). Microorganism capture triggers NADPH oxidase that produces NADP+, superoxide ion (O2ˉ) and hydrogen peroxide
AC
(H2O2). Later myeloperoxidase transforms these species into hydroxyl radicals (°OH) and hypochlorite (ClO-). These newly produced ROS are highly effective in killing and digesting ingested particles including bacteria and fungi (microbicidal activity) (3,4). Inherited dysfunctions of NADPH oxidase or more rarely myeloperoxidase are responsible for defect in ROS production and accumulation of phagocytized products leading to Chronic Granulomatous Disease (CGD)(5). Microbicidal activity can be measured through the capacity of the cells to produce ROS for the diagnosis of CGD.
Tissue infection induces systemic effects with fever and inflammatory metabolism that includes production of acute phase proteins including opsonins. Systemic effects are induced by bacterial components such as lipopolysaccharide (LPS) or
ACCEPTED MANUSCRIPT peptidoglycan (6) and are mediated by the release of cytokines such as IFN, TNF or IL-6 (7,8) preparing peripheral cells for phagocytosis and migration to the inflammatory site. At later stage, cell exhaustion and regulatory mechanisms
T
induce a transient depression of phagocytosis called immunoparalysis (9).
IP
Immunoparalysis can be prolonged in sepsis, extended burning, injury or surgery
SC R
and contributes in the severity of secondary infections leading and consequently a high mortality rate in Intensive Care Units (ICU) (10-12). Immunoparalysis can be analyzed in peripheral blood cells of ICU patients. As an example, the high affinity IgG receptor FcRI (CD64) (13) is up regulated while the low affinity FcRIII
NU
receptor (CD16) on PMNs (14) and HLA DR molecule expression on the surface of monocytes are down-regulated (15,16). These markers have been shown to
MA
predict a poor clinical outcome of patients (17,18). Microbicidal activity can be measured by flow cytometry. Several procedures
D
have been proposed for diagnosis of CGD and can be applied for Immunoparalysis. They are usually based on measurement of production of ROS.
TE
ROS induces oxidization of probes such as Dihydrorhodamine 123 (DHR123) into
CE P
fluorescent Rhodamine 123 (R123) that can be measured in individual cells by flow cytometry (2,19).
These methods can be easily implemented to provide
more precise and reproducible output, e.g. by adding selective markers for granulocytes. Furthermore, in routine cell analysis, the use of Ethylene-diamine-
AC
tetra-acetic acid (EDTA) is recommended as the best preservative anticoagulant (20,21). But by chelating calcium, EDTA not only prevents coagulation but also cell aggregation, complement activation and phagocytosis (22). However, standardization of the test is required in accreditation process according to ISO 15189 (23) and attempts have been recently proposed, even between different systems (instruments, antibodies.) (24-26). In the present report, we propose to implement and standardize a microbicidal test, independently of the reagent and instrument used, for the diagnosis of CGD and we show its interest in monitoring immunoparalysis.
ACCEPTED MANUSCRIPT 2. Material and Method
T
2.1. Blood sample.
IP
The study was performed at University Hospital of Saint-Etienne (France) on blood samples collected from 15 adults patients from cardiac unit without
and eventually ongoing sepsis (group C)
SC R
biological sign of acute inflammation (group B), 19 patients from ICU with recent and 11 healthy volunteers from
occupational medicine (group A). Venous blood was collected on EDTA or lithium
NU
heparin tubes and tested within 12 hours of sampling. Patients from ICU had inflammation status as shown by raised level of CRP (from 14 to 366 mg/L; ref
MA
value <5 mg/L) and ongoing bacterial infection was identified in 8 patients (gram negative bacteria, n = 5, gram positive bacteria, n = 4). For legal reasons, this technical study was performed on anonymous samples. No more medical
D
information was available and measurement of CRP was not clinically justified in
TE
control donors and patients. All donors were older than 35 years and
2.2. Reagents
CE P
approximately equally distributed between males or females.
All reagents for phagocytosis induced oxidative burst assays were taken from
AC
FagoFlowEx® Kit (Cat. No. ED7042, EXBIO, Praha, Czech Republic) containing standardized lyophilized Escherichia coli bacteria (E coli), lyophilized DHR123, lyophilized Phorbol 12-myristate 13-acetate (PMA) as a stimulation control and ready-to-use
lysing
solution.
Labeled
monoclonal
antibodies
CD16-
Phycoerythrin–Cyanine (PE-Cy7; clone 3G8) and CD14-allophycocyanin (APC; clone MEM-15) were also purchased from EXBIO. EDTA samples were supplemented with Ca2+ at 10 mmol/L and calcium Heparinate at 250UI/ml final concentrations. Ca2+ replacement buffer was optimized on the criteria of sample coagulation and activation process. Different doses of calcium and heparin were tested around concentrations usually used. Suboptimal doses induced coagulation or lack of activation. Optimal doses of calcium and heparin were finally validated by comparing 10 individuals
ACCEPTED MANUSCRIPT simultaneously analyzed on heparin and EDTA anti-coagulated blood samples, EDTA being supplemented with Ca and heparin. The assay was also simplified. Cell stimulation and antibody labelling were
T
performed in one step for 20 minutes in 37°C water bath followed by fixation and
IP
red blood cells lysis done according to manufacturer instructions. In order to better preserve the sample, PBS-bovine serum albumin 10 g/L (MP Biomedical
SC R
Solon OH), EDTA (0.5 G/L; MP Biomedical) was added after erythrolysis for data acquisition.
Samples were analyzed on Navios™ flow cytometer (CE-IVD; 3 lasers, 10 colors)
NU
using Navios™ software (Beckman-Coulter; Fullerton, CA) within 1 hour from their final wash. The flow cytometer was daily calibrated on a routine basis using microbeads
(Flowset,
MA
fluorescent
Beckman-Coulter).
R123
median
fluorescence intensity (MdFI) was measured on at least 5.000 PMNs selected on CD16 labelling versus side-scatter (SSC) and on monocytes selected on
D
CD14/SSC. Data analyses were performed using Kaluza™ software (Beckman
TE
Coulter).
CE P
2.3. Standardization between instruments Protocol standardization on time and between laboratories can be managed by keeping fluorescence detections at the same levels on time as daily checked but
AC
also between different instruments using beads with standardized fluorescence intensity. As previously described (26), settings were first optimized on Navios; fluorescence targets were measured using calibration beads (Cyto-Cal Multifluor Calibrator, Thermo Scientific, Carlsbad CA) and voltages adjusted on BD FACSCanto™ II cytometer (BD Biosciences, San Jose, CA) to get the same fluorescence targets. The method was named as Fluorescence Intensity Target Settings procedure (FITS). On these settings, 20 samples, same preparation, were cross-analyzed on the two instruments in delays that did not exceed 3 hours.
2.4. Statistical Analyses
ACCEPTED MANUSCRIPT Statistical analysis was performed using, paired student’s T test, linear regressions and are expressed as median +/-1SD. Coefficient of variation (CV)
T
were calculated as 1SD/Mean.
IP
2.5. Legal issues
This technical study was performed on anonymous samples provided for
MA
NU
SC R
diagnosis purpose in accordance with French Law.
3. Results
D
3.1 .Method optimization: In order to implement the test precision and robustness, we used CD16 and CD14 to select neutrophil PMNs and monocytes respectively
TE
(fig 1a; 2a). The labelling was performed in one step during the activation step. The one step gave same results on preliminary tests as if samples were activated
CE P
and labelled in two steps. The induced microbicidal activity could be easily measured under stimulation with PMA (fig 1d, 2d) or E Coli (fig 1c, 2c) and compared to unstimulated sample (fig 1b, 2b) on the two leukocyte populations
AC
involved in phagocytosis: PMNs (fig 1 and 2b) or monocytes (fig 2c). Calcium supplementation on EDTA samples gave similar results as heparinized samples as shown on one representative sample (fig 2d) with a better sample stability at least up to 24hours storing at 4°C before testing on 5 healthy donors (Fig 2e).
3.2. PMN and monocyte activation test: overall CD16 expression was significantly decreased on granulocytes stimulated with E coli (MdFI = 239.3+93.5, paired t test: p<0.001) and even more when they were stimulated with PMA (139.7+76.8; p<0.0001) as compared to unstimulated cells (307.6+145.1) in 45 tests from the three clinical groups altogether (19 sepsis, 15 non infected patients, 11 healthy donors; fig 3a). The stimulation did not significantly change CD14 expression on monocytes (not shown).
ACCEPTED MANUSCRIPT
E coli induced very strong production of fluorescent R123 by PMNs (MdFI 21.9+23.4, n=45 tests) as compared to unstimulated PMNs (0.9+1.3, p<0.0001).
T
This Rhodamine production induced by E. coli was much heterogeneous (fig 1c)
IP
and significantly higher than under PMA activation (14.2+6.6; p<0.017; fig 1d). In some patients, R123 fluorescence was more or less distributed into 3 peaks of
SC R
progressive intensities: 1.6+0.5; 9.4+2.6; 50.6+9.1 respectively, suggesting 3 different mechanisms of cell activation. The low intensity peak was significantly higher than autofluorescence measured on stimulated samples without DHR123
NU
(MdFI 0.200+0.041 under PMA stimulus of 4 patients) as compared to unstimulated samples (MdFI 0.113+0.009). The highest peak was significantly
MA
higher than values observed with PMA stimulus.
In comparison, monocytes produced low and homogeneous R123 fluorescence
D
under E coli stimulation (2.0+3.2) although it was still significantly higher than
TE
unstimulated monocytes (0.9+1.6, p<0.0038) but lower than under PMA activation
CE P
(3.0+2.0, p<0.0001; table III).
3.3. Method validation: By repeating the E coli stimulation 10 times on two samples on the same day, by the same operator, we could observe a good
AC
repeatability for PMN CD16 labelling (CV =2.8 and 4.0% fig 2a). R123 fluorescence repeatability was not so good on PMN (CV =19.2 and 54.3%; fig 2b) but was better on monocytes (CV =13.4 and 31.6%; fig 2c). Because samples cannot be stored for a long time, reproducibility could not be measured but 20 EDTA samples stored at 4°C were tested for four consecutive days. Surprisingly we observed good E coli induced production of R123 fluorescence by surviving PMN at least on the first 24-48 hours of preservation (mean change at 48 hours: +19.7%, fig 2e), despite the rapid drop in number of surviving PMN. PMN death was difficult to quantify.
3.4. Protocol standardization between two types of instruments: In order to detect the same levels of fluorescence between different instruments, we measured
ACCEPTED MANUSCRIPT median fluorescence intensity (MFI) on Navios from one representative PMA activated sample for Rhodamine (15.3), CD16-PE-Cy7 (120) and CD14 (4.4) and used these values as targets for setting FACSCanto II with a correction factor
T
(x256) considering the differences in scales, giving targets at: Rhodamine (3 917),
IP
CD16-PE-Cy7 (30 720) and CD14 (1 126; Table II). Samples analyzed shown in fig 1 a’b’c’d’ FACSCanto II
SC R
simultaneously on the two systems show very comparable plots (one typical case as compared to Navios fig 1abcd).
Then, 20 (non-infected and healthy) samples were recorded on both instruments showing very close pictures with excellent correlation of fluorescence intensities
NU
on CD16 (fig 2f) and R123 induced by E coli (r² = 0,898) or PMA (r² = 0, 980; fig
MA
2g).
3.5. Clinical interest of the test: Because of the rarity of genetic deficiency (CGD) that would give clear defect in response, we focused on functional deficiency. In
D
order to check if the test could detect functional deficiency in severely ill patients,
TE
we compared PMN capacity to produce ROS in response to E coli between the 3 groups of patients on Navios. First (table I), reduced expression of CD16 could
CE P
be already observed in unstimulated samples of ICU patients (197.4+131.2) as compared to patients from group B (366.0+97.9) and to healthy donors (418.2+81.3 p<0.0001; fig 3a). Similar CD16 down regulation was observed after
AC
ex vivo stimulation in all groups (table I). Furthermore, ICU patient’s PMN had much lower microbicidal capacity as compared to healthy donors when induced by E coli (10.5+11.1 versus 30.1+26.5; p<0.037; fig 3a; table IV) or PMA (10.2+5.9 versus 17.9+3.4; p<0.0002; table IV). Microbicidal activity in Group B patients was not significantly different from healthy donors. Interestingly, E coli response could be stronger than PMA’s response in non-septic patients or healthy donors but this was not observed in ICU patients.
Considering Monocyte activity in healthy donors, E coli could induce some activity (MdFI: 0.67+0.31 compared to unstimulated monocytes 0.18+0.04, p=0.0004; table III) but at lower levels compared to PMA activation (2.31+0.6; p<0.0001; fig 3c).
Interestingly, in stressed, non septic patients (group B) unstimulated
ACCEPTED MANUSCRIPT monocytes had significantly higher activity (2.2+2.3; p=0.0035) as compared to healthy donors. This activity was increased under E coli (4.0+4.7) and PMA (3.9+2.5) stimulations to the same level, not more, than healthy donors. That
T
basal hyperactivity in stressed patients was found not so high in ICU patients
IP
(0.30+0.2; p<0.001) although it was still higher than in healthy donors. ICU Monocytes could further be activated by E coli, and at higher level than healthy
SC R
donors (1.10+0.69; p<0.0317) although PMA stimulation (2.75+1.86) was similar to healthy controls (Table III).
NU
3.6. Approach on the mechanism of E coli induced activation of granulocytes in ICU patients: As E coli induced microbicidal activity was defective (MdFI
MA
=10.6+11.4; fig 4; n=16) when tested on whole blood, we analyzed the possible role of plasma. E coli induced activity was at comparable level as under PMA stimulation (MdFI =10.4+5.6). In healthy donors, washing the blood cells
D
completely inhibit E coli activation. This was restored with plasma restitution from
TE
healthy donor unless the plasma was heat inactivated. In sepsis patients, plasma washing out strongly reduced but did not abolished the E coli activation (MdFI
CE P
2.3+2.3; p=0.0082) as compared to unstimulated PMN (0.26+0.12, p=0.002). Plasma substitution from healthy donors restored the E coli activation (13.5+10.3) at similar level as unwashed samples, but restoration was only partial if donor
AC
plasma was heat inactivated (6.9+3.0, p=0.0056 compared to untreated healthy plasma). Surprisingly, in ICU, E coli could significantly activate cells despite complement inactivation as compared to cells without plasma (p<0.0006) even if the activity was not completely restored as on unwashed cells (p<0.0001; see table V).
4. Discussion Testing microbicidal activity appeared to be a simple, reliable test despite it is a functional test requiring more sample preparation. We could show a clear induction of ROS production using PMA giving a narrow, homogeneous distribution of R123 fluorescence. This provides a good internal and interexperiment quality control. Activation with bacteria are more physiological and
ACCEPTED MANUSCRIPT probably more relevant clinically. Because of high variability in bacteria preparation, it is much safer using commercially available bacteria preparation that guarantees the reproducibility of the stimulus between experiments. The
T
study of phagocytic function and bactericidal activity has important clinical
IP
implications for the detection of genetic deficiency in which clear-cut results are expected (2,27). But this disease is rare and more applications can be clinically
SC R
relevant too. As an example, our data show a partial deficiency in a large proportion of severely ill patients in ICU, in accordance with immuno-paralysis that has been repeatedly reported (2,10,17,27,28). Measuring immunodeficiency
NU
can be of great help in these patients at high risk of secondary infection. The test is not new but we thought it can be easily implemented. By just adding 2
MA
antibodies, we could show that the test gets more precise information, at limited added cost. The sample preparation is not prolonged as labeling can simply be performed during stimulation. On this respect, 15-20 min incubation as
D
recommended by the manufacturer is a good compromise to have short duration
TE
test. A longer incubation might increase the response but this one is already very strong. This 3 color test is easy to set up and we chose fluorochromes with no
CE P
spectral overlap and that can be applied on most instruments. However, other conjugates can be easily used with R123 for instruments with only one laser or no far red detection.
AC
By this technique, two leukocyte populations (neutrophils and monocytes) from the same preparation could be tested for phagocytosis and oxidative burst without the need for further purification and treatment of the cells. Our results show that the two lineages do not have the same behavior. Note that NK cells (CD16+ low SSC) could be used as negative controls for DHR123 activation (29). Going to routine diagnosis in basic laboratories requires robustness and reproducibility for continuous monitoring of patients on time and for comparison between patients and across laboratories. Our proposal for standardization of the test appears to be easy to apply and most valuable upgrading at little or no extra cost, even different instruments based on Fluorescence Intensity Target Setting (“FITS”) as recently described (26). Beads or PMA activation of healthy donors can be used for inter-instrument settings while to E coli activation is too
ACCEPTED MANUSCRIPT heterogeneous to be considered as source for the targets. Our instrument settings are free to use by any lab interested and targets are detailed in table II. Because of clear calcium dependency of bacterial capture, heparin anticoagulant
T
is usually recommended by the manufacturers for this test. However, EDTA
IP
anticoagulant is usually preferred in routine diagnosis (including hemogramm) for its better preservative capacity than heparin and this can be a critical issue
SC R
especially for monitoring granulocytes functionality. Our experience is in favor of EDTA preservative. So because of sample availability and a better preservation, we
would
recommend
using
EDTA
and
we
show
that
the
calcium
NU
supplementation is simple and very convenient. Based on our comparison, the results on EDTA samples should be considered similar to Heparin samples. In
MA
any case, we recommend keeping the same procedure (either heparin or EDTA) for the same study.
Because of their fragility, it is usually recommended to perform any PMN test
D
within 8 hours. We were surprised to see repeatedly that PMN microbicidal
TE
activity was still correct after one night of storage at 4°C. This is not enough to recommend this delay of storage but sometimes, logistic issues are difficult to
CE P
achieve in short delays. The activity still detectable after one night may be biased because most fragile, poorly functional granulocytes are probably lost preferentially in the first few hours of storage. In this respect, using CD16 may
AC
help better select the granulocytes as PMN side scatter rapidly decreases after few hours of storage. CD16 brings information on cell stress. Indeed, we observed that CD16 expression strongly decreases on stimulated cells as it was previously reported in pre-apoptosis (30). The meaning is probably different from Rhodamine readout as CD16 decrease was significantly higher under PMA stimulation compared to E coli while R123 activation was much higher and heterogeneous with E coli compared to PMA activation. If using CD16/CD14 labelling does not increase the time of the experiment, it does increase the reagent cost that can be justified by several advantages. Neutrophils PMN are more precisely selected as CD16 expression is better preserved on time and help separating singlet from micro-aggregates. CD16 levels of expression give good information on cell stress, even before stimulation.
ACCEPTED MANUSCRIPT CD16 changes were observed in ICU patients but also in patients with not such strong inflammatory status. With the use of CD14 and CD16, it is also possible to analyze activity of monocyte subpopulations M1 (CD14+CD16-) and M2
T
(CD14dim or negative and CD16+) even if this distinction is more difficult to do on
IP
activated cells (31). DHR123 activation by PMN was very efficient, even in a short period of incubation; with one or two decades of increase, which makes negligible
SC R
the activation induced autofluorescence changes. However, quantifying R123 fluorescence is not always straight since the distribution is very wide and heterogeneous (see fig 3b). This may partly explain the wide variability of E coli
NU
activation between patients and even on repeats in single patient. Heterogeneous distribution of the response may reflect cell population heterogeneity due to cell
MA
rapid ageing or differences in maturation / pre-activation stages. Technical issues may also explain this heterogeneity through uneven distribution of stimulus per cell because of possible E coli micro-aggregation. Sample preparation has to be
D
very strict with good homogenization of E coli before and after adding the cells in
TE
order to have as much as possible an even distribution of bacteria on PMN. In any case, positivity is usually close to 100% and it is better considering the
CE P
fluorescence intensity shift. Median is of course more representative than mean fluorescence intensity but still do not reflect the discrete distribution among positive cells.
AC
Alternatively, peripheral monocytes were poorly activated by E coli, close to levels of PMA induced activity although pathogen associated products are known to directly activate macrophages. This might be due to a lack of maturation and possibly a different mechanism as compared to granulocytes. In any case, measuring bactericidal activity in monocytes represents a good internal reference for comparison of PMN. Despite that our technical study was not designed to approach clinical significance of the test; our preliminary data have shown strong reduction of capacity of granulocyte to activate DHR123 under E coli stimulation in ICU patients. This corroborated with reduced CD16 expression even observed on unstimulated cells, suggesting a pre-activation status of cells with possible refractoriness or exhaustion as previous well documented, with poor prognosis.
ACCEPTED MANUSCRIPT
In conclusion: measuring bactericidal activity by flow cytometry is easy and quick to perform and can be easily standardized for routine detection of functional
T
deficit specially in primary genetic disorders or in acquired immuno-paralysis as
AC
CE P
TE
D
MA
NU
SC R
IP
frequently observed in ICU.
ACCEPTED MANUSCRIPT Legends
SC R
IP
T
Figures Figure 1. Flow cytometry PMN selection on CD16 vs side scatter of a lysed whole blood specimen on 2 different instruments Navios (a,b,c and d) and FACSCAntoII (a’,b’,c’ and d’). Typical PMN produced rhodamine fluorescence histograms of resting (b, b’) or under E coli (c, c’) or PMA (d, d’) stimulation in one representative donor. E coli stimulus show a strong, heterogeneous response compared to PMA Stimulation.
MA
NU
Figure 2. Optimal E coli induced phagocytosis test was repeated 10 times at the same day on PMN (a, b) and monocytes (c) compared to one non stimulation and one PMA stimulation on one representative sample. Sample preservation on EDTA or Heparin was compared in few samples (d) and EDTA preservation tested on 4 days in 20 samples (e). CD16 labelling and E coli induced Rhodamine fluorescence from PMN were compared between two systems after Fluorescence Intensity target settings using a PMA induced sample, on Navios (Beckman Coulter) and a Canto II (BD Biosciences).
CE P
TE
D
Figure 3. Expression of CD16 (a) and Rhodamine fluorescence (b) on PMN or monocytes (c) were compared between healthy, non-infected and infected patients.
AC
Figure 4. The Role of plasmatic components in E coli induced production of Rhodamine RH123 fluorescence by granulocytes. Whole blood was analyzed alone (unstimulated, column 1) or stimulated with E Coli (column 2) or PMA (column 6). E coli stimulation was also tested on washed blood (column3-5), restituted with Hanks buffer directly (column 3) or with an healthy donor plasma, directly (column 4) or after heat inactivation (column 5) on 18 ICU patients.
ACCEPTED MANUSCRIPT
T
Tables Table I: CD16 levels of expression (median +1SD) on PMN in patients with or without sepsis compared to healthy donors.
SC R
IP
Table II : Standardization between different systems using of Fluorescence Intensity Target Settings considering respective scales.
NU
Table III : Rhodamine fluorescence intensity (mean +1SD) produced by Monocytes under stimulation, in patients with (n=17) or without (n = 16) sepsis compared to healthy donors (n=10).
MA
Table IV: Rhodamine fluorescence intensity (median +1SD) produced by PMN under activation, in patients with or without sepsis compared to healthy donors.
AC
CE P
TE
D
Table V: E coli induced Rhodamine fluorescence by PMN (median +1SD). Role of plasma components in 16 ICU patients comparing whole blood with washed blood and plasma either neat or heat inactivation replacement from one healthy donor.
ACCEPTED MANUSCRIPT
References McDonald JU, Rosas M, Brown GD, Jones SA, Taylor PR. Differential dependencies of monocytes and neutrophils on dectin-1, dectin-2 and complement for the recognition of fungal particles in inflammation. PLoS One 2012;7:e45781. Petropoulos M, Karamolegkou G, Rosmaraki E, Tsakas S. Hydrogen peroxide signals E. coli phagocytosis by human polymorphonuclear cells; up-stream and down-stream pathway. Redox Biol 2015;6:100-105. Marchi LF, Sesti-Costa R, Ignacchiti MD, Chedraoui-Silva S, Mantovani B. In vitro activation of mouse neutrophils by recombinant human interferongamma: increased phagocytosis and release of reactive oxygen species and pro-inflammatory cytokines. Int Immunopharmacol 2014;18:228-35. Letiembre M, Echchannaoui H, Bachmann P, Ferracin F, Nieto C, Espinosa M, Landmann R. Toll-like receptor 2 deficiency delays pneumococcal phagocytosis and impairs oxidative killing by granulocytes. Infect Immun 2005;73:8397-401. Emmendorffer A, Nakamura M, Rothe G, Spiekermann K, Lohmann-Matthes ML, Roesler J. Evaluation of flow cytometric methods for diagnosis of chronic granulomatous disease variants under routine laboratory conditions. Cytometry 1994;18:147-55. De Marzi MC, Todone M, Ganem MB, Wang Q, Mariuzza RA, Fernandez MM, Malchiodi EL. Peptidoglycan recognition protein-peptidoglycan complexes increase monocyte/macrophage activation and enhance the inflammatory response. Immunology 2015;145:429-42. Fernando MR, Reyes JL, Iannuzzi J, Leung G, McKay DM. The pro-inflammatory cytokine, interleukin-6, enhances the polarization of alternatively activated macrophages. PLoS One 2014;9:e94188. Striz I, Brabcova E, Kolesar L, Sekerkova A. Cytokine networking of innate immunity cells: a potential target of therapy. Clin Sci (Lond) 2014;126:593612. Volk HD, Reinke P, Docke WD. Clinical aspects: from systemic inflammation to 'immunoparalysis'. Chem Immunol 2000;74:162-77. Fitrolaki DM, Dimitriou H, Kalmanti M, Briassoulis G. CD64-Neutrophil expression and stress metabolic patterns in early sepsis and severe traumatic brain injury in children. BMC Pediatr 2013;13:31. Trimmel H, Luschin U, Kohrer K, Anzur C, Vevera D, Spittler A. Clinical outcome of critically ill patients cannot be defined by cutoff values of monocyte human leukocyte antigen-DR expression. Shock 2012;37:140-4. Janols H, Wullt M, Bergenfelz C, Bjornsson S, Lickei H, Janciauskiene S, Leandersson K, Bredberg A. Heterogeneity among septic shock patients in a set of immunoregulatory markers. Eur J Clin Microbiol Infect Dis 2014;33:313-24. Li S, Huang X, Chen Z, Zhong H, Peng Q, Deng Y, Qin X, Zhao J. Neutrophil CD64 expression as a biomarker in the early diagnosis of bacterial infection: a metaanalysis. Int J Infect Dis 2013;17:e12-23.
4.
8. 9. 10. 11. 12.
13.
SC R
CE P
7.
AC
6.
TE
D
5.
NU
3.
MA
2.
IP
T
1.
ACCEPTED MANUSCRIPT 14.
21. 22. 23. 24.
25.
26.
27. 28.
SC R
NU
MA
20.
D
19.
TE
18.
CE P
17.
AC
16.
IP
T
15.
Poehlmann H, Schefold JC, Zuckermann-Becker H, Volk HD, Meisel C. Phenotype changes and impaired function of dendritic cell subsets in patients with sepsis: a prospective observational analysis. Crit Care 2009;13:R119. Gouel-Cheron A, Allaouchiche B, Guignant C, Davin F, Floccard B, Monneret G, AzuRea G. Early interleukin-6 and slope of monocyte human leukocyte antigen-DR: a powerful association to predict the development of sepsis after major trauma. PLoS One 2012;7:e33095. Lambert C, Preijers FW, Demirel GY, Sack U. Monocytes and macrophages in flow: An ESCCA initiative on advanced analyses of monocyte lineage using flow cytometry. Cytometry B Clin Cytom 2015. Monneret G, Venet F. Sepsis-induced immune alterations monitoring by flow cytometry as a promising tool for individualized therapy. Cytometry B Clin Cytom 2015. Spittler A, Roth E. Is monocyte HLA-DR expression predictive for clinical outcome in sepsis? Intensive Care Med 2003;29:1211-2. Chen Y, Junger WG. Measurement of oxidative burst in neutrophils. Methods Mol Biol 2012;844:115-24. Davis BH, Dasgupta A, Kussick S, Han JY, Estrellado A, Group IIW. Validation of cell-based fluorescence assays: practice guidelines from the ICSH and ICCS part II - preanalytical issues. Cytometry B Clin Cytom 2013;84:286-90. Banfi G, Salvagno GL, Lippi G. The role of ethylenediamine tetraacetic acid (EDTA) as in vitro anticoagulant for diagnostic purposes. Clin Chem Lab Med 2007;45:565-76. Elsner J, Oppermann M, Czech W, Kapp A. C3a activates the respiratory burst in human polymorphonuclear neutrophilic leukocytes via pertussis toxinsensitive G-proteins. Blood 1994;83:3324-31. Sack U, Barnett D, Demirel GY, Fossat C, Fricke S, Kafassi N, Nebe T, Psarra K, Steinmann J, Lambert C. Accreditation of flow cytometry in Europe. Cytometry B Clin Cytom 2013;84:135-42. Kalina T, Flores-Montero J, van der Velden VH, Martin-Ayuso M, Bottcher S, Ritgen M, Almeida J, Lhermitte L, Asnafi V, Mendonca A and others. EuroFlow standardization of flow cytometer instrument settings and immunophenotyping protocols. Leukemia 2012;26:1986-2010. van de Loosdrecht AA, Alhan C, Bene MC, Della Porta MG, Drager AM, Feuillard J, Font P, Germing U, Haase D, Homburg CH and others. Standardization of flow cytometry in myelodysplastic syndromes: report from the first European LeukemiaNet working conference on flow cytometry in myelodysplastic syndromes. Haematologica 2009;94:1124-34. Solly F, Rigollet L, Baseggio L, Guy J, Borgeot J, Guerin E, Debliquis A, Drenou B, Campos L, Lacombe F and others. Comparable flow cytometry data can be obtained with two types of instruments, Canto II, and Navios. A GEIL study. Cytometry A 2013;83:1066-72. Nordenfelt P. Quantitative assessment of neutrophil phagocytosis using flow cytometry. Methods Mol Biol 2014;1124:279-89. Danikas DD, Karakantza M, Theodorou GL, Sakellaropoulos GC, Gogos CA. Prognostic value of phagocytic activity of neutrophils and monocytes in
ACCEPTED MANUSCRIPT
T
IP
SC R
NU MA D TE
31.
CE P
30.
AC
29.
sepsis. Correlation to CD64 and CD14 antigen expression. Clin Exp Immunol 2008;154:87-97. Peluso I, Morabito G, Riondino S, La Farina F, Serafini M. Lymphocytes as internal standard in oxidative burst analysis by cytometry: a new data analysis approach. J Immunol Methods 2012;379:61-5. Dransfield I, Buckle AM, Savill JS, McDowall A, Haslett C, Hogg N. Neutrophil apoptosis is associated with a reduction in CD16 (Fc gamma RIII) expression. J Immunol 1994;153:1254-63. Fingerle G, Pforte A, Passlick B, Blumenstein M, Strobel M, Ziegler-Heitbrock HW. The novel subset of CD14+/CD16+ blood monocytes is expanded in sepsis patients. Blood 1993;82:3170-6.
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
Figure 1
AC
Figure 2
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
Figure 3
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 4
ACCEPTED MANUSCRIPT Table I: CD16 levels of expression (mean +1SD) on granulocytes in 2 groups or patients with or without sepsis compared to healthy donors.
366.0+97.9 224.4+82.2 161.4+67.5
0,0000
0,0000
0,0000
NU
0,0000
0,0000
0,0087
TE CE P AC
T
Non septic vs healthy ns 0.006 ns P value
IP
418.2+81.3 311.4+66.4 189.2+72.1
D
unstimulated E coli PMA Comparisons E coli vs unstim PMA vs unstim PMA vs E Coli
p
ICU Patients + Sepsis p
197.4+131.2 207.6+96.6 92.8+62.0
SC R
Healthy
MA
Samples
Non septic patients
0,6153 0,0012 0,0000
Sepsis vs healthy <0.0001 0.002 0.003
ACCEPTED MANUSCRIPT Table II: Standardization of fluorescence intensity target settings inter instruments using PMA induced healthy donor sample.
CD16-PE-Cy7
15.3
3 917
120.0
30 720 4.4
NU MA D TE CE P AC
1
SC R
CD14 APC 126
IP
Rhodamine
T
Navios FACSCanto II
ACCEPTED MANUSCRIPT Table III level of Rhodamine fluorescence intensity (mean +1SD) produced by Monocytes under stimulation, in 2 groups of patients with or without sepsis compared to healthy donors.
MA D TE CE P AC
p 0.0063 0.0317 ns
T
ICU 0.3+0.2 1.10+0.69 2.75+1.86
IP
Cardiology p 2.2 + 2.3 0.0035 4.0+4.76 0.0168 3.89+2.5 0.0359 p value 0.0004 0.0891 0.0000 0.0025 0,0000 0,1752
SC R
healthy 0.18+0.04 0.67+0.31 2.31+0.6
0.00011 <0.00011 0,00415
NU
unstimulated E coli PMA comparisons E coli vs unstim PMA vs unstim PMA vs E coli
ACCEPTED MANUSCRIPT
0.37+0.5 10.5+11.1
ns 0.037
16.6+ 6.5
ns P value
10.2+ 5.9
0.0002
0,0010 0,0000 0,0497
TE CE P AC
SC R
0.0021 ns
NU
0,0038 0,0000 ns
Sepsis
p vs healthy
2.0+ 1.8 30.3+27.6
D
PMA Comparisons E coli vs unstim PMA vs unstim PMA vs E coli
healthy
MA
unstimulated E coli
Healthy 0.20+ 0.03 30.1 +26.5 17.9 + 3.4
p vs
IP
Non septic patients
T
Table IV: level of Rhodamine fluorescence intensity (median +1SD) by produced by PMN under activation, in 2 groups of patients with or without sepsis compared to healthy donors. (ns = non significant; vs = versus)
0,0009 0,0000 ns
ACCEPTED MANUSCRIPT Table V: E coli induced Rhodamine fluorescence by granulocytes (median +1SD). Role of plasma components in 16 ICU patients.
13.5+10.3
6.9+3.0
p=0,0012 p=0,0000 NS
CE P AC
SC R
IP
T
Heated healthy plasma
NS
p=0.0001
NU
p=0.0018 p=0.0082
TE
E coli vs unstim PMA vs unstim E Coli vs PMA
Healthy plasma substitution
p=0.0056
MA
unstimulated Stimulated healthy plasma washed cells
D
N = 16 unstimulated E coli PMA Compared to
Whole Washed blood (WB) blood 0.26+0.12 2.3+2.3 10.6+11.4 10.4+5.6
p=0.0006
S0022-1759(16)30220-4 doi:10.1016/j.jim.2016.09.010 JIM 12225
To appear in:
Journal of Immunological Methods
Received date: Revised date: Accepted date:
6 June 2016 11 August 2016 26 September 2016
Please cite this article as: Jeraiby, M., Sidi Yahya, K., Depince-Berger, Lambert, C., Microbicidal activity measured by flow cytometry: optimization and standardization for detection of primary and functional deficiencies, Journal of Immunological Methods (2016), doi:10.1016/j.jim.2016.09.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
IP
T
Microbicidal activity measured by flow cytometry: optimization and standardization for detection of primary and functional deficiencies
SC R
JERAIBY M1, SIDI YAHYA K, DEPINCE-BERGER, LAMBERT C1
MA
NU
Immunology Laboratory, CNRS UMR5307 Labo Georges Friedel (LGF), Pole de Biologie-Pathologie, University Hospital. St Etienne, France
D
Corresponding author: Dr. Claude Lambert, Immunology Laboratory, Pole de Biologie-Pathologie, UnivHospital, CHU St Etienne, F 42055 ST ETIENNE Cedex 2, France
TE
E-mail : [email protected].
CE P
Phone number: +33 4 77 12 05 13
AC
Fax number: + 33 4 77 12 05 52
Abbreviations:
HLA-DR: human leucocyte antigen D-related; CRP: C-reactive protein; PMN: Polymorphonuclear neutrophil; ROS: reactive oxygen species; E coli: Escherichia coli; DHR123: Dihydrorhodamine 123; R123: Rhodamine 123; EDTA: Ethylene Diamine Tetra Acetic Acid; BSA: Bovine serum albumin; PMA: Phorbol 12 Myristate 13Acetate; CV: Coefficient of variation; NADPH: Nicotin amide adenine dinucleotide phosphate; PBS: Phosphate-Buffered Saline; MdFI: Median florescence intensity; dim: diminished; FSC: Forward Scatter; SSC: Side Scatter; FITC: Fluorescein isothiocyanate; PE-Cy7 Phycoerythrine Cyanin 7; APC: Allophycocyanin; NS : non significant; LPS: lipopolysaccharide ; IL :
ACCEPTED MANUSCRIPT interleukine ; ICU : Intensive care unit;
IP
T
Abstract
Microbicidal activity is related to the production of reactive oxygen species (ROS)
SC R
that can be measured by Flow-Cytometry using Rhodamine 123 (R123). Few assays have been proposed to measure ROS production, usually on heparinized
NU
samples but none of them is standardized. Here we propose to improve the test by selecting polymorphonuclears (PMN) and monocytes, labelled and activated in
MA
one step to keep the test short, and to standardize the process even between different systems (i.e. Navios and FACSCanto) using Fluorescence Intensity
TE
D
Target Setting (“FITS”). We applied this test on 15 patients without inflammation, 19 patients from an intensive care unit (ICU) and 11 healthy volunteers. Results:
CE P
provided calcium restitution, we show that the test can be performed on EDTA that is a better sample preservative. The results were highly correlated between
AC
instruments (r² = 0,898). PMN CD16 (and not CD14) expression was altered under stimulation with E. coli (MdFI = 239.3+93.5) or PMA (139.7+76.8) as compared
to
resting
sample
(307.6+145.1).
RH123
was
strongly
and
homogeneously induced by PMA (14.2+6.6) and more heterogeneously by E coli (MdFI 21.9+23.4) as compared to unstimulated PNN (0.9+1.3, p<0.0001). The test useful for genetic disorders but is also for secondary deficiencies as observed in ICU (E coli RH123 MFI = 10.5+11.1 patients vs 30.1+26.5 in healthy donors). In ICU, CD16 expression was already altered on unstimulated samples (MdFI = 197.4+131.2 vs 418, 2+81.3 in healthy donors; p=<0.0001). Bacterial stimulation
ACCEPTED MANUSCRIPT was dependent of the complement that partly explains deficiency to bacterial
IP
T
stimulus in ICU patients.
Keywords: Microbicidal activity, oxidative burst, immunodeficiency,
NU
SC R
Rhodamine, flow cytometry, standardization, granulocytes.
Phagocytosis
of
MA
1. Introduction micro-organisms
or
cell
debris
is
the
main
task
of
D
polymorphonuclear leukocytes (PMNs). It is facilitated by plasma products such as complement components, antibodies and scavengers (1). This mechanism is
TE
named opsonisation. Several steps in opsonisation, including complement activation and adhesion, require cations such as Calcium and Magnesium.
CE P
Phagocytes contain a large arsenal of enzymes for proteolysis and production of reactive oxidative species (ROS) (2). Microorganism capture triggers NADPH oxidase that produces NADP+, superoxide ion (O2ˉ) and hydrogen peroxide
AC
(H2O2). Later myeloperoxidase transforms these species into hydroxyl radicals (°OH) and hypochlorite (ClO-). These newly produced ROS are highly effective in killing and digesting ingested particles including bacteria and fungi (microbicidal activity) (3,4). Inherited dysfunctions of NADPH oxidase or more rarely myeloperoxidase are responsible for defect in ROS production and accumulation of phagocytized products leading to Chronic Granulomatous Disease (CGD)(5). Microbicidal activity can be measured through the capacity of the cells to produce ROS for the diagnosis of CGD.
Tissue infection induces systemic effects with fever and inflammatory metabolism that includes production of acute phase proteins including opsonins. Systemic effects are induced by bacterial components such as lipopolysaccharide (LPS) or
ACCEPTED MANUSCRIPT peptidoglycan (6) and are mediated by the release of cytokines such as IFN, TNF or IL-6 (7,8) preparing peripheral cells for phagocytosis and migration to the inflammatory site. At later stage, cell exhaustion and regulatory mechanisms
T
induce a transient depression of phagocytosis called immunoparalysis (9).
IP
Immunoparalysis can be prolonged in sepsis, extended burning, injury or surgery
SC R
and contributes in the severity of secondary infections leading and consequently a high mortality rate in Intensive Care Units (ICU) (10-12). Immunoparalysis can be analyzed in peripheral blood cells of ICU patients. As an example, the high affinity IgG receptor FcRI (CD64) (13) is up regulated while the low affinity FcRIII
NU
receptor (CD16) on PMNs (14) and HLA DR molecule expression on the surface of monocytes are down-regulated (15,16). These markers have been shown to
MA
predict a poor clinical outcome of patients (17,18). Microbicidal activity can be measured by flow cytometry. Several procedures
D
have been proposed for diagnosis of CGD and can be applied for Immunoparalysis. They are usually based on measurement of production of ROS.
TE
ROS induces oxidization of probes such as Dihydrorhodamine 123 (DHR123) into
CE P
fluorescent Rhodamine 123 (R123) that can be measured in individual cells by flow cytometry (2,19).
These methods can be easily implemented to provide
more precise and reproducible output, e.g. by adding selective markers for granulocytes. Furthermore, in routine cell analysis, the use of Ethylene-diamine-
AC
tetra-acetic acid (EDTA) is recommended as the best preservative anticoagulant (20,21). But by chelating calcium, EDTA not only prevents coagulation but also cell aggregation, complement activation and phagocytosis (22). However, standardization of the test is required in accreditation process according to ISO 15189 (23) and attempts have been recently proposed, even between different systems (instruments, antibodies.) (24-26). In the present report, we propose to implement and standardize a microbicidal test, independently of the reagent and instrument used, for the diagnosis of CGD and we show its interest in monitoring immunoparalysis.
ACCEPTED MANUSCRIPT 2. Material and Method
T
2.1. Blood sample.
IP
The study was performed at University Hospital of Saint-Etienne (France) on blood samples collected from 15 adults patients from cardiac unit without
and eventually ongoing sepsis (group C)
SC R
biological sign of acute inflammation (group B), 19 patients from ICU with recent and 11 healthy volunteers from
occupational medicine (group A). Venous blood was collected on EDTA or lithium
NU
heparin tubes and tested within 12 hours of sampling. Patients from ICU had inflammation status as shown by raised level of CRP (from 14 to 366 mg/L; ref
MA
value <5 mg/L) and ongoing bacterial infection was identified in 8 patients (gram negative bacteria, n = 5, gram positive bacteria, n = 4). For legal reasons, this technical study was performed on anonymous samples. No more medical
D
information was available and measurement of CRP was not clinically justified in
TE
control donors and patients. All donors were older than 35 years and
2.2. Reagents
CE P
approximately equally distributed between males or females.
All reagents for phagocytosis induced oxidative burst assays were taken from
AC
FagoFlowEx® Kit (Cat. No. ED7042, EXBIO, Praha, Czech Republic) containing standardized lyophilized Escherichia coli bacteria (E coli), lyophilized DHR123, lyophilized Phorbol 12-myristate 13-acetate (PMA) as a stimulation control and ready-to-use
lysing
solution.
Labeled
monoclonal
antibodies
CD16-
Phycoerythrin–Cyanine (PE-Cy7; clone 3G8) and CD14-allophycocyanin (APC; clone MEM-15) were also purchased from EXBIO. EDTA samples were supplemented with Ca2+ at 10 mmol/L and calcium Heparinate at 250UI/ml final concentrations. Ca2+ replacement buffer was optimized on the criteria of sample coagulation and activation process. Different doses of calcium and heparin were tested around concentrations usually used. Suboptimal doses induced coagulation or lack of activation. Optimal doses of calcium and heparin were finally validated by comparing 10 individuals
ACCEPTED MANUSCRIPT simultaneously analyzed on heparin and EDTA anti-coagulated blood samples, EDTA being supplemented with Ca and heparin. The assay was also simplified. Cell stimulation and antibody labelling were
T
performed in one step for 20 minutes in 37°C water bath followed by fixation and
IP
red blood cells lysis done according to manufacturer instructions. In order to better preserve the sample, PBS-bovine serum albumin 10 g/L (MP Biomedical
SC R
Solon OH), EDTA (0.5 G/L; MP Biomedical) was added after erythrolysis for data acquisition.
Samples were analyzed on Navios™ flow cytometer (CE-IVD; 3 lasers, 10 colors)
NU
using Navios™ software (Beckman-Coulter; Fullerton, CA) within 1 hour from their final wash. The flow cytometer was daily calibrated on a routine basis using microbeads
(Flowset,
MA
fluorescent
Beckman-Coulter).
R123
median
fluorescence intensity (MdFI) was measured on at least 5.000 PMNs selected on CD16 labelling versus side-scatter (SSC) and on monocytes selected on
D
CD14/SSC. Data analyses were performed using Kaluza™ software (Beckman
TE
Coulter).
CE P
2.3. Standardization between instruments Protocol standardization on time and between laboratories can be managed by keeping fluorescence detections at the same levels on time as daily checked but
AC
also between different instruments using beads with standardized fluorescence intensity. As previously described (26), settings were first optimized on Navios; fluorescence targets were measured using calibration beads (Cyto-Cal Multifluor Calibrator, Thermo Scientific, Carlsbad CA) and voltages adjusted on BD FACSCanto™ II cytometer (BD Biosciences, San Jose, CA) to get the same fluorescence targets. The method was named as Fluorescence Intensity Target Settings procedure (FITS). On these settings, 20 samples, same preparation, were cross-analyzed on the two instruments in delays that did not exceed 3 hours.
2.4. Statistical Analyses
ACCEPTED MANUSCRIPT Statistical analysis was performed using, paired student’s T test, linear regressions and are expressed as median +/-1SD. Coefficient of variation (CV)
T
were calculated as 1SD/Mean.
IP
2.5. Legal issues
This technical study was performed on anonymous samples provided for
MA
NU
SC R
diagnosis purpose in accordance with French Law.
3. Results
D
3.1 .Method optimization: In order to implement the test precision and robustness, we used CD16 and CD14 to select neutrophil PMNs and monocytes respectively
TE
(fig 1a; 2a). The labelling was performed in one step during the activation step. The one step gave same results on preliminary tests as if samples were activated
CE P
and labelled in two steps. The induced microbicidal activity could be easily measured under stimulation with PMA (fig 1d, 2d) or E Coli (fig 1c, 2c) and compared to unstimulated sample (fig 1b, 2b) on the two leukocyte populations
AC
involved in phagocytosis: PMNs (fig 1 and 2b) or monocytes (fig 2c). Calcium supplementation on EDTA samples gave similar results as heparinized samples as shown on one representative sample (fig 2d) with a better sample stability at least up to 24hours storing at 4°C before testing on 5 healthy donors (Fig 2e).
3.2. PMN and monocyte activation test: overall CD16 expression was significantly decreased on granulocytes stimulated with E coli (MdFI = 239.3+93.5, paired t test: p<0.001) and even more when they were stimulated with PMA (139.7+76.8; p<0.0001) as compared to unstimulated cells (307.6+145.1) in 45 tests from the three clinical groups altogether (19 sepsis, 15 non infected patients, 11 healthy donors; fig 3a). The stimulation did not significantly change CD14 expression on monocytes (not shown).
ACCEPTED MANUSCRIPT
E coli induced very strong production of fluorescent R123 by PMNs (MdFI 21.9+23.4, n=45 tests) as compared to unstimulated PMNs (0.9+1.3, p<0.0001).
T
This Rhodamine production induced by E. coli was much heterogeneous (fig 1c)
IP
and significantly higher than under PMA activation (14.2+6.6; p<0.017; fig 1d). In some patients, R123 fluorescence was more or less distributed into 3 peaks of
SC R
progressive intensities: 1.6+0.5; 9.4+2.6; 50.6+9.1 respectively, suggesting 3 different mechanisms of cell activation. The low intensity peak was significantly higher than autofluorescence measured on stimulated samples without DHR123
NU
(MdFI 0.200+0.041 under PMA stimulus of 4 patients) as compared to unstimulated samples (MdFI 0.113+0.009). The highest peak was significantly
MA
higher than values observed with PMA stimulus.
In comparison, monocytes produced low and homogeneous R123 fluorescence
D
under E coli stimulation (2.0+3.2) although it was still significantly higher than
TE
unstimulated monocytes (0.9+1.6, p<0.0038) but lower than under PMA activation
CE P
(3.0+2.0, p<0.0001; table III).
3.3. Method validation: By repeating the E coli stimulation 10 times on two samples on the same day, by the same operator, we could observe a good
AC
repeatability for PMN CD16 labelling (CV =2.8 and 4.0% fig 2a). R123 fluorescence repeatability was not so good on PMN (CV =19.2 and 54.3%; fig 2b) but was better on monocytes (CV =13.4 and 31.6%; fig 2c). Because samples cannot be stored for a long time, reproducibility could not be measured but 20 EDTA samples stored at 4°C were tested for four consecutive days. Surprisingly we observed good E coli induced production of R123 fluorescence by surviving PMN at least on the first 24-48 hours of preservation (mean change at 48 hours: +19.7%, fig 2e), despite the rapid drop in number of surviving PMN. PMN death was difficult to quantify.
3.4. Protocol standardization between two types of instruments: In order to detect the same levels of fluorescence between different instruments, we measured
ACCEPTED MANUSCRIPT median fluorescence intensity (MFI) on Navios from one representative PMA activated sample for Rhodamine (15.3), CD16-PE-Cy7 (120) and CD14 (4.4) and used these values as targets for setting FACSCanto II with a correction factor
T
(x256) considering the differences in scales, giving targets at: Rhodamine (3 917),
IP
CD16-PE-Cy7 (30 720) and CD14 (1 126; Table II). Samples analyzed shown in fig 1 a’b’c’d’ FACSCanto II
SC R
simultaneously on the two systems show very comparable plots (one typical case as compared to Navios fig 1abcd).
Then, 20 (non-infected and healthy) samples were recorded on both instruments showing very close pictures with excellent correlation of fluorescence intensities
NU
on CD16 (fig 2f) and R123 induced by E coli (r² = 0,898) or PMA (r² = 0, 980; fig
MA
2g).
3.5. Clinical interest of the test: Because of the rarity of genetic deficiency (CGD) that would give clear defect in response, we focused on functional deficiency. In
D
order to check if the test could detect functional deficiency in severely ill patients,
TE
we compared PMN capacity to produce ROS in response to E coli between the 3 groups of patients on Navios. First (table I), reduced expression of CD16 could
CE P
be already observed in unstimulated samples of ICU patients (197.4+131.2) as compared to patients from group B (366.0+97.9) and to healthy donors (418.2+81.3 p<0.0001; fig 3a). Similar CD16 down regulation was observed after
AC
ex vivo stimulation in all groups (table I). Furthermore, ICU patient’s PMN had much lower microbicidal capacity as compared to healthy donors when induced by E coli (10.5+11.1 versus 30.1+26.5; p<0.037; fig 3a; table IV) or PMA (10.2+5.9 versus 17.9+3.4; p<0.0002; table IV). Microbicidal activity in Group B patients was not significantly different from healthy donors. Interestingly, E coli response could be stronger than PMA’s response in non-septic patients or healthy donors but this was not observed in ICU patients.
Considering Monocyte activity in healthy donors, E coli could induce some activity (MdFI: 0.67+0.31 compared to unstimulated monocytes 0.18+0.04, p=0.0004; table III) but at lower levels compared to PMA activation (2.31+0.6; p<0.0001; fig 3c).
Interestingly, in stressed, non septic patients (group B) unstimulated
ACCEPTED MANUSCRIPT monocytes had significantly higher activity (2.2+2.3; p=0.0035) as compared to healthy donors. This activity was increased under E coli (4.0+4.7) and PMA (3.9+2.5) stimulations to the same level, not more, than healthy donors. That
T
basal hyperactivity in stressed patients was found not so high in ICU patients
IP
(0.30+0.2; p<0.001) although it was still higher than in healthy donors. ICU Monocytes could further be activated by E coli, and at higher level than healthy
SC R
donors (1.10+0.69; p<0.0317) although PMA stimulation (2.75+1.86) was similar to healthy controls (Table III).
NU
3.6. Approach on the mechanism of E coli induced activation of granulocytes in ICU patients: As E coli induced microbicidal activity was defective (MdFI
MA
=10.6+11.4; fig 4; n=16) when tested on whole blood, we analyzed the possible role of plasma. E coli induced activity was at comparable level as under PMA stimulation (MdFI =10.4+5.6). In healthy donors, washing the blood cells
D
completely inhibit E coli activation. This was restored with plasma restitution from
TE
healthy donor unless the plasma was heat inactivated. In sepsis patients, plasma washing out strongly reduced but did not abolished the E coli activation (MdFI
CE P
2.3+2.3; p=0.0082) as compared to unstimulated PMN (0.26+0.12, p=0.002). Plasma substitution from healthy donors restored the E coli activation (13.5+10.3) at similar level as unwashed samples, but restoration was only partial if donor
AC
plasma was heat inactivated (6.9+3.0, p=0.0056 compared to untreated healthy plasma). Surprisingly, in ICU, E coli could significantly activate cells despite complement inactivation as compared to cells without plasma (p<0.0006) even if the activity was not completely restored as on unwashed cells (p<0.0001; see table V).
4. Discussion Testing microbicidal activity appeared to be a simple, reliable test despite it is a functional test requiring more sample preparation. We could show a clear induction of ROS production using PMA giving a narrow, homogeneous distribution of R123 fluorescence. This provides a good internal and interexperiment quality control. Activation with bacteria are more physiological and
ACCEPTED MANUSCRIPT probably more relevant clinically. Because of high variability in bacteria preparation, it is much safer using commercially available bacteria preparation that guarantees the reproducibility of the stimulus between experiments. The
T
study of phagocytic function and bactericidal activity has important clinical
IP
implications for the detection of genetic deficiency in which clear-cut results are expected (2,27). But this disease is rare and more applications can be clinically
SC R
relevant too. As an example, our data show a partial deficiency in a large proportion of severely ill patients in ICU, in accordance with immuno-paralysis that has been repeatedly reported (2,10,17,27,28). Measuring immunodeficiency
NU
can be of great help in these patients at high risk of secondary infection. The test is not new but we thought it can be easily implemented. By just adding 2
MA
antibodies, we could show that the test gets more precise information, at limited added cost. The sample preparation is not prolonged as labeling can simply be performed during stimulation. On this respect, 15-20 min incubation as
D
recommended by the manufacturer is a good compromise to have short duration
TE
test. A longer incubation might increase the response but this one is already very strong. This 3 color test is easy to set up and we chose fluorochromes with no
CE P
spectral overlap and that can be applied on most instruments. However, other conjugates can be easily used with R123 for instruments with only one laser or no far red detection.
AC
By this technique, two leukocyte populations (neutrophils and monocytes) from the same preparation could be tested for phagocytosis and oxidative burst without the need for further purification and treatment of the cells. Our results show that the two lineages do not have the same behavior. Note that NK cells (CD16+ low SSC) could be used as negative controls for DHR123 activation (29). Going to routine diagnosis in basic laboratories requires robustness and reproducibility for continuous monitoring of patients on time and for comparison between patients and across laboratories. Our proposal for standardization of the test appears to be easy to apply and most valuable upgrading at little or no extra cost, even different instruments based on Fluorescence Intensity Target Setting (“FITS”) as recently described (26). Beads or PMA activation of healthy donors can be used for inter-instrument settings while to E coli activation is too
ACCEPTED MANUSCRIPT heterogeneous to be considered as source for the targets. Our instrument settings are free to use by any lab interested and targets are detailed in table II. Because of clear calcium dependency of bacterial capture, heparin anticoagulant
T
is usually recommended by the manufacturers for this test. However, EDTA
IP
anticoagulant is usually preferred in routine diagnosis (including hemogramm) for its better preservative capacity than heparin and this can be a critical issue
SC R
especially for monitoring granulocytes functionality. Our experience is in favor of EDTA preservative. So because of sample availability and a better preservation, we
would
recommend
using
EDTA
and
we
show
that
the
calcium
NU
supplementation is simple and very convenient. Based on our comparison, the results on EDTA samples should be considered similar to Heparin samples. In
MA
any case, we recommend keeping the same procedure (either heparin or EDTA) for the same study.
Because of their fragility, it is usually recommended to perform any PMN test
D
within 8 hours. We were surprised to see repeatedly that PMN microbicidal
TE
activity was still correct after one night of storage at 4°C. This is not enough to recommend this delay of storage but sometimes, logistic issues are difficult to
CE P
achieve in short delays. The activity still detectable after one night may be biased because most fragile, poorly functional granulocytes are probably lost preferentially in the first few hours of storage. In this respect, using CD16 may
AC
help better select the granulocytes as PMN side scatter rapidly decreases after few hours of storage. CD16 brings information on cell stress. Indeed, we observed that CD16 expression strongly decreases on stimulated cells as it was previously reported in pre-apoptosis (30). The meaning is probably different from Rhodamine readout as CD16 decrease was significantly higher under PMA stimulation compared to E coli while R123 activation was much higher and heterogeneous with E coli compared to PMA activation. If using CD16/CD14 labelling does not increase the time of the experiment, it does increase the reagent cost that can be justified by several advantages. Neutrophils PMN are more precisely selected as CD16 expression is better preserved on time and help separating singlet from micro-aggregates. CD16 levels of expression give good information on cell stress, even before stimulation.
ACCEPTED MANUSCRIPT CD16 changes were observed in ICU patients but also in patients with not such strong inflammatory status. With the use of CD14 and CD16, it is also possible to analyze activity of monocyte subpopulations M1 (CD14+CD16-) and M2
T
(CD14dim or negative and CD16+) even if this distinction is more difficult to do on
IP
activated cells (31). DHR123 activation by PMN was very efficient, even in a short period of incubation; with one or two decades of increase, which makes negligible
SC R
the activation induced autofluorescence changes. However, quantifying R123 fluorescence is not always straight since the distribution is very wide and heterogeneous (see fig 3b). This may partly explain the wide variability of E coli
NU
activation between patients and even on repeats in single patient. Heterogeneous distribution of the response may reflect cell population heterogeneity due to cell
MA
rapid ageing or differences in maturation / pre-activation stages. Technical issues may also explain this heterogeneity through uneven distribution of stimulus per cell because of possible E coli micro-aggregation. Sample preparation has to be
D
very strict with good homogenization of E coli before and after adding the cells in
TE
order to have as much as possible an even distribution of bacteria on PMN. In any case, positivity is usually close to 100% and it is better considering the
CE P
fluorescence intensity shift. Median is of course more representative than mean fluorescence intensity but still do not reflect the discrete distribution among positive cells.
AC
Alternatively, peripheral monocytes were poorly activated by E coli, close to levels of PMA induced activity although pathogen associated products are known to directly activate macrophages. This might be due to a lack of maturation and possibly a different mechanism as compared to granulocytes. In any case, measuring bactericidal activity in monocytes represents a good internal reference for comparison of PMN. Despite that our technical study was not designed to approach clinical significance of the test; our preliminary data have shown strong reduction of capacity of granulocyte to activate DHR123 under E coli stimulation in ICU patients. This corroborated with reduced CD16 expression even observed on unstimulated cells, suggesting a pre-activation status of cells with possible refractoriness or exhaustion as previous well documented, with poor prognosis.
ACCEPTED MANUSCRIPT
In conclusion: measuring bactericidal activity by flow cytometry is easy and quick to perform and can be easily standardized for routine detection of functional
T
deficit specially in primary genetic disorders or in acquired immuno-paralysis as
AC
CE P
TE
D
MA
NU
SC R
IP
frequently observed in ICU.
ACCEPTED MANUSCRIPT Legends
SC R
IP
T
Figures Figure 1. Flow cytometry PMN selection on CD16 vs side scatter of a lysed whole blood specimen on 2 different instruments Navios (a,b,c and d) and FACSCAntoII (a’,b’,c’ and d’). Typical PMN produced rhodamine fluorescence histograms of resting (b, b’) or under E coli (c, c’) or PMA (d, d’) stimulation in one representative donor. E coli stimulus show a strong, heterogeneous response compared to PMA Stimulation.
MA
NU
Figure 2. Optimal E coli induced phagocytosis test was repeated 10 times at the same day on PMN (a, b) and monocytes (c) compared to one non stimulation and one PMA stimulation on one representative sample. Sample preservation on EDTA or Heparin was compared in few samples (d) and EDTA preservation tested on 4 days in 20 samples (e). CD16 labelling and E coli induced Rhodamine fluorescence from PMN were compared between two systems after Fluorescence Intensity target settings using a PMA induced sample, on Navios (Beckman Coulter) and a Canto II (BD Biosciences).
CE P
TE
D
Figure 3. Expression of CD16 (a) and Rhodamine fluorescence (b) on PMN or monocytes (c) were compared between healthy, non-infected and infected patients.
AC
Figure 4. The Role of plasmatic components in E coli induced production of Rhodamine RH123 fluorescence by granulocytes. Whole blood was analyzed alone (unstimulated, column 1) or stimulated with E Coli (column 2) or PMA (column 6). E coli stimulation was also tested on washed blood (column3-5), restituted with Hanks buffer directly (column 3) or with an healthy donor plasma, directly (column 4) or after heat inactivation (column 5) on 18 ICU patients.
ACCEPTED MANUSCRIPT
T
Tables Table I: CD16 levels of expression (median +1SD) on PMN in patients with or without sepsis compared to healthy donors.
SC R
IP
Table II : Standardization between different systems using of Fluorescence Intensity Target Settings considering respective scales.
NU
Table III : Rhodamine fluorescence intensity (mean +1SD) produced by Monocytes under stimulation, in patients with (n=17) or without (n = 16) sepsis compared to healthy donors (n=10).
MA
Table IV: Rhodamine fluorescence intensity (median +1SD) produced by PMN under activation, in patients with or without sepsis compared to healthy donors.
AC
CE P
TE
D
Table V: E coli induced Rhodamine fluorescence by PMN (median +1SD). Role of plasma components in 16 ICU patients comparing whole blood with washed blood and plasma either neat or heat inactivation replacement from one healthy donor.
ACCEPTED MANUSCRIPT
References McDonald JU, Rosas M, Brown GD, Jones SA, Taylor PR. Differential dependencies of monocytes and neutrophils on dectin-1, dectin-2 and complement for the recognition of fungal particles in inflammation. PLoS One 2012;7:e45781. Petropoulos M, Karamolegkou G, Rosmaraki E, Tsakas S. Hydrogen peroxide signals E. coli phagocytosis by human polymorphonuclear cells; up-stream and down-stream pathway. Redox Biol 2015;6:100-105. Marchi LF, Sesti-Costa R, Ignacchiti MD, Chedraoui-Silva S, Mantovani B. In vitro activation of mouse neutrophils by recombinant human interferongamma: increased phagocytosis and release of reactive oxygen species and pro-inflammatory cytokines. Int Immunopharmacol 2014;18:228-35. Letiembre M, Echchannaoui H, Bachmann P, Ferracin F, Nieto C, Espinosa M, Landmann R. Toll-like receptor 2 deficiency delays pneumococcal phagocytosis and impairs oxidative killing by granulocytes. Infect Immun 2005;73:8397-401. Emmendorffer A, Nakamura M, Rothe G, Spiekermann K, Lohmann-Matthes ML, Roesler J. Evaluation of flow cytometric methods for diagnosis of chronic granulomatous disease variants under routine laboratory conditions. Cytometry 1994;18:147-55. De Marzi MC, Todone M, Ganem MB, Wang Q, Mariuzza RA, Fernandez MM, Malchiodi EL. Peptidoglycan recognition protein-peptidoglycan complexes increase monocyte/macrophage activation and enhance the inflammatory response. Immunology 2015;145:429-42. Fernando MR, Reyes JL, Iannuzzi J, Leung G, McKay DM. The pro-inflammatory cytokine, interleukin-6, enhances the polarization of alternatively activated macrophages. PLoS One 2014;9:e94188. Striz I, Brabcova E, Kolesar L, Sekerkova A. Cytokine networking of innate immunity cells: a potential target of therapy. Clin Sci (Lond) 2014;126:593612. Volk HD, Reinke P, Docke WD. Clinical aspects: from systemic inflammation to 'immunoparalysis'. Chem Immunol 2000;74:162-77. Fitrolaki DM, Dimitriou H, Kalmanti M, Briassoulis G. CD64-Neutrophil expression and stress metabolic patterns in early sepsis and severe traumatic brain injury in children. BMC Pediatr 2013;13:31. Trimmel H, Luschin U, Kohrer K, Anzur C, Vevera D, Spittler A. Clinical outcome of critically ill patients cannot be defined by cutoff values of monocyte human leukocyte antigen-DR expression. Shock 2012;37:140-4. Janols H, Wullt M, Bergenfelz C, Bjornsson S, Lickei H, Janciauskiene S, Leandersson K, Bredberg A. Heterogeneity among septic shock patients in a set of immunoregulatory markers. Eur J Clin Microbiol Infect Dis 2014;33:313-24. Li S, Huang X, Chen Z, Zhong H, Peng Q, Deng Y, Qin X, Zhao J. Neutrophil CD64 expression as a biomarker in the early diagnosis of bacterial infection: a metaanalysis. Int J Infect Dis 2013;17:e12-23.
4.
8. 9. 10. 11. 12.
13.
SC R
CE P
7.
AC
6.
TE
D
5.
NU
3.
MA
2.
IP
T
1.
ACCEPTED MANUSCRIPT 14.
21. 22. 23. 24.
25.
26.
27. 28.
SC R
NU
MA
20.
D
19.
TE
18.
CE P
17.
AC
16.
IP
T
15.
Poehlmann H, Schefold JC, Zuckermann-Becker H, Volk HD, Meisel C. Phenotype changes and impaired function of dendritic cell subsets in patients with sepsis: a prospective observational analysis. Crit Care 2009;13:R119. Gouel-Cheron A, Allaouchiche B, Guignant C, Davin F, Floccard B, Monneret G, AzuRea G. Early interleukin-6 and slope of monocyte human leukocyte antigen-DR: a powerful association to predict the development of sepsis after major trauma. PLoS One 2012;7:e33095. Lambert C, Preijers FW, Demirel GY, Sack U. Monocytes and macrophages in flow: An ESCCA initiative on advanced analyses of monocyte lineage using flow cytometry. Cytometry B Clin Cytom 2015. Monneret G, Venet F. Sepsis-induced immune alterations monitoring by flow cytometry as a promising tool for individualized therapy. Cytometry B Clin Cytom 2015. Spittler A, Roth E. Is monocyte HLA-DR expression predictive for clinical outcome in sepsis? Intensive Care Med 2003;29:1211-2. Chen Y, Junger WG. Measurement of oxidative burst in neutrophils. Methods Mol Biol 2012;844:115-24. Davis BH, Dasgupta A, Kussick S, Han JY, Estrellado A, Group IIW. Validation of cell-based fluorescence assays: practice guidelines from the ICSH and ICCS part II - preanalytical issues. Cytometry B Clin Cytom 2013;84:286-90. Banfi G, Salvagno GL, Lippi G. The role of ethylenediamine tetraacetic acid (EDTA) as in vitro anticoagulant for diagnostic purposes. Clin Chem Lab Med 2007;45:565-76. Elsner J, Oppermann M, Czech W, Kapp A. C3a activates the respiratory burst in human polymorphonuclear neutrophilic leukocytes via pertussis toxinsensitive G-proteins. Blood 1994;83:3324-31. Sack U, Barnett D, Demirel GY, Fossat C, Fricke S, Kafassi N, Nebe T, Psarra K, Steinmann J, Lambert C. Accreditation of flow cytometry in Europe. Cytometry B Clin Cytom 2013;84:135-42. Kalina T, Flores-Montero J, van der Velden VH, Martin-Ayuso M, Bottcher S, Ritgen M, Almeida J, Lhermitte L, Asnafi V, Mendonca A and others. EuroFlow standardization of flow cytometer instrument settings and immunophenotyping protocols. Leukemia 2012;26:1986-2010. van de Loosdrecht AA, Alhan C, Bene MC, Della Porta MG, Drager AM, Feuillard J, Font P, Germing U, Haase D, Homburg CH and others. Standardization of flow cytometry in myelodysplastic syndromes: report from the first European LeukemiaNet working conference on flow cytometry in myelodysplastic syndromes. Haematologica 2009;94:1124-34. Solly F, Rigollet L, Baseggio L, Guy J, Borgeot J, Guerin E, Debliquis A, Drenou B, Campos L, Lacombe F and others. Comparable flow cytometry data can be obtained with two types of instruments, Canto II, and Navios. A GEIL study. Cytometry A 2013;83:1066-72. Nordenfelt P. Quantitative assessment of neutrophil phagocytosis using flow cytometry. Methods Mol Biol 2014;1124:279-89. Danikas DD, Karakantza M, Theodorou GL, Sakellaropoulos GC, Gogos CA. Prognostic value of phagocytic activity of neutrophils and monocytes in
ACCEPTED MANUSCRIPT
T
IP
SC R
NU MA D TE
31.
CE P
30.
AC
29.
sepsis. Correlation to CD64 and CD14 antigen expression. Clin Exp Immunol 2008;154:87-97. Peluso I, Morabito G, Riondino S, La Farina F, Serafini M. Lymphocytes as internal standard in oxidative burst analysis by cytometry: a new data analysis approach. J Immunol Methods 2012;379:61-5. Dransfield I, Buckle AM, Savill JS, McDowall A, Haslett C, Hogg N. Neutrophil apoptosis is associated with a reduction in CD16 (Fc gamma RIII) expression. J Immunol 1994;153:1254-63. Fingerle G, Pforte A, Passlick B, Blumenstein M, Strobel M, Ziegler-Heitbrock HW. The novel subset of CD14+/CD16+ blood monocytes is expanded in sepsis patients. Blood 1993;82:3170-6.
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
Figure 1
AC
Figure 2
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
Figure 3
TE
D
MA
NU
SC R
IP
T
ACCEPTED MANUSCRIPT
AC
CE P
Figure 4
ACCEPTED MANUSCRIPT Table I: CD16 levels of expression (mean +1SD) on granulocytes in 2 groups or patients with or without sepsis compared to healthy donors.
366.0+97.9 224.4+82.2 161.4+67.5
0,0000
0,0000
0,0000
NU
0,0000
0,0000
0,0087
TE CE P AC
T
Non septic vs healthy ns 0.006 ns P value
IP
418.2+81.3 311.4+66.4 189.2+72.1
D
unstimulated E coli PMA Comparisons E coli vs unstim PMA vs unstim PMA vs E Coli
p
ICU Patients + Sepsis p
197.4+131.2 207.6+96.6 92.8+62.0
SC R
Healthy
MA
Samples
Non septic patients
0,6153 0,0012 0,0000
Sepsis vs healthy <0.0001 0.002 0.003
ACCEPTED MANUSCRIPT Table II: Standardization of fluorescence intensity target settings inter instruments using PMA induced healthy donor sample.
CD16-PE-Cy7
15.3
3 917
120.0
30 720 4.4
NU MA D TE CE P AC
1
SC R
CD14 APC 126
IP
Rhodamine
T
Navios FACSCanto II
ACCEPTED MANUSCRIPT Table III level of Rhodamine fluorescence intensity (mean +1SD) produced by Monocytes under stimulation, in 2 groups of patients with or without sepsis compared to healthy donors.
MA D TE CE P AC
p 0.0063 0.0317 ns
T
ICU 0.3+0.2 1.10+0.69 2.75+1.86
IP
Cardiology p 2.2 + 2.3 0.0035 4.0+4.76 0.0168 3.89+2.5 0.0359 p value 0.0004 0.0891 0.0000 0.0025 0,0000 0,1752
SC R
healthy 0.18+0.04 0.67+0.31 2.31+0.6
0.00011 <0.00011 0,00415
NU
unstimulated E coli PMA comparisons E coli vs unstim PMA vs unstim PMA vs E coli
ACCEPTED MANUSCRIPT
0.37+0.5 10.5+11.1
ns 0.037
16.6+ 6.5
ns P value
10.2+ 5.9
0.0002
0,0010 0,0000 0,0497
TE CE P AC
SC R
0.0021 ns
NU
0,0038 0,0000 ns
Sepsis
p vs healthy
2.0+ 1.8 30.3+27.6
D
PMA Comparisons E coli vs unstim PMA vs unstim PMA vs E coli
healthy
MA
unstimulated E coli
Healthy 0.20+ 0.03 30.1 +26.5 17.9 + 3.4
p vs
IP
Non septic patients
T
Table IV: level of Rhodamine fluorescence intensity (median +1SD) by produced by PMN under activation, in 2 groups of patients with or without sepsis compared to healthy donors. (ns = non significant; vs = versus)
0,0009 0,0000 ns
ACCEPTED MANUSCRIPT Table V: E coli induced Rhodamine fluorescence by granulocytes (median +1SD). Role of plasma components in 16 ICU patients.
13.5+10.3
6.9+3.0
p=0,0012 p=0,0000 NS
CE P AC
SC R
IP
T
Heated healthy plasma
NS
p=0.0001
NU
p=0.0018 p=0.0082
TE
E coli vs unstim PMA vs unstim E Coli vs PMA
Healthy plasma substitution
p=0.0056
MA
unstimulated Stimulated healthy plasma washed cells
D
N = 16 unstimulated E coli PMA Compared to
Whole Washed blood (WB) blood 0.26+0.12 2.3+2.3 10.6+11.4 10.4+5.6
p=0.0006