We hypothesise that TLR4 activity may play a key role in pathogenesis of pre-eclampsia through one or more of its endogenous ligands (e.g. heparan sulphate, fibrinogen, fibronectin and hyaluronan), produced by the ischaemic placenta. Additionally these endogenous ligands may play some role in activation of leukocytes and endothelial dysfunction, resulting in the exaggerated systemic inflammatory response seen in this condition. On the other hand it may represent a protective adaptive mechanism of the human host to combat the dysregulation of angiogenesis that occurs in PE from reduced VEGF expression, itself secondary to sFlt1 release as a consequence of placental dysfunction.
To identify expression levels of TLR4 and TLR2 on peripheral blood monocytes and its different subpopulations from pre-eclamptic patients compared to those from normal pregnancy and non pregnant controls.
To characterise the expression levels of TLR4 and TLR2 in human umbilical vein endothelial cells (HUVEC) from pre-eclamptic patients versus normal pregnant women.
To determine which endogenous ligands can activate TLR4 in monocytes and HUVEC cells by investigating the pro- and anti-inflammatory cytokines responses when they are exposed to TLR4 ligands.
To investigate the expression of adhesion molecules, angiogenic factors, markers of apoptosis, chemokines and cytokines by endothelial cells (HUVEC) and monocytes when exposed to ligands of TLR4 during co-culture experiments.
To determine the effects of inhibiting TLR4 on markers of monocyte activation and endothelial cell damage and dysfunction.
In summary, to determine whether TLR4 signalling plays a modulatory role in interactions between monocytes and endothelial cells in the pathogenesis of PE.
3. Materials and Methods:
3.1 Materials:
3.1.1 Ethical approval:
Ethical approval for this study was received from South Yorkshire Research Ethics Committee (09/H1310/12). Detailed information sheet was given to all study participants and written informed consent was obtained.
3.1.2 Subjects:
Women with established pre-eclampsia <34 weeks gestation diagnosed by the criteria of the International Society for the Study of Hypertension in Pregnancy (ISSHP) (55) were recruited to the study. The patients were approached in the Day Assessment Unit of Jessop Wing Hospital when the diagnosis was confirmed. Healthy normotensive pregnant controls were recruited from women attending the routine antenatal clinics at the Jessop Wing. Healthy non-pregnant female volunteers were also studied to determine baseline non-gestational levels of the study proteins of interest. The non pregnant women had normal menstrual cycles and were not on hormonal contraception. Experiments have been performed on blood samples from six cases of pre-eclampsia, nine normal pregnant women and seven non pregnant women (Table 2).
Table 2: Study groups
Donor
Age (years)
*Mean ±SD
Gestational Age (weeks)
*Mean ±SD
Parity
*range
Systolic pressure
(mm Hg)
*Mean ±SD
Diastolic
Pressure
(mm Hg)
*Mean ±SD
Proteinureia (+ dipstick)
*range
Pre-eclampsia
Healthy pregnant
Non pregnant
33 ± 6.3
27.6 ± 2.2
32 ± 2.2
32 ± 2.4
26 ± 1.6
1-2
1-4
156.4 ± 4.6
110 ± 10
93.4 ± 9
70 ± 5
2-3
0
3.1.3 Samples:
Following the consent obtained from the study groups an additional 12 ml of fresh blood was collected from each subject at the time of a scheduled blood collection into a tube containing EDTA which prevents clotting of the blood. Blood so obtained was then immediately taken to the research laboratories at the Academic Unit of Obstetrics and Gynaecology at the Jessop Wing for isolation of monocytes, storage of aliquotted serum samples and further laboratory work.
3.1.4 Laboratory equipments:
Cell culture hood was used to prevent cell contamination during the isolation of either peripheral blood monocytes or human umbilical vein endothelial cells.
Different sizes of pipettes (200μl-1000μl, 50μl-200μl and 2μl-20μl).
50ml Falcon tubes.
Microcentrifuge eppendorf tubes.
Centrifuge (MSE Mistral 2000, Calserve Ltd).
4-8â°C Laboratory refrigerator.
-20 â°C and -80 â°C Freezers.
Light microscope was used for cells counting (Zeiss, West Germany) with Ã-10 magnification power.
Incubator with 37 â°C temperature and 5% COâ‚‚.
3.1.5 Reagent and Media:
Table 3: Reagents and media, which were of the highest quality available, were used for monocytes isolation and culture:
Reagent
Supplier
Catalogue Number
Hank's Balanced Solution (HBSS) without Phenol red, Ca2+ or Mg2+
Fetal Bovine Serum (FBS)
Ficoll-Paque
IMDM without l-Glutamine
l-Glutamine
Phosphate Buffered Saline (PBS)
Penicillin-streptomycin
Lonza
Lonza
GE Healthcare
Lonza
Sigma
Sigma
Sigma
BE10-547F
DE14-80
17-1440-02
12-726F
019K2360
1008036
9H0387
Table 4: Bacterial and endogenous ligands of TLR2 and TLR4 (stimulators):
Stimulator
Supplier
Catalogue Number
Lipopolysaccharides (rough strains) from Escherichia coli EH 100
Peptidoglycans from staphylococcus
Fibrinogen from human plasma
Heparan Sulfate Proteoglycan from basement membrane
select-Hyaluronan molecular mass: 125-175 kDa
Fibronectin from human serum
Sigma-Aldrich
Sigma-Aldrich
Sigma-Aldrich
Sigma-Aldrich
Sigma-Aldrich
BD Biosciences
L9641
77140
F3879
H4777
S0201
354008
Table 5: Monoclonal antibodies were used in the Flow Cytometry method:
Monoclonal antibody
Supplier
Catalogue Number
Phycoerythrin (PE) conjugated antibody - TLR4
FITC conjugated antibody - TLR2
APC-Alexa Flour 780 conjugated antibody - CD14
e Fluor 450 conjugated antibody - CD16
Allophycocyonin (APC) conjugated
antibody - Tie2
Precyp5.5-cy5.5 conjugated antibody - (MHCII) HLA-DR
PE conjugated antibody - mouse IgG2a isotype control
FITC conjugated antibody - mouse IgG2a-k isotype control
APC conjugated antibody - mouse IgG1-k isotype control
Precyp5.5-cy5.5 conjugated antibody - mouse IgG2a-k isotype control
e Bioscience
e Bioscience
e Bioscience
e Bioscience
R&D systems
BD Bioscience
e Bioscience
e Bioscience
BD Bioscience
BD Bioscience
12-9917-42
11-9922
27-0149
48-0168
FAB3131A
552764
12-4724-42
11-4724-73
555751
558020
Table 6: Reagents and consumables were used for human umbilical vein endothelial cells (HUVEC) isolation and culture:
Reagent
Supplier
Catalogue Number
M199 10X solution, 500m
MEM, 500ml
Gelatin, from porcine skin, 100g
Endothelial Cell Growth Supplement, 50mg
Heparin Sodium Salt
Collagenase Type IV
HEPES buffer 1M, 100ml
New born calf serum
Trypsin-EDTA
Amphotericin solution
Ported IV cannula 14G-orange
Injection bungs
Plastic tubing clamps
Syringe filter unit (0.02ïm; filtrospur)
Tissue culture flasks(25 cm2)
Fisher
Lonza
Sigma
Tebu Bio
Sigma
Sigma
Lonza
Sigma
Sigma
VWR
SP services
Vygon
VWR
Sarstedt
Greiner bio-one
VX21180021
BE12-611F
G1890-100G
1003
H3149
C5138
BE17-737
N4637-100ML
T4299-100ML
AMREK721-20
IN/223P
880.5
229-0332
83.1826.001
690160
3.2 Methods:
3.2.1 Isolation of Peripheral Blood Mononuclear Cells (PBMCs):
Peripheral blood mononuclear cells (PBMCs) were isolated from fresh venous blood on the top of Ficoll-Paque and centrifugation using a protocol previously described (56) as listed below:
The blood Mixed with Hank's balanced solution, HBSS (without phenol red, Ca2+ or Mg2+) enriched with 2% (v/v) fetal bovine serum (FBS) at rate of 1:1 into 50 ml falcon tube.
The mixture of blood and HBSS was overlaid slowly onto top of 20 ml Ficoll- Paque solution in the 50ml falcon tube.
The falcon tube was centrifuged for 40 minutes at 450g/1400RPM under 4â°C.
The PBMCs layer (visualised as a distinct white layer of cells between the upper serum and lower Ficoll layers) was removed and transferred to a fresh tube.
The cells were washed three times by re-suspending the pelleted cells with 10ml of HBSS (without phenol red, Ca2+ or Mg2+) enriched with 2% FBS and centrifuged for ten minutes at 450g/1400RPM each time.
The PBMCs were counted using a haemocytometer method. The mean cell for each patient approximated to 26 Ã-10â¶cells/ml. These cells were divided into two parts. The first part was used for monocytes culturing and stimulation with ligands of TLR2 and TLR4. The second part was used for flow cytometry for determination of monocyte phenotypes.
3.2.2 Monocyte Culturing and Stimulation:
The isolated PBMCs were then seeded in 12-well tissue culture flat bottom plate (Greiner bio-one/catalogue number: 665180) at a density of 1.8 Ã- 10ⶠcells/ well. The cells were incubated with medium (IMDM supplemented with 2% FBS, 2mM L-Glutamine and penicillin-streptomycin) at 37 â°C with 5% COâ‚‚. After two hours of incubation the medium was removed and the monocytes, which adhere to the surface of the plates, were washed twice with cold PBS. The medium was then replaced (IMDM supplemented with 2% FBS, 2mM l-Glutamine and penicillin-streptomycin) for monocyte cells in the tissue culture plate.
The monocytes were then treated with 10ng/ml Lipopolysaccharides (rough strains) from Escherichia coli, 5µg/ml Peptidoglycan from Staphylococcus aureus, 10µg/ml Heparan Sulfate Proteoglycan, 5µg/ml Fibronectin, 2mg/ml Fibrinogen and 100µg/ml select-Hyaluronan of molecular mass: 125-175 kDa as described in previous studies (57) (58) & (41). All ligands were purchased from Sigma-Alderich except Fibronectin supplied by BD Bioscience. The stimulated monocytes were incubated for 24 hours at 37 â°C with 5% COâ‚‚ and conditioned medium was collected after this time. Medium was aliquoted into small volumes and frozen at -20 â°C until analysis was undertaken.
3.2.3 Flow Cytometry:
Flow Cytometry was used to identify expression levels of TLR2, TLR4, human leukocyte class II DR antigens (HLA-DR) and tyrosine kinase receptor (Tie2) on the peripheral blood monocytes. It is a technique used for analysing cells and sub-cellular particles by passing them through a fluid system and laser beam. Then absorbed lights and fluorescent intensities are measured as indicator for cell size and granularity. Additionally, if these particles have been labelled with fluorochromes, specific surface proteins are measured by laser excitation of these dyes.
3.2.3.1 Mononuclear Cells Stained with Monoclonal Antibodies:
The PBMCs were labelled with the following monoclonal antibodies: FITC conjugated antibody against TLR2, PE conjugated antibody against TLR4, APC-Alexa Fluor 780 conjugated antibody against CD14, e Flour450 conjugated antibody against CD16, APC conjugated antibody against Tie2 and Precyp5.5-cy5.5 conjugated antibody against HLA-DR.
Initially the cells (3.2Ã-10ⶠcells/tube) were distributed into four microcentrifuge eppendorf tubes and blocked with 5µl of anti human Fc receptor blocking reagent (MACS-Miltenyi Biotic/catalogue number:1 30-059-901) and refrigerated for 10 minutes. Then cells were mixed with monoclonal antibodies as follows:
The first eppendorf tube had unstained cells only.
The second one had isotype negative controls matched for monoclonal antibodies which were mouse IgG2a-ĸ conjugated to FITC, IgG2a conjugated to PE, IgG1ĸ conjugated to APC and IgG2a-k conjugated to Precyp5.5-cy5.5.
The third eppendorf tube had PBMCs stained with all monoclonal antibodies (5μl of CD14 conjugated to APC-Alexa Fluor 780 antibody, 5μl of CD16 conjugated to e Flour450 antibody, 20μl of TLR2 conjugated to FITC antibody, 5μl of TLR4 conjugated to PE antibody, 20μl of Tie2 conjugated to APC antibody and 5μl of HLA-DR conjugated to Precyp5.5-cy5.5 antibody) as mentioned in company's sheets.
And the last eppendorf tube had cells mixed with 5μl of CD14 conjugated to APC-Alexa Fluor 780 antibody, 5μl of CD16 conjugated to e Flour450 antibody and isotype negative controls matched for TLR2, TLR4, HLA-DR and Tie2 monoclonal antibodies - the fluorescence minus one method.
All eppendorf tubes were incubated in refrigerator for 30 minutes. Cells were then washed twice by cold BD flow cytometry stain buffer (FSB/catalogue no: 554656) and centrifuged at 5000RPM for 2 minutes each time.
3.2.3.2 Flow Cytometric Analysis:
Analysis was done using a multi-laser and detector analyser (LSR II Flow Cytometer from BD Bioscience) at the Flow Cytometry department in the core facility of University of Sheffield Medical School. FACS DIVA-BD Bioscience software was used for analysis. Auto compensation was calculated using anti mouse compensation beads sets obtained from BD Bioscience (catalogue number: 552843), was performed as directed by the manufacturer. The different blood cell populations were identified in the forward scatter (FSC) and side scatter (SSC) dot plot. A gate specific for the monocytes population was created and a total of 10,000 events were acquired. Monocyte sub-populations were identified according to CD14 and CD16 surface markers as defined previously by Ziegler-Heitbrock (18, 21) .
3.2.4 Cytometric Bead Array:
Cytometric bead array technique was used to measure IL-10, an anti-inflammatory cytokine, and TNF-ï¡, a pro-inflammatory cytokine, within monocyte conditioned medium. BD cytometric bead array is a method used to measure multiple soluble proteins such as cytokines in small volumes (<50μl). BD-CBA kits contain suspended beads bound with antibodies that capture soluble protein of interest in the sample (Figure 6). These beads are incubated with the sample and then incubated further in the presence of fluorescent antibody conjugated with the phycoeryithrin (PE) fluorochrome. Additionally, a set of standards is prepared that enables protein measuring between 10 and 2500 pg/ml. Thereafter, soluble analytes are analysed by detection of a fluorescence-based, flow cytometric technique. This assay was performed following the instructions of the master buffer kit manual (BD-catalogue number: 558683) at the Flow Cytometry Core Facility in the Sheffield University Medical School. BD FACS Array flow cytometry machine and FCAP Array software were used for analysis.
3.2.5. NF-ï«B Experiment:
The nuclear factor-kappa B (NF-ï«B) is a dimer formed from five subunits (p100/p52, p105/p50, RelA (p65), RelB and c-Rel (59). The most commonly studied dimer is that consisting of the p50 and p65 subunits. During unstimulated, resting conditions the p50-p65 heterodimer is coupled to I-kappa Bï¡ (Iï«Bï¡). After stimulation with ligands of TLR2 and TLR4, the Iï«Bï¡ is phosphorylated causing dissociation of Iï«Bï¡ -p50-p65 complex. This results in nuclear translocation of NF-ï«B subunits, which induces expression of inflammatory cytokines (Figure 7). Thus the induction of signalling pathways by ligands of TLR2 and TLR4 can be studied with staining of intracellular phosphorylated NF-ï«B p65 experiment.
Following the isolation, culturing and stimulation of monocytes with 100ng/ml LPS (TLR4-stimulator) or 5μg/ml PG (TLR2-stimulater) for 15 minutes as described above, the following experiment was performed. The cell fixation and permeabilization were done as directed in the BD-Bioscience protocol for staining of intracellular phosphorylated NF-ï«B. After cells were stimulated, they were fixed immediately by pre-warmed BD Cytofix buffer (catalogue no: 554655) and incubated at 37â°C for 10 minutes. Cell centrifugation was done at 300g for 5 minutes and supernatants were removed. Then cells were permeabilized by adding BD Phosflow Perm Buffer (catalogue no: 558052) and incubated for 30 minutes on ice. Then the cells were washed twice by re-suspending them in BD Pharminogen Stain Buffer (catalogue no: 554656) and centrifuged at 300g for 5 minutes each time. Afterwards, cells were stained by Alexa Fluor 647 anti-NF-ï«B p65 (BD-catalogue no: 558422) or its matched negative isotype control (BD-catalogue no: 557903). Cells were then incubated at room temperature for 30 minutes in the dark and washed once with BD pharminogen stain buffer. Thereafter, cells were analysed by LSRII flow cytometer as described previously.
Figure 6: Overview of the Cytometric Bead Array procedure. The CBA beads are incubated with the sample and fluorescent antibodies, a set of standards is prepared that enables protein measuring between 10 and 2500 pg/ml. soluble analytes are analysed by FACS Array Flow Cytometry. MFI: median fluorescence intensity.
Figure 7: NF-ĸB activation pathway. The TLR2/4 ligands induce phosphorylation of the IĸB kinase α (IKBα). Once phosphorylated, the p50-p65: NF-ĸB is translocated to the nucleus where it can induce expression of inflammatory cytokines.
3.2.6 Isolation and Culture of Human Umbilical Vein Endothelial Cells (HUVECs):
Endothelial cells were isolated from freshly collected umbilical cords using standard published techniques used regularly in the Department of Cardiovascular Science, Medical School in the University of Sheffield (60, 61).
3.2.6.1 Buffer preparation:
Cord Collection Medium
5ml of penicillin-streptomycin solution, 5ml of amphotericin and 6ml of sodium bicarbonate (7.5%) were added to 500ml of 1X MEM. Note: fresh medium was prepared for each cord collection.
M199 Medium
50ml of 10X M199 was diluted in 430ml of distilled water, and 5ml of penicillin-streptomycin (10,000 IU/ml), 5ml l-glutamine and 10ml of sodium bicarbonate, was added.
Complete Growth Medium
M199 was supplemented with 10% newborn calf serum, 10% fetal calf serum, 20μg/ml of endothelial growth supplement and 90μg/ml of heparin.
0.1% (w/v) Collagenase solution
For each umbilical cord 0.02g of collagenase from Clostridium histolyticum was dissolved in 20ml of M199. The solution was passed through a syringe filter (gauge is 0.2ïm) into a sterile tube. Collagenase solution was prepared fresh or used within 2 days if stored at 4â°C.
Gelatin Coated Flasks
0.2% (w/v) gelatin solution in distilled water was used to coat tissue culture flasks. Sufficient volume was added to cover the surface and left to coat the flasks for a minimum of 30 minutes before use.
3.2.6.2 Procedure:
Umbilical cords were collected in sterile containers, containing 100ml of cord collection media, from the labour word at Jessop Wing Hospital and stored at 4â°C until used.
Working under a Class II hood, all surfaces were wiped with alcohol wipes and a clean Pyrex dish was used for umbilical cord manipulation.
Any damaged umbilical cords with holes/tears or those less than 10cm were discarded. The clamped ends and traumatised regions of cord were cut off.
Excess blood was removed by gently massaging the umbilical cord along its length.
The umbilical vein was cannulated with a 14G plastic cannula and tightly secured with string.
The umbilical vein was flushed through by perfusing with 30-40ml of medium M199 using 50ml syringe. This was continued until the vein was free of remaining blood.
The free end of umbilical cord was secured with a surgical clamp and an injection bung was placed over the cannula. 0.1% collagenase solution was injected into the vein using 20ml syringe and then the cord was incubated at room temperature for 10 minutes.
The digested layer of endothelial cells was collected into a sterile 50ml tube. The vein was flushed with 10-15ml of medium M199 to collect any partially dislodged cells. The cell suspension was centrifuged at 1000RPM for 5 minutes at room temperature and the supernatant was carefully removed.
Pelleted cells were resuspended in 5ml of complete growth media and transferred to gelatin coated tissue culture flasks. Endothelial cells were then incubated at 37â°C, 5% COâ‚‚.
24 hours after isolation, the medium was removed and the adherent endothelial cell layer was washed once with PBS, to remove any non-adherent cells. Endothelial cells were re-fed with complete growth media and thereafter, fed every two days until they reached confluence.
Typically endothelial cells reach confluence within 8-10 days. At confluence, cells were washed once with PBS and incubated with a small volume (1-2ml) of 0.1% trypsin / 0.02% EDTA in PBS, until the cells rounded up and detached. Endothelial cells were collected in a sterile 50ml tube containing 5-10ml of complete growth medium. The flask was washed with an additional 1-2ml of trypsin/EDTA and was added to the cell suspension.
Cells were centrifuged for 5 minutes at 1000RPM and the supernatant was carefully removed. Endothelial cells were resuspended in complete growth medium and were split in a 1:3 or 1:4 ratio onto gelatin-coated flasks or plates.
3.2.7 Statistical Analysis:
Results are illustrated as mean ± standard error of the mean (SEM). 2 sample t-tests were performed to determine any significant differences between study groups. P values < 0.05 were considered to be significant.
4. Results:
4.1. Monocyte Subpopulations:
Changes in the proportion of the peripheral blood monocyte subpopulations from pre-eclampsia, normal pregnant and non-pregnant women are presented in the Figures 8-11 below:
Figure 9: The proportion of CD14+CD16+ monocytes, in non-pregnant, normal pregnant and pre-eclamptic women. Values are illustrated as mean ±SEM.
Figure 10: The proportion of CD14highCD16- monocytes in non-pregnant, normal pregnant and pre-eclamptic women. Values are illustrated as mean ±SEM.
Figure 11: The proportion of CD14highCD16+ monocytes in non pregnant, normal pregnant and pre-eclamptic women. Values are illustrated as mean ±SEM.
Figure 9 shows that the proportion of the CD14+CD16+ monocytes in preeclampsia was significantly higher than in normal pregnant and non-pregnant women (P<0.001). However this monocyte subtype was significantly lower in normal pregnancy compared to non-pregnant women (P<0.001).
On the other hand, the proportion of the CD14highCD16- monocytes was highly increased in normal pregnant women compared to non-pregnant women and those with PE (P<0.001) as shown in Figure 10. In contrast it was significantly lower in PE than non-pregnant women (P<0.05).
The histograms in Figure 11, show that the proportion of CD14highCD16+ was lower in both PE and normal pregnant women compared to non pregnant but this difference did not achieve statistical significance (P = 0).
4.2. Cell Surface Expression of TLR4, TLR2 and HLA-DR:
Our results from the LSR II flow cytometry method showed significant differences in TLR2, TLR4 and HLA-DR expression on peripheral blood monocytes in samples from non-pregnant, pregnant and pre-eclamptic women as shown in the Figures and sections below.
From the graphs depicted in Figure 12A below, it can be seen that the mean percentage of the peripheral blood monocytes expressing TLR4 was higher in PE patients compared to normal pregnant and non-pregnant women (P<0.001). Moreover it was significantly higher in normal pregnancy than non-pregnant women (P<0.001). The mean percentage of the peripheral blood monocytes expressing TLR2 was greater in preeclampsia compared to normal pregnancy (P<0.05) and non pregnancy controls (P<0.001) (Figure 12-B). The mean percentage of peripheral blood monocytes expressing HLA-DR was significantly lower in PE than normal pregnancy (P<0.001), but higher in PE compared to non pregnant control women (P<0.05) (Figure 12C).
A B
C
Figure 12A-C: changes of the cell surface expression of TLR4 and TLR2 and HLA-DR on the peripheral blood monocytes from non-pregnant, normal pregnant (NP) and pre-eclampsia (PE) women. Values are shown as mean ±SEM.
Furthermore, the expression of TLR4, TLR2 and HLA-DR on the monocyte subpopulations has been identified in the flow cytometry histograms as shown in the Figures below:
Figure 13: Flow cytometry histograms of TLR4 expression on total MOs, CD14highCD16- monocytes, CD14highCD16+ monocytes and CD14+CD16+ monocytes from different study groups. The vertical axis showing cell counts, with horizontal axis representing fluorescence on a logarithmic scale.
Figure 14: Changes in the TLR4 expression on the monocyte subpopulations from study groups. Results are illustrated as mean ±SEM.
Our results demonstrated that the TLR4 expression in the three different monocyte subtypes studied differed for pre-eclamptic as well as for normotensive women but did not differ in nonpregnant women. In the pregnant subjects there was an increasing expression of TLR4 as the monocyte subtypes expressing more inflammation markers (Figure14 above). However no significant differences between TLR4 expression on the different monocyte subpopulations except in normal pregnancy.
Figure 15: Flow cytometry histograms of TLR2 expression on total MOs, CD14highCD16- monocytes, CD14highCD16+ monocytes and CD14+CD16+ monocytes from different study groups. The vertical axis showing cell counts, with horizontal axis representing fluorescence on a logarithmic scale.
Figure 16: Changes in the TLR2 expression on the monocyte subpopulations from study groups. Results are illustrated as mean ±SEM.
In all study groups, the mean percentage of CD14+CD16+ monocytes that expressed TLR2 was higher compared to the CD14highCD16- and CD14highCD16+ monocytes, as shown in graph above (Figure 16). This result achieved statistical significance for samples from normal pregnant women only (P < 0.05).
Figure 17: Flow cytometry histograms of HLA-DR expression on total MOs, CD14highCD16- monocytes, CD14highCD16+ monocytes and CD14+CD16+ monocytes from different study groups. The vertical axis showing cell counts, with horizontal axis representing fluorescence on a logarithmic scale.
Figure 18: Changes in the HLA-DR expression on the monocyte subpopulations from study groups. Results are illustrated as mean ±SEM.
The graph above (Figure 18) shows that the mean percentage of CD14highCD16+ monocytes which expressed HLA-DR expression was greater than the CD14highCD16- and CD14+CD16+ monocytes for samples from women with pre-eclampsia.
4.3 Tie-2 Compensation:
We observed a problem with compensation in our experiments studying Tie2 expression in the LSRII flow cytometry machine. The red 660/20 laser (APC-Tie2 antibody) was highly overlapped with the red 780 laser (APC-Alexa Fluor 780-CD14 antibody). The next control step was done, to solve this problem. After the PBMCs were isolated from fresh venous blood, cells were labelled with monoclonal antibodies as done before. In addition to CD14 conjugated to APC-Alexa Fluor 780 antibody, cells were stained with CD14 conjugated to Alaxa Flour 700 antibody obtained from biolegend (catalogue number: 325614). Thus the following tubes were added:
Eppendorf tube had PBMCs stained with 5μl of CD14 conjugated to Alexa Flour 700 antibody, 5μl of CD16 conjugated to e Flour450 antibody, 20μl of TLR2 conjugated to FITC antibody, 5μl of TLR4 conjugated to PE antibody, 20μl of Tie2 conjugated to APC antibody and 5μl of HLA-DR conjugated to Precyp5.5-cy5.5 antibody) as mentioned in company's sheets.
Eppendorf tube had cells mixed with 5μl of CD14 conjugated to Alexa Flour 700 antibody, 5μl of CD16 conjugated to e Flour450 antibody and isotype negative controls matched for TLR2, TLR4, HLA-DR and Tie2 monoclonal antibodies - the fluorescence minus one method.
Eppendorf tube had cells stained with 5μl of CD14 conjugated to APC-Alexa Fluor 780 antibody and 20μl of APC - Tie2 antibody.
Eppendorf tube had cells labelled with 5μl of CD14 conjugated to Alexa Flour 700 antibody and 20μl of APC -Tie2antibody.
Eppendorf tube had cells labelled with 5μl of CD14 conjugated to APC-Alexa Fluor 780 antibody and 20μl of APC - mouse IgG1-k isotype control.
Eppendorf tube had cells stained with 5μl of CD14 conjugated to Alexa Flour 700 antibody and 20μl of APC - mouse IgG1-k isotype control.
Then cells were analysed with LSRII flow cytometry machine as described previously. The FACS DIVA - BD Bioscience software showed that the red 660/20 laser (APC - Tie2 antibody) was incompatible with the red 780 laser (APC-Alexa Fluor 780 - CD14 antibody). On the other hand, it showed a perfect result for Tie2 expression with the red 730 laser (Alexa Flour 700 - CD14 antibody) as shown in Figure 19. Therefore, from this control step we plan to use Alexa Flour 700 - CD14 antibody instead of APC-Alexa Fluor 780 - CD14 antibody for the next series of samples from study groups.
4.4 Inflammatory Cytokines:
We tested the ability of various TLR ligands to modulate secretion of cytokines from monocyte donors. The following TLR2/4 ligands were used in the study: LPS; Lipopolysaccharides, PDG; Peptidoglycans, FC; Fibronectin, FG; Fibrinogen, HS; Heparan Sulfate Proteoglycan, HA; Hyaluronan. The Figures below presents the results obtained from the preliminary analysis of cytometric bead array.
A
B
Figure 20 A-B: IL-10 and TNF-α inflammatory cytokines, elaboration by monocytes from study groups stimulated with endogenous ligands of TLR4 and bacterial ligands of TLR2/TLR4. Abbreviations: LPS; Lipopolysaccharides, PG; Peptidoglycans, FC; Fibronectin, FG; Fibrinogen, HS; Heparan Sulfate Proteoglycan and HA; Hyaluronan. Results are shown as mean ± SEM.
As shown above from Figure 20A-B, in the pre-eclampsia women TNF-α, the pro inflammatory cytokine, and IL-10, an anti-inflammatory cytokine, were not changed with TLR4 endogenous ligands, but they were increased with bacterial ligands of TLR4/TLR2. On the other hand, TNF-α was increased with fibronectin, hyaluronan and bacterial ligands of TLR4/TLR2 whereas IL-10 did not change with endogenous ligands of TLR4 from normal pregnant women. In non pregnancy controls, TNF-α was increased with both endogenous ligands of TLR4 and bacterial ligands of TLR4/TLR2 but IL-10 did not changed.
4.5 NF-ï«B Experiment:
Unfortunately the results of NF-ĸB experiment were falsely positive for isotype controls than anti NF-ĸB antibody in LPS and PDG stimulated cells as seen in Figure: 21. This result was due to a low cell count used in this experiment and a lot of debris which falsely stained for the isotype control.
Figure 21: Flow cytometry dot plot and histograms showing the PBMCs stained with NF-ĸB and its matched isotype control after stimulation by LPS or PG. A showed the PBMC in the forward scatter (FSC) and side scatter (SSC) dot plot with monocytes gating. B & D histograms showed MOs falsely positive for isotype controls with LPS & PDG stimulation respectively. C & E histograms showed MOs stained with anti NF-KB antibodies and stimulated by LPS & PDG respectively.
5. Discussion:
5.1 Monocyte subpopulations:
Prior studies have been shown that the pre-eclampsia is characterised by increased production of inflammatory cytokines and abnormal activation of inflammatory cells such as monocytes (62, 63). Although there is strong evidence that monocytes are activated in pre-eclampsia, no published study has investigated the peripheral blood monocytes subpopulations from PE patients. Our hypothesis that the monocyte populations with an inflammatory phenotype will be expressed in a greater proportion in PE was borne out by our observations.
In the current study, the proportion of CD14+CD16+ monocytes was significantly increased in pre-eclampsia compared to normal pregnant and non pregnant women (P<0.001). This observation is consistent with the exaggerated systemic inflammatory response seen in PE and reported by other studies (63). Also it is consistent with other published studies which have reported an increase in CD14+CD16+ monocytes in infectious and inflammatory conditions such as rheumatoid arthritis, haemodialysis, atherosclerosis, Kawasaki disease, septic shock, and human immunodeficiency virus (HIV) infection (21, 64-69).
In contrast to our observation that the proportion of CD14+CD16+ monocytes was significantly increased in preeclampsia, the proportion of CD14highCD16- monocytes was significantly decreased in this condition compared to normal pregnancy (P<0.001). Also in normal pregnancy, the reduction in the CD14+CD16+ monocytes proportion appeared to be secondary to the increase in the proportion of the CD14highCD16- monocytes.
Zeigler-Heitbrock and his colleagues have suggested that human monocytes subpopulations represent different stages of monocyte maturation. They suggest that the CD14+CD16+ monocytes are the more mature phenotype with the characteristics of tissue macrophages (18, 70). However there remains considerable debate regarding the implications of the finding of these subtypes. Additionally, in vitro maturation studies have shown that CD14+CD16+ monocytes can be also be differentiated from the more common CD14highCD16- monocytes (70).
Therefore, it is plausible that the increased immature CD14highCD16- monocytes from normal pregnancy may differentiate into the apparently more mature phenotype CD14+CD16+ cells at end stages of pregnancy. This may contribute to the mild inflammatory response present with normal pregnancy. It is plausible that his maturation process of monocytes could occur at earlier stages in pre-eclamptic women, resulting in the exaggerated systemic inflammatory response seen in this condition. Furthermore, a trigger factor that is expressed in pre-eclampsia might be inducing early monocyte maturation. Further studies of normotensive pregnant women at the stage of pre-labour stage may be required to refute or confirm this hypothesis.
We had hypothesised that endogenous ligands of TLR4 may be implicated in the pathogenesis of abnormal vascular reactivity and dysfunction in pre-eclampsia. However our preliminary observations stimulating monocytes with some of the well described ligands did not demonstrate significant inflammatory cytokine expression, neither did they seem to increase the percentage of monocytes with an inflammatory phenotype. If endogenous ligands contribute in the way we postulated, this is likely to be partial and may be well involved endogenous ligands which we have not studied in these experiments.
5.2 Cell Surface Expression of TLR4, TLR2 and HLA-DR:
As mentioned in the literature review, all TLRs are expressed at the maternal-fetal interface (46). TLR4 has been demonstrated to activate innate immune system in pregnancy. Moreover, it has the ability to activate inflammatory cytokines profile and promote angiogenesis via VEGF. VEGF in turn is also known to be involved in preventing the endothelial dysfunction seen in pre-eclampsia (45, 46).
It was hypothesized that TLR4 activity may play a key role in pathogenesis of pre-eclampsia through one or more of its endogenous ligands (e.g. heparan sulphate, fibrinogen, fibronectin and hyaluronan), produced by the ischaemic placenta. To the best of our knowledge, this is the first study to investigate the expression of TLR2 and TLR4 on human peripheral blood monocytes. Additionally no studies have investigated endogenous TLR4 ligands expression in women with PE.
The results of this study show that the TLR4 and TLR2 expression on the peripheral blood monocytes was significantly increased in PE compared to normal pregnant and non-pregnant women. Up regulation of TLR2 and TLR4 expression on the peripheral blood monocytes from PE women suggest a role for this pattern recognition receptor in the exaggerated systemic inflammatory response seen in this condition but it is unclear whether this occurs as a cause or consequence of PE. Also our preliminary results show that the TLR4 and TLR2 expression was increased on the CD14+CD16+ monocytes. However the observed difference in TLR4 and TLR2 expression on monocyte subpopulations in our study groups did not achieve statistical significance, perhaps owing to the limited sample size. Studies are ongoing to increase the number of samples studied.
In contrast to earlier studies which demonstrated no significant changes of the HLA-DR expression on monocytes between pre-eclampsia and normal pregnancy (62). The current study found the expression of HLA-DR significantly lower on the peripheral blood monocytes from PE compared to normal pregnancy. The reason for this observation is as yet unclear.
5.3 Inflammatory Cytokines:
The present study was designed to specify which endogenous ligands can activate TLR4 at monocytes culture via determination of inflammatory cytokines with cytometric bead array method. As shown above from preliminary CBA results, TNF-α, the pro inflammatory cytokine, and IL-10, an anti-inflammatory cytokine, were not changed with TLR4 endogenous ligands, but they were increased with bacterial ligands of TLR4/TLR2, in pre-eclampsia. On the other hand, TNF-α expression by monocytes from normal pregnant women was increased with fibronectin, hyaluronan and bacterial ligands of TLR4/TLR2 whereas IL-10 expression did not change when treated with endogenous ligands of TLR4. For non-pregnant samples, TNF-α was increased with both endogenous ligands of TLR4 and bacterial ligands of TLR4/TLR2 but IL-10 did not changed. However, the detected changes were not significant. The same work is continuing to achieve a larger sample size for each of the study cohorts (PE, normal pregnant and non pregnant women).
5.4 NF-ĸB Experiment:
The purpose of the study of phosphorylated NF-ï«B p65 was to determine the induction of the intracellular signalling pathway by ligands of TLR4. The result of this experiment was falsely positive for isotype control rather than anti-NF-ĸB antibody in lipopolysaccharide-, and peptidoglycan-stimulated cells (Figure 21). This result was due to a low cell count used in this experiment and a lot of debris which falsely stained for the isotype control. Thus, future studies will employ a different method to identify phosphorylated intracellular NF-ĸB, namely an ELISA for NF-ĸB p65 from SA Bioscience. It is highly quantitative and sensitive method which can be analysed phosphorylated NF-ĸB p65 within stimulated cells in 96-well culture plate. Furthermore, this method is published by recent cancer researches (71-73).
6. Further Work:
6.1 Serum Studies:
We will determine serum levels of endogenous ligands of TLR2 and TLR4 in the 3 groups of subjects we are studying. Serum has been frozen from these subjects (pre-eclamptic patients, normal pregnant women and non pregnant women) at -80Câ° for these further studies by radioimmunoassay and ELISA.
7.2 Human Umbilical Vein Endothelial Cell (HUVEC) Culture:
Currently, I am practising HUVECs isolation and culture experiment in the Cardiovascular Research Department at the Medical School of the University of Sheffield. When the technique is optimised, I am going to collect umbilical cord sections from PE and normal pregnancy women for further studies.
Umbilical Cord sections will be used to derive HUVEC cells for cultures using validated and published methodology as described above (60, 61). Our experiments will compare the effects of endogenous ligands of TLR4 and bacterial ligands of TLR2/TLR4 in the HUVEC cultures via using cytometric bead array. Real time PCR will be used to determine TLR2/4 expression in the HUVEC culture.
6.3 Co-culture Experiments:
We will investigate the interaction between monocytes and HUVEC. When endothelial cells undergo inflammatory activation, the increased expression of cell adhesion molecules promotes the adherence of monocytes. Adhesion molecule expression is induced by inflammatory cytokines such as IL-10 and TNF-α. We will determine whether TLR4 ligands contribute to the stimulation and elaboration by monocytes of proinflammatory cytokines which may then promote adhesion molecule expression by endothelial cells as a mechanism of the endothelial damage that epitomises pre-eclampsia. We will mount appropriate negative and positive control experiments to determine the cause of our observations (for example using antagonists of TLR4).
