Periodontally healthy, generalized periodontitis [stage III, grade B, (20)] and MIPP patients (cases with inderdental loss of attachment ≥5 mm at first molars and incisors and with radiographic evidence of vertical bone loss at the same sites) were recruited prior to periodontal treatment from the Periodontal Department, Hebrew University- Hadassah, Ein Karem between 2017-2019. Periodontal diagnosis was determined using radiographs and detailed periodontal clinical charts. Inclusion criteria included: patients 18y old or older, generally healthy and willing to participate in the study. Exclusion criteria included: smoking, pregnancy and lactating, antibiotic consumption within 6 months of participation, drug/alcohol abuse, and chronic consumption of any drug.

Cells were obtained on the morning of the experiments from whole blood in standard Lithium Heparin vacutainer tubes. Neutrophils were isolated from whole blood by density gradient centrifugation (Polymorphprep, Axis-shield, Oslo, Norway), according to the manufacturer’s instructions. In brief, blood was layered on an equal volume of Polymorphprep in a 15ml tube and centrifuged for 30min, 500g at room temperature (RT). The lower cellular white band contained PMNs and was separated, eluted with red blood cells lysis buffer (Invitrogen) and centrifuged for 10 min, 400g at RT.

Neutrophils were incubated with the different bacteria (A. actinomycetemcomitans, JP2 or Pg) in a ten-fold multiplicity of infection (MOI 10) in a 24 well plate (5 × 10⁵ cells/well) for 90min at 37°C. After incubation samples were transferred to collection tubes and centrifuged (7min/25⁰c/1500 rpm). The supernatants were then decanted and the cells were resuspended in 100 µL PBS and 5 µL propidium iodide (PI, Sigma) was added to each sample. After 15 min incubation in the dark, the samples were analyzed by flow cytometry (BD Accuri™ C6 Cytometer, San Jose, CA, USA). The experiments were performed in triplicate for 8 different patients with MIPP and 8 healthy controls.

Neutrophils (5 × 10⁴ cells/well) from healthy controls and MIPP patients were incubated with A. actinomycetemcomitans or JP2 at a MOI-10 ratio in a 96-well black plate (PerkinElmer, Waltham, MA). HOCl (0.0006%) served as positive control. DCFH-DA (7′-dichlorofluorescein diacetate, Sigma–Aldrich, Israel) was added to each well as a marker for ROS. Immediately after addition of the bacteria, the plate was transferred to a florescence plate reader (infinite plate reader; Tecan, Männedorf, Switzerland) at 37˚C with readings (490nm excitation and 535nm emission) every 5 min for a total 150min.

The experiments were repeated in triplicate for 8 MIPP patients and 8 healthy controls.

Neutrophils were incubated with the bacteria (5 × 10⁵ cells/well; MOI-10) in 24 well plates. After 90 min, samples from all the groups were seeded on 5% sheep blood agar (Hay Labs, Rehovot, Israel) in eight 10-fold dilutions. The plates were incubated at 37°C for one week in a 5% CO₂ chamber, after which colonies were counted and colony forming units (CFU) were calculated.

The experiments were repeated in triplicate for 8 patients with MIPP and 8 healthy controls.

Similarly to the above assays, neutrophils were incubated with the different bacteria for 3 hrs (verses un-exposed naïve controls) and centrifuged at 1200RPM at RT for 20min. The supernatants were collected and stored for protein-based analysis and the pellets were eluted in Trisol (Invitrogen) for RT-PCR analysis. All samples were stored at -80⁰ C until analysis.

A human cytokine array kit, hematopoietic receptor and non-hematopoietic receptor array kits (all from R&D systems, Minneapolis, MN, USA), were used according to the manufacturer’s instructions.

For the cytokinome assay, pooled samples from all the cases (equal volumes) were mixed by groups (MIPP and periodontally healthy), separately and used according to the manufacturer’s instruction.

For the neutrophil receptor assay, HL60 cells (Human acute promyelocytic leukemia cell line 60) were lysed in EDTA lysis buffer (10mM Tris-HCl, 2mM EDTA, 1% SDS in PBS) supplemented with protease inhibitors (Sigma), centrifuged at 25,000g at 4°C for 10 min and incubated with JP2 (at MOI 10) for 20min at RT. Then the lysates with the JP2 and adhering proteins were centrifuged at 4000 RPM for 20 min at RT and resuspended in stripping buffer ​(15 g/l glycine, 1 g/l SDS, 1% Tween 20 in DDW) for 10 min at RT to detach the proteins. After 0.22µm filtration to remove the JP2, the eluted proteins were used for the arrays.

“Image lab” (BIO-RAD, Hercules, CA, USA) software was used for semi-quantification of each protein by dot size and intensity.

Four cytokines, IL8, IL10, TNFα, mCSF (all from R&D Systems), chosen according to the results of the cytokine profiler, were further validated using ELISA. Samples were analyzed as individual values for each patient. In brief, ELISA plates were coated with capture antibody overnight at RT. The plates were then washed with wash buffer (0.05% tween diluted 20 x in PBS) and blocked with 1% bovine serum albumin-PBS) for 1 hr at RT. The plates were then washed, and samples and standards were added for a further 2 hrs incubation at RT. Secondary antibodies, followed by streptavidin-HRP and substrate solution (TMB, R&D systems) were then used and the reaction was terminated by the addition of 2N H₂SO₄. The results were analyzed with a spectrophotometer (ELISA reader) at 540nm wavelength.

Similarly to the ELISA assay, samples were analyzed for each patient. RNA was extracted from the cell samples with the TRIzol™ Reagent (Invitrogen), according to the manufacturer’s instructions. The purified RNAs were translated into cDNA with a QuantaBio kit (Beverly, MA, USA(according to the manufacturer’s instructions and the synthesized cDNAs were amplified with SYBR Green real-time PCR and specific primers for the genes of interest (ABI Prism 7300 System, Applied Biosystems, Foster City, CA, USA). The 20 μl reaction volume contained 10 μl PerfeCTa SYBR Green FastMix (Quantabio, Beverly, MA, USA), 500 nmol/L each of the forward and reverse primesr, and 5 μl cDNA. The thermal profile for SYBR Green real-time RT-PCR was 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 58°C for 1 min.

The sequences of each of the primers and their sources are: IL8- F: ATGACTTCCAAGCTGGCCGTGGCT, R: TCTCAGCCCTCTTCAAAAACTTCTC (24); IL10: F: GCCTAACATGCTTCGAGATC, R: TGATGTCTGGG CTTGGTT (25); mCSF: F: 5′-TAGCCACATGATTGGGAGTG-3′, R: 5′-TATCTCTGAAGCGCATGGTG-3′ (26); TNFa: F: GAAAGCATGATCCGGGACGTG, R: GATGGCAGAGAGGAGGTTGAC (27); GAPDH (as house-keeping gene(: F: AGTTGGGATAGGGCCTCTCTT, R: TCCCACTCTTCCACCTTCGA (28). The results were analyzed according to the ΔΔCt protocol.

HL60 cells (ATCC) were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM) supplemented with 20% fetal bovine serum. For differentiation into neutrophil-like cells, the cell medium was supplemented with 1.4% DMSO for 4 days, and they are thereafter referred to as neutrophils (29).

A 0.2% trypan blue solution was added to HL60 cells incubated with JP2, and after 15 min on ice, the cells were washed with PBS and analyzed by flow cytometry. When stated, 4% paraformaldehyde in PBS was used to fix JP2 and/or HL60 on ice for 20 min before JP2 incubation with the neutrophils. When stated, a stripping buffer was added to the HL60 cells incubated with JP2, which were then analyzed by flow cytometry.

HL60 cells were cultured in IMDM supplemented with 20% fetal bovine serum. For differentiation into neutrophils, the cell medium was supplemented with 1.4% DMSO for 4 days, and the cells are thereafter referred to as neutrophils (29). The neutrophils, at 5X10⁵cells/100uL, were incubated with the following blocking antibodies: TREM-1, CD18, Galectin-3, Siglec-5, DPP4/CD26, CD40, IL-6 R, chitinase 3 like 1, L-selectin, CD93 and CXCL/IL-8 for 1 hr at 37°C and 5% CO₂. They were then incubated with JP2 for 90min, stained with PI for 15 min on ice and analyzed with flow cytometry.

When stated, primary neutrophils from MIPP patients, instead of the differentiated HL60 cells, were isolated with Polymorphprep (Axis-shield).

Assays of neutrophil cytotoxicity, survival and ROS were performed in triplicate for eight subjects for each of the healthy and MIPP groups.

The data were analyzed with a statistical software package (SigmaStat; Jandel Scientific, San Rafael, CA, USA). One-way repeated measure analysis of variance (RM ANOVA) was applied to test the significance of the differences between the treated groups. If the results were significant, inter-group differences were tested for significance with Student’s t test and the Bonferroni correction for multiple testing.

First, primary neutrophils from periodontally healthy, generalized periodontitis (stage III grade B) and MIPP subjects (n=3 for each group) were each incubated with Porphyromonas gingivalis (Pg), non-pathogenic (A. actinomycetemcomitans) or the MIPP pathogen A. actinomycetemcomitans clone (JP2). Following 90 min incubation, cells were collected for analysis of necrosis and bacterial clearance.

The results showed that JP2 and A. actinomycetemcomitans survival was similar in all the neutrophil groups ( Figure 1A ). Pg showed reduced survival in both periodontitis groups compared with that of the periodontally healthy group ( Figure 1A ). Greater resistance to neutrophil necrosis was observed in the MIPP group compared with that in the periodontally healthy and generalized periodontitis groups ( Figure 1B , p<0.0001).

Based on the above results, the next experiment was conducted with MIPP and healthy control neutrophils (n=10 per group), and with A. actinomycetemcomitans and JP2 as infectious agents. The results were similar to those described in Figure 1 , for both necrosis and bacterial survival ( Figure 2A ), suggesting a true trend in neutrophil behavior.

Next, we examined the reactivity of neutrophils to infection, by measuring intracellular reactive oxidative species (ROS). The only significant differences were between MIPP and healthy control neutrophils in response to JP2: MIPP neutrophils exhibited significantly lower levels of ROS production compared with that of the healthy neutrophils, both as cumulative ROS (as the area under the curve values, AUC) and as endpoint after 90 min bacterial infection ( Figure 2B , p<0.001).

Secretome analysis was performed to corroborate the above findings and single out the unique cytokine pattern of MIPP versus periodontally healthy neutrophils. To that end, a profiler assay was used to map the neutrophil cytokine secretome (101 cytokines and chemokines, Figure 3A and Supplementary Figure 1 ).

MIPP naïve and JP2 infected neutrophils expressed reduced inflammatory levels of most of the tested cytokines compared with those of periodontally healthy neutrophils ( Figure 3B ). Chemokine expression levels also showed reduced expression in MIPP neutrophils compared with those in periodontally healthy neutrophils ( Figure 3C ).

As the profiler assay is a semi-quantitative assay, we further quantified selected cytokines (IL-8, IL-10, M-CSF, and TNFα) using ELISA and RT-PCR. The ELISA results (in pg/ml) showed significantly reduced levels of the above cytokines in MIPP neutrophils ( Figure 3D ), which were corroborated by qRT-PCR ( Figure 3E ).

Next, we examined the nature of the interaction between JP2 and neutrophils.

Using the human cell line HL60 and JP2 for this purpose, we examined whether the bacterium adheres to the neutrophils and if so, whether the interaction occurs extracellularly or intracellularly. dHL60 cells were incubated with fluorescein isothiocyanate- (FITC) stained JP2 (29). Trypan blue was used to quench extracellular FITC staining. Gating strategy is presented in Supplementary Figure 3 . The results showed that the dHL60 were all positive for FITC staining, and the FITC-stained cells were also positive for necrosis (PI, Figure 4A ). Trypan blue abrogated the FITC signal on the dHL60, suggesting that the bacteria adhere to the cells extracellularly.

We next examined whether the interaction between JP2 and the neutrophils is protein-mediated. We denaturized the surface proteins (by fixation with paraformaldehyde) on the neutrophils and/or JP2 before the infection. The results showed that denaturation of the proteins on neutrophils abrogated JP2 attachment, whereas the same process on JP2 did not ( Figure 4B ). Furthermore, stripping off the inter-protein association (with a stripping buffer) detached the JP2 from the neutrophils ( Figure 4C ).

Next, a wide receptor profile was made in order to single out potential receptors that interact with JP2. A lysate of neutrophils (dHL60) was incubated with JP2, and then the JP2s (with the attached proteins from the cell lysate) were collected, the inter-protein association was stripped, and the JP2 was removed by filtration. The extracted proteins showed specific targets compared with the entire neutrophil lysate ( Figure 5A ). Also, the specific protein levels were higher than the whole extraction samples ( Figure 5B ) (for those most enriched by the JP2 immunoprecipitation see Table 1 and Figure 5C ).

The enriched receptors (TREM-1, CD18, Galectin-3, Siglec-5, DPP4/CD26, CD40, IL-6 R, chitinase 3-like 1, L-selectin, CD93 and CXCL/IL-8) were then functionally tested by blocking each one before JP2-dHL60 interaction and cell necrosis was examined. Of the tested receptors, only blocking of CD18 rescued the neutrophils from necrosis upon infection with JP2 ( Figure 6A , P<0.05).

Finally, in order to validate that CD18 indeed has a role in the necrosis resistance of the MIPP neutrophils, CD18 on primary neutrophils from MIPP cases was blocked and then challenged with JP2 and neutrophil necrosis was examined. Indeed, in comparison with the necrosis rescue effect in healthy neutrophils by CD18 blocking ( Figure 6A ), the same procedure did not affect MIPP neutrophil necrosis, showing that the CD18 in these cells is functionally different from that in periodontally healthy neutrophils ( Figure 6B ).