Alpha-1 antitrypsin inhibits fractalkine-mediated monocyte-lung endothelial cell interactions

Abstract

Chronic obstructive pulmonary disease (COPD) is characterized by nonresolving inflammation fueled by breach in the endothelial barrier and leukocyte recruitment into the airspaces. Among the ligand-receptor axes that control leukocyte recruitment, the full-length fractalkine ligand (CX3CL1)-receptor (CX3CR1) ensures homeostatic endothelial-leukocyte interactions. Cigarette smoke (CS) exposure and respiratory pathogens increase expression of endothelial sheddases, such as a-disintegrin-and-metalloproteinase-domain 17 (ADAM17, TACE), inhibited by the anti-protease α-1 antitrypsin (AAT). In the systemic endothelium, TACE cleaves CX3CL1 to release soluble CX3CL1 (sCX3CL1). During CS exposure, it is not known whether AAT inhibits sCX3CL1 shedding and CX3CR1⁺ leukocyte transendothelial migration across lung microvasculature. We investigated the mechanism of sCX3CL1 shedding, its role in endothelial-monocyte interactions, and AAT effect on these interactions during acute inflammation. We used two, CS and lipopolysaccharide (LPS) models of acute inflammation in transgenic Cx3cr1^gfp/gfp mice and primary human endothelial cells and monocytes to study sCX3CL1-mediated CX3CR1⁺ monocyte adhesion and migration. We measured sCX3CL1 levels in plasma and bronchoalveolar lavage (BALF) of individuals with COPD. Both sCX3CL1 shedding and CX3CR1⁺ monocytes transendothelial migration were triggered by LPS and CS exposure in mice, and were significantly attenuated by AAT. The inhibition of monocyte-endothelial adhesion and migration by AAT was TACE-dependent. Compared with healthy controls, sCX3CL1 levels were increased in plasma and BALF of individuals with COPD, and were associated with clinical parameters of emphysema. Our results indicate that inhibition of sCX3CL1 as well as AAT augmentation may be effective approaches to decrease excessive monocyte lung recruitment during acute and chronic inflammatory states.

NEW & NOTEWORTHY Our novel findings that AAT and other inhibitors of TACE, the sheddase that controls full-length fractalkine (CX3CL1) endothelial expression, may provide fine-tuning of the CX3CL1-CX3CR1 axis specifically involved in endothelial-monocyte cross talk and leukocyte recruitment to the alveolar space, suggests that AAT and inhibitors of sCX3CL1 signaling may be harnessed to reduce lung inflammation.

INTRODUCTION

One of the chronic obstructive pulmonary disease (COPD) hallmarks is nonresolving inflammation, characterized by a breach in endothelial mechanical barriers and accumulation of leukocytes into the airspaces (1). The leukocyte-endothelium cross talk is essential for initiation and resolution of inflammation. In homeostasis, ligand-receptor pairs like P-L selectins or CX3CL1-CX3CR1 are constitutively expressed on endothelium and leukocytes, respectively (2). The ligands bind weakly to the receptors, allowing leukocytes to “patrol” the endothelium (3) and perform “housekeeping” phagocytosis of membrane’s debris and microparticles (4) or postinjury tissue repair (5). Inflammation, cigarette smoke (CS) exposure, or shear stress shifts the ligand-receptor repertoire to tighter binding pairs like E-selectin-CD11b/CD18 or -α4β1 that facilitate leukocyte adherence and transendothelial migration (6–8).

The CX3CL1-CX3CR1 axis is linked to recruitment, tethering, and accumulation of perivascular Ly6G^-Ly6C^lowCX3CR1⁺ monocytes in the dermis, kidney, lung, and brain (4, 9–11). The full-length CX3CL1-CX3CR1⁺ cross talk orchestrated monocyte crawling, endothelium patrolling (4), and resolution of injury after excisional skin-, lupus nephritis-, ischemia-reperfusion, or traumatic brain injury (10–13). Contrarily, sCX3CL1 is shed off the endothelium following extracellular proteolysis of full-length CX3CL1 by ADAM10 in homeostasis or ADAM17 (TACE) during inflammation (14). sCX3CL1 fuels microglia survival and proliferation (15), diminishes CX3CR1 expression on CD8⁺ T cell (16), acts as an indirect chemoattractant for leukocyte recruitment by increasing integrin α4β1 and αvβ3 receptors under shear stress conditions (17) and in CS-exposed lungs (9), and increases proinflammatory TNF-α and IL-6 production by lung interstitial CX3CR1⁺ monocytes (18).

Alpha-1 antitrypsin deficiency (AATD) is a major genetic risk factor in COPD particularly when combined with environmental CS exposure (19). AAT is a serine protease inhibitor with highest activity not only against neutrophil elastase (20) but also against a spectrum of proteases involved in lung inflammation, including TACE (21–23). Individuals with COPD and AATD present with either functionally altered (e.g., oxidized in COPD) or insufficient AAT (due to intrahepatic polymerization of mutated Z protein in AATD) (24). In these individuals, TACE activity in the lung endothelium is unchecked, because the endothelium is unable to synthesize local, intracellular AAT (25). The systemic and lung endothelium is dependent on the extracellular, circulating AAT, altered in COPD by CS-induced oxidation, or in AATD by intrahepatic polymerization and low circulating AAT levels (26).

Tissue inhibitors of metalloproteinases (TIMPs) are known endogenous TACE inhibitors showing therapeutic benefit in the treatment of inflammation and cancer (27, 28). Whether AAT has an effect on TACE activity, CX3CL1-CX3CR1 axis, and leukocyte-endothelial cross talk during acute lung inflammation is not known. Aerosolized ATT administered to rats restored in situ airway macrophages phagocytosis (22) and increased survival after Pseudomonas aeruginosa pneumonia by decreasing lung inflammation in AAT^+/+ transgenic mice (29).

In this study, we used gain- and loss-of-function, in vivo and ex vivo approaches to investigate the inhibitory effect of AAT on CX3CL1-CX3CR1 axis and monocyte-endothelium cross talk during LPS and CS exposures. Our results show that prophylaxis and treatment with AAT significantly decreased sCX3CL1 shedding, increased full-length CX3CL1 endothelial expression, and decreased CX3CR1⁺ monocyte adhesion and transendothelial migration into the airspaces. These effects were posttranslational, associated with TACE-inhibition, as AAT treatment had no effect on Cx3cl1 transcription and protein expression.

Some of the results of these studies have been previously reported in the form of abstracts (30).

MATERIALS AND METHODS

Reagents

Chemicals and reagents, including P. aeruginosa-derived LPS were purchased from Sigma Aldrich (St. Louis, MO), unless otherwise stated. Phorbol 12-myristate 13-acetate (PMA) was from Abcam (Cambridge, MA). TNF-α Protease Inhibitor-1 (TAPI₁) and GM6001 inhibitor were from ESD Millipore (Billerica, MA). Human CX3CL1 chemokine domain blocking antibody (Clone No. 81506) was from R&D Systems. Anti-human CX3CL1 (Clone No. 51637), anti-human CX3CR1 (2A9-1), anti-human CD62E (clone 13D5), and anti-HA tag (ab9110) antibodies were from R&D Systems, BioLegend, and Abcam, respectively. Human AAT (Prolastin-C) was from Grifols (Los Angeles, CA).

Animals

All experiments were approved and were conducted in compliance with National Jewish Health (NJH) Institutional Animal Care and Use Committee (IACUC) guidelines. Six- to eight-week-old B6.129P2(Cg)-Cx3cr1^tm1Litt/J or CX3CR1^gfp/gfp (KO) and C57B/l6 mice (Jackson Laboratories) were bred to generate CX3CR1^gfp/+ (HET) and CX3CR1^+/+ (WT) littermates. The CX3CR1^gfp/gfp knock-in/knockout mice express enhanced green fluorescent protein (EGFP) at the cx3cl1 locus that allows tracking of the GFP⁺ CX3CR1-KO monocyte, dendritic, and natural killer (NK) cells. As per Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines, an equal number of male and female CX3CR1^gfp/gfp KO and WT animals were used for LPS and CS exposure experiments.

LPS administration.

Mice, 8–12 wk old, received one dose of PBS or LPS (50 µg in 50 µL) via oropharyngeal aspiration. After LPS administration, animals (n = 4–8 mice/group) were allowed to recover in the surgical area. The animals were euthanized 72 h (day 3) post-LPS administration when plasma, BAL, and lungs were collected (31).

Acute CS exposure.

Seven-to-ten mice, 6–8 wk old, were exposed in a total body exposure chamber (TE-10Z, Teague Enterprises, Woodland, CA) for 5 h/day, 5 days/wk, for 1 mo or to CS or to ambient air (32). CS was generated from research-grade cigarettes 3R4F from the Kentucky Tobacco Research and Development Center (Lexington, KY). The cigarettes were smoked at a rate that achieved 90–120 ng/m³ microparticles in the exposure chamber.

AAT administration.

AAT, 20 mg/kg body wt or equal volume PBS was given intraperitoneally 1 day before LPS administration or once weekly, during the last 3 wk of CS exposure.

BAL fluid, plasma, and lung tissue harvest was performed as previously described (33). Briefly, mice were anesthetized with isoflurane and euthanized by bilateral pneumothorax. Plasma was collected via cardiac puncture in syringes prefilled with 100 μL of citrate buffer. Lungs were perfused via the heart with 10 mL of PBS. After neck dissection and midline tracheostomy, the BAL needle was secured in place and 1 mL of cold sterile saline containing 2 mM EDTA was instilled and recovered three times. Cell suspension from all three BAL aliquots was combined and washed twice with a wash buffer (PBS, 9% FBS, 0.5 mM EDTA). After the last wash, cells were spun at 300 g for 8 min and processed for flow cytometry.

Flow cytometry.

For total CD45⁺ counts, 100 μL aliquot of the BAL fluid was blocked with CD16/CD32 (clone 93, eBioscience, ThermoFisher), stained with CD45 (30-F11, BD, Franklin Lakes, NJ), and mixed with 123count eBeads (eBioscience). Cells used for total cell counts were stained and fixed without any centrifugation to avoid variability introduced by pelleting and aspirating. Using the absolute concentration of the counting beads added and the ratio of total CD45⁺ events to total bead events, the concentration of CD45⁺ cells was determined. BAL cell suspension was first blocked with CD16/CD32 (eBioscience), and stained with anti-mouse CD45 (30-F11, BD), Ly6G (1A8, BioLegend), Ly6C (HK1.4, BioLegend), CD64 (X54–5/7.1, BD), CD11c (N418, eBioscience), F4/80 (BM8, eBioscience), CD11b (M1/70, eBioscience), Siglec-F (E50–2440, BD), and CX3CR1 (SA011F11, BioLegend, only in wild-type mice) antibodies. Flow data, which included a minimum of 10,000 CD45⁺ events for each sample, was collected using LSR II cytometer (BD) and analyzed using FlowJo (34). The gating strategy to identify the recruited CX3CR1⁺ monocytes is detailed in Supplemental Fig. S1.

Cell Culture

Primary human lung microvascular endothelial cells (HLMVECs) from distinct donors (biological replicates) were purchased from Lonza (Walkersville, MD) and maintained in a complete culture medium consisting of EGM-2MV supplemented with growth factors and 5% FBS (CC-3202).

Primary human pulmonary artery endothelial cells (HPAECs) were purchased from Invitrogen (Carlsbad, CA) and maintained in a complete culture medium consisting of M200 media supplemented with Low Serum Growth Supplement (LSGS, Gibco). Only cells below passage 8 were used for experiments. HPAECs were transfected with N-, C-terminal, and mutated C-terminal human influenza hemagglutinin (HA)-tagged human Cx3cl1 ORF mammalian expression plasmids (Sino Biological Inc., Wayne, PA) using nucleofection kit (Lonza, VVPI-1001) and Amaxa Nucleofector (U-017 program) (35). Cx3cl1 gene site-directed mutagenesis at the putative TACE cleavage site was accomplished using QuikChangeII mutagenesis kit (Agilent Technologies). Primer sequences are listed in Table 1. Introduction of the deletions was verified by sequencing following PCR amplification and E. coli transformation. QIAGEN Midiprep kit was used to generate endotoxin-free constructs that were subsequently used for transfection of HPAEC cells.

Table 1.

Site-directed mutagenesis primer sequences

Construct Name	Primer ID	Sequence (5′ to 3′)
Sequence (5′ to 3′)	L324-F	GGACCCCCAGAGGGGCGTCCTTATC
	L324-R	GATAAGGACGCCCCTCTGGGGGTCC
ΔVal326(GTC)	V326-F	CAGAGGCTGGGCCTTATCACTCCTG
	V326-R	CAGGAGTGATAAGGCCCAGCCTCTG
ΔThr329(ACT)	T329-F	GCGTCCTTATCCCTGTCCCTGACGC
	T329-R	GCGTCAGGGACAGGGATAAGGACGC
V331-R	V326-F	CCTTATCACTCCTCCTGACGCCCAG
	V331-R	CTGGGCGTCAGGAGGAGTGATAAGG
ΔAla334(GCC)	A334-F	CTCCTGTCCCTGACCAGGCTGCCAC
	A334-R	GTGGCAGCCTGGTCAGGGACAGGAG

Human peripheral blood-derived monocytes (hPBDMs) were freshly isolated from healthy donor plasma, which was collected in the BD Vacutainer CPT system containing sodium citrate anticoagulant and blood separation media composed of a thixotropic polyester gel and a FICOLL Hypaque solution. After centrifugation (1,600 g, 20 min, 23°C), the gel portion of the medium formed a barrier separating the mononuclear cells and plasma. hPBDMs were subsequently negatively selected using Dynabeads Untouched Human Monocytes kit (Invitrogen) as per the manufacturer’s protocol. Purified hPBDM were maintained in FBS-free RPMI-1640 culture medium containing 1% Pen/Strep, 1% nonessential amino acids, 2% sodium pyruvate, and 20 mM HEPES.

Human monocytes cell line THP-1 (American Type Culture Collection, ATCC) cells were cultured in RPMI-1640 supplemented with 10% heat-inactivated FBS, 0.05 mM 2-mercaptoethanol, 100 U/mL penicillin, and 0.1 mg/mL streptomycin.

Endothelial cell treatments.

Serum-deprived (no FBS) EGM-2MV or M200 media was used to create conditions of AAT deprivation. FBS was used as the only source of AAT in cell culture conditions. TACE activators PMA (100 ng/mL, 1 h), LPS (100 ng/mL, 1–16 h), or CS (5%, 16 h) were added to the serum-free HLMVEC or HPAEC for the specified amount of time, as previously published (14, 36, 37) in the presence or absence of TACE inhibitors: TAPI₁ (50 μg/mL, 1 h), GM6001 (2 μM, 1 h), and AAT (500 μg/mL, 1–16 h). Aqueous CS extract (100%) was prepared from filtered research-grade cigarettes (1R3F, Kentucky Tabaco Research and Development Center). Similarly, the ambient air control extract (AC) was prepared by bubbling ambient air into sterile PBS, as previously described (22).

Rolling and Adhesion Flow Studies

HLMVEC were seeded at 1.5 × 10⁶ cell/mL density in μ-Slides I^0.4 Luer ibiTreat chamber (Ibidi GmbH, Martinsried, Germany) and maintained in complete EGM2-MV media, under static conditions, for 2 to 4 h, until achieving a confluent monolayer. Confluent HLMVEC were then treated with 500 μg/mL AAT and 100 ng/mL LPS for 16 h, 50 μg/mL TAPI-1 for 1 h, or 5 μg/mL neutralizing anti-sCX3CL1 antibodies for 30 min, all in serum-free conditions. THP-1 monocytes were resuspended at 10 × 10⁶ cell/mL in EGM-2 media and loaded into a 10 mL Gastight syringes (Hamilton Co, Reno, NV), which was connected to a KD Scientific (KDS) KDS 100 Infusion Pump (Analytical West Inc. Corona, CA) using sterile 1/16 × 1/8 out diameter silicone tubing (Tygon SPT 3350, USP Corp. Lima, OH). The μ-Slides I^0.4 Luer was then connected to the silicone tubing with Luer adaptors rendering a closed circulating system. After removing air bubbles, THP-1 cell flow was initiated by running the KDS pump at 1.3 mL/h to achieve a unidirectional laminar flow and low shear stress (38) in the range of 12–15 dynes/cm². The system was placed on a heated stage of the Zeiss 200 M inverted microscope and time-lapse videos of two different randomly selected fields on the μ-Slides I^0.4 Luer were obtained with the ×20 objective at 30 frames/s for over 3 min. “Rolling events” were defined as THP-1 cell moving over the HLMVEC monolayer for > 3 frames and “adhesion events” as THP-1 cell stopping onto the HLMVEC monolayer for ≥ 2 frames. The time-lapse video images were analyzed using the nearest neighbor tracker algorithm of TrackMate, a FIJI plug-in.

Migration Assay

Cell migration was measured using Chemicon QCM 96-well 3-μm chemotaxis assay (Millipore, Temecula, CA) as per the manufacturer’s protocol.

TACE Activity

TNF-α converting enzyme (TACE) activity was assessed in the membrane fractions of HLMVEC as chemiluminescence using SensoLyte 520 TACE Activity Assay Kit (Anaspec, Fremont, CA) as per the manufacturer’s instructions.

Urea, sCX3CL1, and MCP-1 ELISA

Urea assay (MAK006-1KT, Sigma Aldrich), sCX3CL1 (DCX310, R&D Systems), and MCP-1 (DCP00, R&D Systems) ELISA were measured as per the manufacturer’s protocols.

Western Blotting

Cells were harvested in RIPA lysis buffer. Cytosolic and membrane fractions were extracted with the membrane protein extraction kit (Biovision, Milpitas, CA), using the manufacturer’s protocol. Whole cell lysates or membrane fractions were loaded in equal amounts, as determined by bicinchoninic acid (BCA) protein analysis (Pierce). Proteins were separated by SDS-PAGE and transferred onto a PVDF membrane followed by immunoblotting. The chemiluminescent signal was detected using ECL-plus (Amersham, NJ) and normalized using anti-GAPDH (1:5,000) antibody.

Human Subjects

Plasma from healthy nonsmokers, smokers, and subjects with COPD enrolled in COPDGene study (39) (n = 216) and from 28 Pi*Z AATD subjects off (n = 14) and on (n = 14) augmentation therapy, as well as BAL fluid from 49 Pi*Z AATD subjects off (n = 18) and on (n = 29) augmentation therapy enrolled in Genomic Research in Alpha-1 Antitrypsin Deficiency and Sarcoidosis (GRADS; 40) study were used. Despite using similar instilled saline volumes, we corrected for the variability of BAL fluid return volumes, by using the ratio of urea concentration in the BAL fluid to that in the serum to calculate the volume of epithelial lining fluid (ELF) recovered after each BAL procedure. We have used the following formulas: Urea_BAL/Urea_plasma × Vol_BAL = ELF and [sCX3CL1] × Vol_BAL/ELF = sCX3CL1_ELF as previously described (41, 42). All protocols were approved by the NJH IRB, COPDGene, and GRADS ancillary studies. Patient characteristics are summarized in Table 2. Plasma sCX3CL1 levels association with clinical (age, waist circumference, BMI, history of diabetes, sex, and clinical center), spirometry (percent predicted FEV₁, FEV₁/FVC, FEV₁, DLCO percent predicted), and radiology data [adjusted lung density (g/L, PD15adj), airway wall thickness (AWT), square root of wall area of a 10-mm airway lumen perimeter (Pi10 SRWA), parametric response mapping (PRM) % air trapping: PRM in the nonemphysematous area of the lung with air trapping; PRM other: PRM in other abnormal, e.g., fibrosis are of the lung; PRM normal: PRM in the normal area of the lung; PRM % emphysema: PRM in the emphysematous area of the lung] was tested.

Table 2.

Patient characteristics

	COPDGene				AATD		GRADS
	Plasma						BALf
Patient Characteristics	Never Smoker n = 13	Current or Former Smoker No COPD n = 79	Mild COPD n = 67	Advanced COPD n = 57	Pi*Z No COPD n = 9	Pi*Z COPD n = 19	Pi*Z No COPD n = 15	Pi*Z COPD n = 32
sCX3CL1 (ng/mL) median ± IQR	0.7 (0.6)	0.2 (0.4)	0.3 (0.7)	0.5 (0.7)	0.34 (0.1)	0.44 (0.1)	277.78 (639.5)	474.6 (1,344.4)
AAT augmentation, n (%)	0	0	0	0	1 (11)	14 (74)	3 (20.0%)	26 (81.3%)
Age (yr) means ± SD	58.5 (7.0)	65.7 (7.4)	72.0 (8.0)	69.2 (7.6)	57 (17)	57 (23)	50.6 (11.00)	60.5 (10.9)
Males n (%)	4 (30.8)	40 (50.6)	40 (59.7)	36 (63.2)	3 (27)	4 (21)	5 (33.3%)	15 (46.9%)
BMI (kg/m2) means ± SD	25.4 (3.9)	29.3 (6.2)	28.1 (4.8)	27.1 (6.8)	28 (11)	25 (10)	27.0 (4.2)	28.4 (5.6)
Waist circumference (cm) mean ± SD	90.8 (11.1)	99.0 (14.9)	99.2 (12.5)	98.6 (17.2)	NA	NA	NA	NA
History of diabetes n (%)	0 (0)	15 (19.0)	12 (17.9)	5 (8.8)	NA	NA	NA	NA
Current smoker n (%)	0 (0)	19 (24.1)	11 (16.4)	10 (17.5)	0	0	1 (6.7)	0
ATS pack-years median ± IQR	0 (0)	44.4 (27.5)	44.6 (31.7)	48.0 (41.7)	6	23.6 (13)	0.00 (0.75)	7.00 (24.00)
FEV1 (L) means ± SD	3.1 (0.8)	2.7 (0.7)	2.0 (0.7)	0.9 (0.3)	1.18 (0.7)	2.1(1.4)	3.29 (0.90)	1.73 (0.66)
FEV1/FVC means ± SD	0.8 (0.0)	0.8 (0.0)	0.6 (0.1)	0.4 (0.1)	NA	NA	97.33 (11.85)	55.41 (15.99)
FEV1 %predicted means ± SD	103.7 (9.9)	93.3 (11.7)	72.4 (16.8)	34.6 (10.1)	77.1 (37)	42 (21.8)	0.78 (0.05)	0.47 (0.10)
DLCO % predicted means ± SD	23.3 (3.7)	23.1 (3.6)	23.1 (4.1)	22.8 (3.4)	68.6 (39)	51.8 (26.8)	97.86 (22.45)	67.45 (23.91)
Emphysema (% LAA <−950 HU) median ± IQR	0.6 (1.4)	1.1 (1.8)	5.9 (13.8)	21.7 (24.6)	NA	NA	2.50 (3.30)	16.30 (19.10)
PD15adj (g/L) means ± SD	85.8 (10.4)	91.0 (17.3)	71.4 (21.6)	53.4 (21.5)	NA	NA	78.38 (12.27)	51.48 (17.62)
PRMAir Trapping median ± IQR	4.5 (5.0)	8.1 (6.9)	22.7(12.6)	33.8 (12.2)	NA	NA	NA	NA
PRMEmphysema median ± IQR	0.1 (0.4)	0.2 (0.6)	4.1 (11.4)	20.9 (25.7)	NA	NA	NA	NA
PRMNormal median ± IQR	94.2 (6.5)	89.6 (7.8)	68.2 (28.6)	39.6 (30.3)	NA	NA	NA	NA
PRMOther median ± IQR	0.6 (1.6)	1.0 (1.3)	2.5 (2.8)	2.0 (2.3)	NA	NA	NA	NA
AWT means ± SD	0.9 (0.1)	1.0 (0.2)	1.1 (0.2)	1.1 (0.2)	NA	NA	NA	NA
Pi10 means ± SD	1.7 (0.2)	1.9 (0.4)	2.3 (0.6)	2.8 (0.5)	NA	NA	NA	NA

AAT, Alpha-1 antitrypsin; AATD, Alpha-1 antitrypsin deficiency; ATS, American Thoracic Society; AWT, Airway wall thickness; BMI, body mass index; COPDGene, Genetic epidemiology of COPD; DLCO, diffusion capacity for carbon monoxide; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; GRADS, the genomic research in alpha-1 antitrypsin deficiency and sarcoidosis; HU, Hounsfield unit; IQR, interquartile range; LAA, low attenuation area; PD15, lung density measured as the volume-adjusted 15th percentile density; PRM, parametric response mapping; PRM % air trapping, PRM in the nonemphysematous area of the lung with air trapping; PRM other, PRM in other abnormal, e.g. fibrosis are of the lung; PRM normal, PRM in the normal area of the lung; PRM % emphysema, PRM in the emphysematous area of the lung; Pi10 (SRWA), square root of wall area of a 10-mm airway lumen perimeter (Pi10).

Statistical Analysis

Statistical analysis was performed with GraphPad Prism v7.0 (La Jolla, CA) and Statiscal Analysis Software (SAS), using unpaired Student’s t test, one-way ANOVA with Tukey’s post hoc multiple comparisons test was used when comparing the means of all groups, regardless of control, or nonparametric tests as appropriate. Plasma sCX3CL1 between different COPD groups was evaluated using Kruskal–Wallis test, followed by a pairwise Wilcoxon test. Plasma sCX3CL1 association with clinical parameters was tested using Spearman correlation. In other analysis, plasma sCX3CL1 levels were natural log-transformed and multivariable linear regression was used to evaluate the associations of sCX3CL1 with spirometry and radiology data independent of clinical parameters. Models included are age, waist circumference, BMI, history of diabetes, sex, and clinical center. For FEV₁ percent predicted and DLCO percent predicted only waist circumference, history of diabetes, and clinical center were used for modeling. This was an exploratory analysis, the significance is shown at the P value level, adjustment for multiple comparisons (FDR level) is not shown because the clinical phenotypes evaluated have high levels of association with each other, hence they are not truly independent variables. P < 0.05 was considered statistically significant.

RESULTS

Alpha-1 Antitrypsin In Vivo Effect on CX3CL1 Shedding and Leukocyte Transendothelial Migration

To test the AAT effect on monocyte transendothelial migration and its dependence on CX3CL1-CX3CR1 axis, we used two models of acute inflammation where WT and CX3CR1^gfp/gfp (KO) mice received one dose of LPS or 1-mo CS exposure (Fig. 1A). At 3 days after LPS or 1-mo and 6-mo after CS exposure, sCX3CL1 shedding in BALF, but not in plasma, was increased, independent of mice genotype (Fig. 1, B and C, Supplemental Fig. S2A). Prophylaxis with exogenous AAT normalized sCX3CL1 BALF levels in the WT (Fig. 1, B and C), but only partially in CX3CR1^gfp/gfp (KO) mice after LPS-exposure (Fig. 1, B and C). Lower sCX3CL1 in AAT-treated mice was associated with significantly decreased transendothelial migration of CX3CR1⁺ monocytes (Fig. 1D), neutrophils (Fig. 1E), and CD11b⁺ recruited monocytes (Fig. 1F) only in the LPS-exposed WT mice, but not in the and CX3CR1^gfp/gfp (KO) mice that had an elevated sCX3CL1 even after AAT administration. Similarly, AAT treatment administered once weekly for 3 wk during CS exposure significantly decreased sCX3CL1 BALF levels, transendothelial migration of CX3CR1⁺ monocytes (Fig. 1G), neutrophils (Fig. 1H), and CD11b⁺ recruited monocytes (Fig. 1I) in the WT, but not in CX3CR1^gfp/gfp mice. The AAT effect on CD11b^-CD11c⁺ airspace macrophages in the LPS model was insignificant, and we detected a modest effect of CS exposure in the WT mice (Supplemental Fig. S2, B and C). Interestingly, during homeostatic conditions in mice that received oropharyngeal PBS or were exposed to ambient air, AAT affected neither leukocyte numbers nor sCX3CL1 shedding in the BALf (Table 3). At baseline, despite a deregulated CX3CL1-CX3CR1 axis with higher sCX3CL1 in the plasma and BALf of CX3CR1^gfp/gfp (KO) mice (Fig. 1, B and C), we did not see a significantly different number of BALf neutrophils, airspace macrophages, or CX3CR1⁺ monocytes (Table 3).

Figure 1.

Endothelial transmigration of CX3CR1⁺ monocytes during acute inflammation in mice is dependent on AAT effect on the CX3CL1-CX3CR1 axis. A: schematic of two acute inflammation models using LPS (50 μg/mouse) administered via oropharyngeal aspiration on day 1 and CS exposure, 5 days/wk for 4 wk. Intraperitoneal PBS or AAT (20 mg/kg body wt) was administered prophylactically, one individual dose the day prior to LPS administration or as a treatment, weekly dose for 3 wk during 1-mo CS exposure, respectively. B and C: plasma and BALf sCX3CL1 levels measured by ELISA in CX3CR1^gfp/gfp (n = 8–10 mice/group) and WT littermates (n = 4–7 mice/group) at day 3 post LPS administration (B) or at 1-mo post CS exposure (C) treated or not with intraperitoneal AAT. D–I: flow cytometry analysis of CX3CR1⁺ (CD45⁺CD11b⁺Ly6G^-F4/80⁺Ly6C^-CX3CR1⁺) monocytes; neutrophils (CD45⁺CD11b⁺Ly6G⁺); and recruited monocytes (CD45⁺CD11b⁺Ly6G^-F4/80⁺). D: CX3CR1⁺ monocytes, E: neutrophils, and F: recruited monocytes absolute counts in the BAL fluid at day 3 post LPS administration and AAT pretreatment. G: CX3CR1⁺ monocytes, H: neutrophils and I: recruited monocytes counts in the BAL fluid after 1-mo CS exposure and AAT treatment. Data pooled from three independent LPS- or CS-exposure experiments are presented as means ± SD, one-way ANOVA, followed by Tukey’s multiple-comparisons, #P < 0.05 vs. PBS-treated mice, *P < 0.05 vs. LPS-treated or CS-exposed mice; ^P < 0.05 vs. LPS-treated or CS-exposed WT mice, ‡P < 0.05 vs. PBS-treated WT mice. AAT, α-1 antitrypsin; CS, cigarette smoke; LPS, lipopolysaccharide; WT, wild type.

Table 3.

AAT effect on sCX3CL1 level and BAL leukocytes at baseline

Baseline AAT Effect	WT PBS	WT AAT	CX3CR1gfp/Gfp PBS	CX3CR1gfp/Gfp AAT
LPS acute inflammation model
BAL PMN (×105 cells)	0.13 ± 0.07	0.03 ± 0.02	0.1 ± 0.12	0.09 ± 0.08
BAL CD11b+ macrophages (×105 cells)	0.07 ± 0.04	0.06 ± 0.06	0.11 ± 0.12	0.1 ± 0.007
BAL CX3CR1+ macrophages (×105 cells)	0.003 ± 0.002	0.005 ± 0.003	0.03 ± 0.04	0.02 ± 0.01
BAL CD11c+ macrophages (×105 cells)	1.4 ± 0.37	1.8 ± 0.6	1.2 ± 0.6	1.6 ± 1.4
BAL sCX3CL1 (ng/mL)	0.7 ± 0.19	0.64 ± 0.04	2.5 ± 0.8	2.1 ± 0.7
Plasma sCX3CL1 (ng/mL)	3.5 ± 0.54	4.1 ± 0.02	72.1 ± 5.1	74.9 ± 1.3
CS acute inflammation model
BAL PMN (×105 cells)	0.02 ± 0.01	0.02 ± 0.01	0.02 ± 0.01	0.02 ± 0.01
BAL CD11b+ macrophages (×105 cells)	0.22 ± 0.04	0.32 ± 0.2	0.38 ± 0.07	0.42 ± 0.28
BAL CX3CR1+ macrophages (×105 cells)	0.004 ± 0.001	0.005 ± 0.002	0.002 ± 0.0001	0.002 ± 0.0007
BAL CD11c+ macrophages	0.9 ± 0.3	0.9 ± 0.4	1.3 ± 0.8	1.1 ± 0.6
BAL sCX3CL1 (ng/mL)	0.8 ± 0.2	0.7 ± 0.23	1.3 ± 0.17	1.6 ± 0.09
Plasma sCX3CL1 (ng/mL)	8.1 ± 0.12	8.1 ± 0.75	85.8 ± 1.49	83.8 ± 1.62

Values are means ± SE. AAT, α-1 antitrypsin; BAL, bronchoalveolar lavage; CS, cigarette smoke; LPS, lipopolysaccharide; WT, wild type.

Our data suggest a lung-specific, anti-inflammatory effect of AAT linked to reducing CX3CR1⁺ monocyte transendothelial migration into the alveolar compartment. The AAT inhibitory effect on monocyte migration was dependent on an intact CX3CL1-CX3CR1 axis since we did not detect a similar effect in CX3CR1^gfp/gfp (KO) mice, with increased sCX3CL1 levels and CX3CR1⁺ monocytes in the BALF, a sign of a disrupted CX3CL1-CX3CR1 axis.

Effect of AAT on sCX3CL1 Shedding and Monocyte-Endothelium Rolling and Adhesion

During homeostasis, HLMVEC cultured in full serum media, or serum-deprived media supplemented with AAT shed less sCX3CL1 than HLMVEC maintained in serum-deprived conditions (Fig. 2A). We noticed similar AAT effect on sCX3CL1 shedding from HLMVEC exposed to CS extract (Fig. 2E) and from large vessel endothelial cells, HPAEC (Fig. 2, B and C). Subsequently, AAT increased full-length CX3CL1 expression on HLMVEC and HPAEC exposed to serum deprivation at 1 h as measured by FACS using intracellular staining against a C-terminal, intracytoplasmic CX3CL1 epitope (Fig. 2, D and E), but not at 16 h (Supplemental Fig. S3, A and B). In PMA-stimulated HLMVECs cultured in serum-deprived conditions, AAT recapitulated the effect of GM6001, a TACE chemical inhibitor, by decreasing sCX3CL1 shedding (Fig. 2F). Furthermore, AAT, similarly to GM6001, inhibits TACE activity at the plasma membrane of PMA-stimulated HLMVEC (Fig. 2G). AAT had no effect on Cx3cl1 gene expression (Fig. 2H), though serum deprivation and PMA stimulation increased Cx3cl1 gene expression (Fig. 2H and Supplemental Fig. S3C). AAT effect was specific to CX3CL1-CX3CR1 axis because in PMA-treated HLMVEC AAT did not affect MCP-1 proinflammatory cytokine secretion (Supplemental Fig. S3D) or E-selectin (CD62E) membrane integrin expression (Supplemental Fig. S3E). Interestingly, LPS treatment in prolonged serum-deprived conditions increased Cx3cl1 gene expression 10-fold higher than serum-deprivation alone, resulting mainly in sCX3CL1 shedding and minimal changes in full-length CX3CL1 expression (Supplemental Fig. S3, F–I). At later time points, following 16 h LPS or CS-extract stimulation, AAT did not have an appreciable effect on sCX3CL1 shedding or endothelial full-length CX3CL1 expression (Supplemental Fig. S3, F–H).

Figure 2.

AAT decreases sCX3CL1 shedding by the endothelium in a TACE-dependent manner. sCX3CL1 shedding measured by ELISA (A and B) and full-length CX3CL1 membrane expression measured by FACS (C and D) in HLMVEC (A and C) and HPAEC (B and D) stimulated by acute serum deprivation (0% serum, 1 h) and treated with AAT (500 μg/mL, 1 h) vs. full media, control conditions. E and F: sCX3CL1 shedding by HLMVECs stimulated by serum deprivation and CS (5%, 16 h, E) and PMA (100 ng/mL, 1 h, F) and treated with AAT (500 μg/mL, 1 h) or the TACE inhibitor, GM6001 (2 μM, 1 h). G: TACE activity in the membrane fraction of HLMVECs stimulated by serum deprivation and PMA (100 ng/mL, 1 h) and treated with AAT (500 μg/mL, 1 h) or GM6001 (2 μM, 1 h). H: Cx3cl1 gene transcription in HLMVECs stimulated by acute and prolonged serum deprivation (0% serum, 1 h or 16 h) and treated with AAT (500 μg/mL, 1 h or 16 h) vs. full media, control conditions. All panels show data points from ≥ 3 independent experiments and means ± SE, one-way ANOVA followed by Tukey’s multiple comparisons, #P < 0.05 vs. serum, *P < 0.05 vs. serum deficiency or vs. PMA. AAT, α-1 antitrypsin; HLMVECs, human lung microvascular endothelial cells; HPAEC, human pulmonary artery endothelial cell; PMA, phorbol 12-myristate 13-acetate; TACE, TNF-α-converting enzyme.

These experiments suggest that AAT time-dependently inhibits sCX3CL1 endothelial shedding via a posttranslational mechanism involving TACE inhibition.

We next thought to investigate the specific processes, e.g., rolling, adhesion, and migration implicated in monocyte recruitment. We exposed ex vivo THP-1, primary human monocytes and endothelial-monocyte coculture to sCX3CL1 and proinflammatory stimuli to induce endothelial sCX3CL1 shedding.

Stimulation with sCX3CL1 time-dependently decreased the levels of CX3CR1 (Supplemental Fig. S4A), disrupted full length CX3CL1-CX3CR1 interactions, and increased levels of integrin αMβ2 (CD11b/CD18, Supplemental Fig. S4A) at hPBDMs plasma membrane. Next, we used THP-1 monocytes, which exhibit a more consistent CX3CR1 expression than hPBDMs isolated from different donors (Supplemental Fig. S4B). THP-1 monocytes were loaded into a Gastight syringe and added to a HLMVEC monolayer seeded in μ-slide chambers using a KDS100 pump to deliver a continuous flow at low shear stress (12–15 dynes/cm²). We measured monocyte rolling and adhesion using time-lapse microscopy. We defined adhesion events as the number of THP-1 monocytes stopping onto the HLMVEC monolayer for ≥ 2 frames and above the threshold (mean +2 SD, red line in Fig. 3, B and C) calculated in full serum, control conditions. First, we tested the ability of our system to detect differences in monocyte rolling, adhesion, and migration after LPS treatment, a known stimulus for sCX3CL1 shedding (Supplemental Fig. S4, C and D).

Figure 3.

Monocyte adhesion to endothelium and migration is dependent on AAT effect on CX3CL1–CX3CR1 axis. A–C: monocyte adhesion to HLMVECs monolayer assay measured using time-lapse microscopy and Zeiss 200 M-inverted microscope. A unidirectional and low shear stress flow of THP-1 monocytes inside μ-Slides I^0.4 Luer ibiTreat chamber with a HLMVEC monolayer was achieved and maintained with a KDS pump set at 1.3 mL/h. A: representative static images of THP-1 monocytes (round and clear, top) adhesion to HLMVECs (cobblestone-like, bottom) after 16 h in full media (5% serum, i), serum deprived (0%, 16 h, ii), serum-deprived media supplemented with AAT (500 μg/mL, 16 h, iii), or with TACE inhibitor, TAPI-1 (50 μM, 16 h, iv). Scale bar: 50 μm. B: quantification of THP-1 monocyte adhesion to HLMVECs from time-lapse videos of conditions indicated in (A) using TrackMate nearest neighbor tracker algorithm. Individual dots represent “adhesion events”, aka number of THP-1 cell stopping onto the HLMVEC monolayer for ≥ 2 frames. C: quantification of THP-1 monocytes adhesion to HLMVECs stimulated by serum deprivation (0%, 16 h) or serum-deprived media supplemented with anti-sCX3CL1 inhibitory antibodies (5 μg/mL, 30 min prior to imaging). All quantifications were normalized to full media as control. Red dashed line represent means ± 2SD in full media (control) condition. D and E: migration of THP-1 (0.5 × 10⁵cells/mL) seeded in a chemotaxis tray toward conditioned medium of HLMVECs maintained in full media, stimulated by serum deprivation (0% serum, 16 h), or serum-deprived media supplemented with AAT (500 μg/mL, 16 h) vs. CM (D) or toward sCX3CL1 (50 ng/mL, 4 h) and/or LPS (100 ng/mL, 16 h) vs. THP-1 control media (CM) expressed as relative fluoresce units (RFUs) (E). All panels show data points from ≥3 independent experiments. Data are presented as means ± SE, one-way ANOVA followed by Tukey’s multiple-comparisons or t test, respectively, #P < 0.05 vs. CM THP-1, *P < 0.05 vs. serum-deficiency. AAT, α-1 antitrypsin; HLMVECs, human lung microvascular endothelial cells; TACE, TNF-α-converting enzyme.

Compared to conditions that preserve full-length CX3CL1 endothelial expression, such as full serum media (Fig. 3Ai, 3B, Supplemental Video S1), serum-depleted media supplemented with AAT (Fig. 3Aiii, 3B, Supplemental Video S3), treatment with TACE inhibitor (TAPI-1, Fig. 3Aiv, 3B, Supplemental Video S4), or blocking antibodies against sCX3CL1 (Fig. 3C), the conditions with high sCX3CL1 shedding such as serum depletion (Fig. 3Aii, B and C, Supplemental Video S2) had significantly higher number of THP-1 monocyte adhere to the endothelium, hence, decreased rolling ability.

In addition, the conditioned medium of serum-deprived HLMVEC (high sCX3CL1) significantly increased THP-1 migration versus conditioned medium of HLMVEC maintained in full serum or in serum-depleted media supplemented with AAT (low sCX3CL1, Fig. 3D). sCX3CL1 failed to increase THP-1 migration independently, but increased LPS-stimulated THP-1 migration (Fig. 3E).

These experiments suggest that sCX3CL1 and high sCX3CL1 shedding conditions increased monocyte adhesion and migration, whereas AAT and TACE inhibitors of CX3CL1 shedding inhibited these processes.

The Inhibitory Effect of AAT on sCX3CL1 Shedding Is TACE Mediated

To further investigate AAT specificity on TACE activity we measured AAT effect on wild-type and mutated CX3CL1 shedding. In HPAEC transfected with a wild-type CX3CL1 plasmid harboring HA-tag at the C-terminal (Fig. 4A and Supplemental Fig. S5A) we measured full length (110 kDa) and cleaved (27 kDa) CX3CL1 overexpression. Again, LPS and PMA treatments under serum-deprived conditions increased cleaved (27 kDa)/full length (110 kDa) CX3CL1 ratio as measured by Western blotting, rendering TACE activation and sCX3CL1 shedding (Fig. 4, B and C). Treatment with TACE inhibitors, AAT and GM6001 decreased cleaved/full length CX3CL1 ratio at the plasma membrane, similar to baseline, full serum conditions (Fig. 4, B and C). However, AAT and GM6001 had no effect on the cleaved/full length CX3CL1 ratio in HPAEC transfected with mutant, C-terminal HA-tagged CX3CL1 plasmid harboring a point mutation (ΔLeu324) at the TACE cleavage site (Fig. 4, D and E). Other point mutations around the TACE cleavage site (ΔThr₃₂₉, ΔAla₃₃₄, ΔVal₃₃₁, and ΔVal₃₂₆) did not protect against serum deprivation or PMA-induced cleavage of full-length CX3CL1 (Supplemental Fig. S5C). Total (cleaved + full length) wild-type and mutant CX3CL1 were not significantly changed by PMA and LPS stimulation, nor by AAT or GM6001 treatments (Supplemental Fig. S5, B and D).

Figure 4.

AAT decreases cleaved/mb-bound CX3CL1 ratio in endothelium in a TACE-dependent manner. A: schematic of wild-type and mutant full length CX3CL1 cleaved by TACE into soluble (sCX3CL1) and cleaved CX3CL1 isoforms. HA-tag was introduced in the C-domain to allow tracking via Western blotting of the C-domain containing CX3CL1 isoforms: full length membrane-bound, full length intracellular, and cleaved CX3CL1. CX3CL1 mutants were obtained by amino acid deletion (ΔLeu₃₂₄, ΔVal₃₂₆, ΔThr₃₂₉, ΔVal₃₃₁, and ΔAla₃₃₄) within the putative TACE cleavage site in the extracellular N-domain. B–E: representative immunoblots (B, D) of cleaved (∼27 kDa) and mb-bound full-length (∼90 kDa) CX3CL1 and densitometry (C, E) of the 27 kDa/90 kDa ratio in HPAEC overexpressing wild-type (B, C) or mutant (D, E) HA-tagged CX3CL1 plasmid (5 μg/1 × 10⁶ cells, 48 h). Cells were exposed to proinflammatory stimuli PMA (100 ng/mL, 1 h) or LPS (100 ng/mL, 16 h) in the presence or absence of AAT (500 μg/mL, 1 h and 16 h) or GM6001 (2 μM, 1 h) for the indicated time. All panels show data points from ≥3 independent experiments. Means ± SE, one-way ANOVA with Tukey’s multiple-comparisons, #P < 0.05 vs. full media, *P < 0.05 vs. serum-free PMA or LPS. AAT, α-1 antitrypsin; HPAEC, human pulmonary artery endothelial cell; LPS, lipopolysaccharide; PMA, phorbol 12-myristate 13-acetate; TACE, TNF-α-converting enzyme.

These experiments suggest that the effect of AAT on the CX3CL1 membrane expression is dependent on an intact amino acid sequence in the full-length CX3CL1 molecule that contains the active TACE-binding site.

Soluble CX3CL1 Levels in Plasma and BALf of Individuals with COPD and AATD and Associations with Clinical Parameters

We next evaluated the CX3CL1-CX3CR1 axis in 216 subjects at risk for or diagnosed with COPD, recruited for the COPDGene cohort, by measuring plasma sCX3CL1 levels and their correlations with markers of disease severity (Table 4). Compared with subjects at risk for but without COPD (n = 79, 0.2 ± 0.4 ng/mL), sCX3CL1 was increased in individuals COPD (n = 137, 0.41 ± 0.5 ng/mL, Fig. 5A). We noticed significantly higher sCX3CL1 levels in stage 3–4 GOLD COPD (n = 57, 0.55 ± 0.4 ng/mL, p = 0.047) compared with subjects at risk (Supplemental Fig. S6A). In 28 Pi*Z individuals with or without COPD sCX3CL1 levels were comparable (0.38 ± 0.1 ng/mL) with the COPDGene subjects (0.5 ± 0.4 ng/mL, Fig. 5, A and B). Plasma sCX3CL1 was increased in Pi*Z individuals with COPD (0.45 ± 0.1 ng/mL) compared with those without COPD (0.3 ± 0.1 ng/mL, P = 0.009, Fig. 5B). Similar to plasma, after correction for urea BALf/plasma ratio, BALf sCX3CL1 levels were increased in Pi*Z individuals with COPD compared with those without COPD L (P = 0.0001, Fig. 5C). In this small cohort, AAT augmentation therapy had no effect on plasma and BALf sCX3CL1 levels in individuals with AATD on (0.42 ± 0.14 ng/mL and 97.3 ± 180, dark circles, Fig. 5, B and C) or off AAT augmentation therapy (0.39 ± 0.1 ng/mL and 14.93 ± 22.9 in plasma and BALf, respectively, white circles). There was a significant association between plasma and BALf sCX3CL1 and %emphysema, and between plasma sCX3CL1 and parametric response mapping (PRM) in the normal, other abnormal (e.g., fibrosis) and emphysematous areas of the lung [PRM^normal, PRM^other, PRM^emphysema], but not with exacerbation frequency, FEV₁, FEV₁/FVC ratio, DLCO, or radiological measurements of airway disease as measured by bronchial wall thickness of inner perimeter of 10 mm airway (Pi10), or airway wall thickness (AWT, Table 4). In a multivariate regression analysis that included age, sex, waist circumference, BMI, history of diabetes, and the clinical site of enrollment, each log unit increase in plasma sCX3CL1 level explained a small, 0.00596% but statistically significant variance of radiologic parameters of small airway disease as measured by PRM in nonemphysematous areas with air trapping (PRM^airtrapping, R² estimate = 0.00596, P = 0.045). In this multivariate regression analysis, the plasma sCX3CL1 level was not associated with variation in exacerbation frequency, FEV₁, FEV₁/FVC ratio, DLCO, or radiological measurements of airway disease as measured by bronchial wall thickness of inner perimeter of 10 mm airway (Pi10), or airway wall thickness (data not shown). We next evaluated the effect of active smoking on sCX3CL1. Similar to previous reports that acute inflammatory states, e.g., induced by ischemic stroke (43) or major orthopedic and cardiac surgeries (44), decrease sCX3CL1, we found that compared to healthy never smokers (n = 13, 0.7 ± 0.6 ng/mL), current smokers without COPD (n = 19, 0.2 ± 0.4 ng/mL) had lower plasma sCX3CL1 (Supplemental Fig. S6B).

Table 4.

sCX3CL1 association with COPD parameters

	COPDGene Plasma		GRADS BALf
Parameter	Rho	P Value	Rho	P Value
PRMair trapping	0.17	0.02	NA	NA
PRMemphysema	0.14	0.05	NA	NA
PRMnormal	−0.14	0.04	NA	NA
PRMOther	0.14	0.04	NA	NA
%Emphysema, −950 HU	0.14	0.05	0.29	0.049
PD15adj, g/L	−0.12	0.06	−0.29	0.05
AWT	−0.03	0.64	NA	NA
Pi10 (SRWA)	0.008	0.91	NA	NA
FEV1 (L)	−0.10	0.13	−0.19	0.20
FEV1/FVC	−0.07	0.29	−0.20	0.18
FEV1 % predicted	−0.06	0.4	−0.25	0.10
DLCO % predicted	−0.06	0.39	−0.10	0.51

Analysis: Spearman correlation using SAS. Legend: AWT, Airway wall thickness; COPD, chronic obstructive pulmonary disease; DLCO, diffusion capacity for carbon monoxide; FEV1: forced expiratory volume in 1 second; FVC: forced vital capacity; GRADS, Genomic Research in Alpha-1 Antitrypsin Deficiency and Sarcoidosis; HU, Hounsfield unit; PRM, parametric response mapping; PRM % air trapping, PRM in the nonemphysematous area of the lung with air trapping; PRM other, PRM in other abnormal, e.g., fibrosis are of the lung; PRM normal, PRM in the normal area of the lung; PRM % emphysema, PRM in the emphysematous area of the lung; Pi10 (SRWA), square root of wall area of a 10-mm airway lumen perimeter (Pi10). The bold values are statistically significant.

Figure 5.

CX3CL1-CX3CR1 axis in COPD. A–C: plasma and BALf sCx3CL1 levels in COPD subjects enrolled in COPDGene cohort (A) and Pi*Z AATD individuals (B and C) measured by ELISA. BALf sCX3CL1 levels are normalized by the volume of epithelial lining fluid (ELF) imputed from BALf/plasma urea ratio and BAL fluid volume. Median ± IQR, Mann–Whitney test, P < 0.05 vs. smokers (current and former) or Pi*Z individuals without COPD. Pi*Z AATD individuals on augmentation therapy are depicted by dark circles (B and C). AATD, α-1 antitrypsin deficiency; COPD, chronic obstructive pulmonary disease; IQR, interquartile range.

DISCUSSION

Our findings demonstrate that exogenous AAT and pharmacological TACE inhibitors enable monocyte-endothelium rolling thus decreasing adhesion and migration into the alveolar space. Interestingly, AAT effect appears to be lung-specific, as AAT decreased sCX3CL1 shedding in the BALF, but not in plasma of mice exposed to LPS or CS or in plasma of Pi*Z individuals with COPD on AAT augmentation therapy. AAT inhibition of sCX3CL1 shedding was TACE-mediated, as AAT had no effect on CX3CL1 transcription or translation. Moreover, AAT effect on sCX3CL1 shedding required an intact CX3CL1-CX3CR1 axis, as AAT had minimal effect in the CX3CR1^gfp/gfp (KO) mice or in HLMVECs transfected with TACE-resistant Cx3cl1-mutated construct (ΔLeu₃₂₄). In chronic inflammatory conditions like CS-related COPD and AATD, plasma sCX3CL1 was mildly, but significantly increased and it was associated with radiological parameters of small airway disease, making CX3CL1-CX3CR1 axis a possible biomarker and therapeutic target for detection and treatment of those smokers and individuals with AATD whose small airway disease progresses toward established emphysema.

Noncanonical functions of AAT, aside from inhibiting neutrophil elastase, have established AAT as a major protective molecule against lung injury in individuals with COPD and AATD. Several AAT noncanonical functions, like inhibition of neutrophil chemotaxis and endothelium proinflammatory activation are mediated via inhibition of proteases, especially TACE, along with caspases, matrix metalloproteinases, or serine proteases associated with complement cascade (45, 46). Here we report that AAT is an endogenous TACE inhibitor and elaborate on our previous work, that partially, AAT anti-inflammatory effect on lung endothelium is mediated via noncanonical inhibition of TACE with downstream effects on CX3CL1-CX3CR1 axis (47).

Although TACE-deficient mice were born with hypoplastic lungs and vasculogenesis delay (48), conditional TACE-deficient mice had attenuated goblet cell metaplasia after intratracheal neutrophil elastase administration (49), suggesting that temporary decrease in TACE activity may reduce lung injury. Moreover, reduced TACE expression in the Adam17^ex/exgp130^F/F mice protected against airspace enlargement of the emphysema-prone gp130^F/F mice, whereas ADAM8^−/− mice were predisposed to emphysema development (50, 51). In addition, inhibitory molecules (e.g., INCB3619, KP457, and INCB84298) and monoclonal antibodies [D1(A12) or A9(B8)] targeting members of ADAMs proteinases family, including TACE, showed promising clinical results in various malignancies, traumatic brain and spine injury, inflammatory arthritis, and diabetic wound (27). These inhibitors of ADAMs proteinases have not been tested in COPD. Monoclonal antibodies against downstream TACE products (e.g., infliximab against TNFα) were administered in patients with COPD with cachexia with no effect on lung inflammation and minor effects on systemic inflammation (52).

Future therapies targeting sCX3CL1, e.g., inhibitory antibodies or endothelial delivery of a TACE-resistant Cx3cl1 could protect full-length CX3CL1 binding to CX3CR1⁺ and retain CX3CR1⁺ monocytes intravascular. Introduction of a transgene encoding for sCX3CL1 (Cx3cl1^105Δ) in CX3CL1^−/− mice was sufficient to allow lamina propria CX3CR1⁺ monocytes to extend protrusions toward the intestinal lumen, suggesting that sCX3CL1 is responsible for CX3CR⁺ monocyte migration. Contrarily, introduction of Cx3cl1^395AA transgene encoding full-length CX3CL1 restored the normal number of Ly6C^-CX3CR⁺ circulating monocytes (53). In a renal injury model, CX3CR1⁺ monocytes stayed intravascular and maintained endothelial health via microparticles clearance (4). Similarly, in Parkinson and Alzheimer dementia models CX3CR1⁺ microglia protected against microglial neurotoxicity, whereas CX3CR1^gfp/gfp mice showed increased microglial activation and neuronal loss (54, 55). Finally, in a diabetic wound model, F4/80⁺CX3CR1⁺ macrophages at the wound site promoted granulation tissue only in WT mice with intact CX3CL1-CX3CR1 axis, but not in CX3CR1^gfp/gfp mice (56). In CS exposure and LPS lung injury models, we show AAT decreased specifically CD11b⁺Ly6G^-F4/80⁺Ly6C^-CX3CR1⁺ monocyte transmigration into the alveoli.

In the LPS model, AAT decreased neutrophil migration as well, suggesting that CX3CL1-CX3CR1 axis encompasses neutrophil- and monocyte-endothelial cross talk during LPS injury. Previous reports show that AAT decreases IL-8, IL-6, and MCP-1 secretion; these chemokines facilitate primarily, but not exclusively, neutrophil chemotaxis (57). The integrin αMβ2 (CD11b/CD18) receptor was involved in leukocyte migration toward these chemokines; and we noticed a stimulatory effect of recombinant sCX3CL1 on this monocyte receptor. Our findings that high sCX3CL1 states, like in LPS- and CS-exposed CX3CR1^gfp/gfp mice or sCX3CL1 administration led to CX3CR1⁺ monocyte transendothelial migration and higher CD11b/CD18 expression on hPBMCs support the concept that sCX3CL1 mediates monocyte migration via other receptors than CX3CR1.

Interestingly, plasma sCX3CL1 was not increased in our murine models of acute inflammation, mainly because intratracheal LPS administration and whole body CS exposure targeting direct lung injury had inconsequential effect on plasma sCX3CL1 levels, similar to findings described post-LPS exposure in CX3CR1^gfp/gfp mice that interrogated behavioral alterations in response to sepsis (58). Our results in COPDGene subjects suggest a relationship between plasma sCX3CL1 and endothelial injury in individuals with small airway disease, preceding emphysema development. The mild positive association between plasma sCX3CL1 and PRM^airtrapping in these patients suggests that sCX3CL1 may serve as an early biomarker of emphysema because small airway disease, comprising small airways narrowing and disappearance precedes emphysema development (59). Similar to long-term (6 mo) CS-exposed WT mice, in advanced COPD plasma and BALF sCX3CL1 were increased and sCX3CL1 levels showed a trend toward association with % emphysema and adjusted lung density (PD15_adj). This suggests that sCX3CL1 may serve as a progression biomarker in smokers developing emphysema. This is not surprising since smoking cessation, vascular density, and vascular remodeling in advanced disease may alter sCX3CL1 shedding. Our study found that BALF sCX3CL1 levels were increased in AATD with COPD, even in the absence of CS exposure, but lack of PRM measurements in the GRADS cohort did not allow us to gauge BALF sCX3CL1 association with early emphysema. Similar to plasma, we did not find an effect of AAT therapy on sCX3CL1 shedding in the BALF in AATD on therapy patients, raising questions about the effect of AAT on TACE activity and CX3CL1-CX3CR1 axis in established COPD. A limitation of our study is limited number of AATD without COPD (early disease) treated with AAT therapy to test its effect on BALF sCX3CL1 shedding in early AATD lung disease.

We focused on the CX3CL1-CX3CR1 axis in the lung because of its involvement in homeostatic rather than proinflammatory monocyte-endothelium interactions. We recognize that, considering the clinical relevance of lung and liver dysfunction in AATD, a comparative study of lung versus liver monocyte-endothelium interactions would be informative and that further studies, e.g., in AAT-deficient mice will help clarify the homeostatic versus proinflammatory role of CX3CL1-CX3CR1 in the liver. We also acknowledge that we did not investigate other ligand-receptor interactions important in monocyte-endothelial signaling including those involving integrins, CCL2, CD40L, or TLR7 (60, 61). Considering the redundancy of these axes, where the endothelial ligands could interact with multiple leukocyte receptors, we cannot rule out possible off-target AAT effects outside that of CX3CL1-CX3CR1, which could be addressed by future studies utilizing double (ccl2^−/−CX3CR1^gfp/gfp) or triple transgenic mice (ccl2^−/−tlr7^−/−CX3CR1^gfp/gfp).

Our findings define a new AAT noncanonical function, that of CX3CL1-CX3CR1 axis modulation to orchestrate monocyte-endothelial cross talk. Based on our current data that an intact CX3CL1-CX3CR1 axis is necessary to decrease sCX3CL1 endothelial shedding, CX3CR1⁺ monocyte adhesion and transendothelial migration into the alveoli, we propose that early administration of AAT or TACE inhibitors should be considered for all AATD, as well as smoker individuals with functional AAT deficiency that exhibit small airway disease and test if such intervention would reduce emphysema development and progression. Specificity of the sCX3CL1 correlations with small airway disease parameters and the mean clinically important difference for the plasma sCX3CL1, not an established clinical biomarker, will require validation cohorts. Airway AAT administration, via nebulization, may be superior to systemic AAT therapy considering the lung-specific AAT effect seen in our animal models. More specific therapies targeting endothelial TACE activity or monoclonal inhibitory antibodies against sCX3CL1 may become exciting new avenues in the treatment of early emphysema.

DATA AVAILABILITY

Data will be made available upon reasonable request.

SUPPLEMENTAL DATA

Supplemental Fig. S1: https://doi.org/10.6084/m9.figshare.21905292.

Supplemental Fig. S2: https://doi.org/10.6084/m9.figshare.21905313.

Supplemental Fig. S3: https://doi.org/10.6084/m9.figshare.21905310.

Supplemental Fig. S4: https://doi.org/10.6084/m9.figshare.21905316.

Supplemental Fig. S5: https://doi.org/10.6084/m9.figshare.21905304.

Supplemental Fig. S6: https://doi.org/10.6084/m9.figshare.21905298.

Supplemental Fig. S7: https://doi.org/10.6084/m9.figshare.23250650.

Supplemental Videos S1–S4: https://doi.org/10.6084/m9.figshare.21905322.

GRANTS

This work was supported by Alpha-1 2014 Inaugural Gordon L. Snider Scholar Award and NIH Grant K08 HL141770 (to K.A.S.), NIH Grant R01HL141264, and Flight Attendant Medical Research Institute (FAMRI) Grant CIA150041 (to F.G.), NIH Grant R01HL077328 (to I.P.), NIH Grants U01HL089897 and U01HL089856 (R.P.B. and COPDGene investigators), Genomic Research in Alpha-1 Antitrypsin Deficiency and Sarcoidosis study (GRADS) was supported by NIH Awards U01HL112707, U01 HL112695 from the National Heart, Lung, and Blood Institute, and UL1TRR002535 to the Colorado Clinical and Translational Sciences Institute (CCTSI) to R.A.S., C.S., and L.M.

DISCLAIMERS

The views expressed in this article do not communicate an official position of the Alpha-1 Foundation. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.M., K.A.P., Q.L., R.A.S., L.M., C.S., R.P.B., and K.A.S. conceived and designed research; A.M., K.N., F.G., S.W., Q.L., I.E., B.W., D.C., M.J.J., R.A.S., L.M., I.P., and K.A.S. performed experiments; A.M., K.N., F.G., K.A.P., S.W., Q.L., I.E., B.W., D.C., M.J.J., R.A.S., L.M., I.P., and K.A.S. analyzed data; A.M., K.N., F.G., I.E., D.C. M.J.J., L.M., R.P.B., and K.A.S. interpreted results of experiments; A.M., I.E., and K.A.S. prepared figures; A.M., K.N., I.E., I.P., and K.A.S. drafted manuscript; A.M., K.N., F.G., K.A.P., I.E., B.W., D.C., C.S., I.P., and K.A.S. edited and revised manuscript; A.M., K.N., F.G., K.A.P., S.W., Q.L., I.E., B.W., D.C., M.J.J., R.A.S., L.M., C.S., R.P.B., I.P., and K.A.S. approved final version of manuscript.

APPENDIX

COPDGene Study Disclaimer and Acknowledgments

COPDGene phase 3.

Grant support and disclaimer.

The project described was supported by Award No. U01 HL089897 and Award No. U01 HL089856 from the National Heart, Lung, and Blood Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health.

COPD Foundation funding.

COPDGene is also supported by the COPD Foundation through contributions made to an Industry Advisory Board that has included AstraZeneca, Bayer Pharmaceuticals, Boehringer-Ingelheim, Genentech, GlaxoSmithKline, Novartis, Pfizer, and Sunovion.

COPDGene Investigators: core units.

Administrative Center: James D. Crapo, MD (PI); Edwin K. Silverman, MD, PhD (PI); Barry J.Make, MD; Elizabeth A. Regan, MD, PhD

Genetic Analysis Center: Terri H. Beaty, PhD; Peter J. Castaldi, MD, MSc; Michael H. Cho, MD, MPH; Dawn L. DeMeo, MD, MPH; Adel El Boueiz, MD, MMSc; Marilyn G. Foreman, MD, MS; Auyon Ghosh, MD; Lystra P. Hayden, MD, MMSc; Craig P. Hersh, MD, MPH;Jacqueline Hetmanski, MS; Brian D. Hobbs, MD, MMSc; John E. Hokanson, MPH, PhD; Wonji

Kim, PhD; Nan Laird, PhD; Christoph Lange, PhD; Sharon M. Lutz, PhD; Merry-Lynn McDonald, PhD; Dmitry Prokopenko, PhD; Matthew Moll, MD, MPH; Jarrett Morrow, PhD; Dandi Qiao, PhD; Elizabeth A. Regan, MD, PhD; Aabida Saferali, PhD; Phuwanat Sakornsakolpat, MD; Edwin K. Silverman, MD, PhD; Emily S. Wan, MD; Jeong Yun, MD, MPH

Imaging Center: Juan Pablo Centeno; Jean-Paul Charbonnier, PhD; Harvey O. Coxson, PhD;Craig J. Galban, PhD; MeiLan K. Han, MD, MS; Eric A. Hoffman, Stephen Humphries, PhD;Francine L. Jacobson, MD, MPH; Philip F. Judy, PhD; Ella A. Kazerooni, MD; Alex Kluiber;David A. Lynch, MB; Pietro Nardelli, PhD; John D. Newell, Jr., MD; Aleena Notary; Andrea Oh, MD; Elizabeth A. Regan, MD, PhD; James C. Ross, PhD; Raul San Jose Estepar, PhD; Joyce Schroeder, MD; Jered Sieren; Berend C. Stoel, PhD; Juerg Tschirren, PhD; Edwin Van Beek, MD, PhD; Bram van Ginneken, PhD; Eva van Rikxoort, PhD; Gonzalo Vegas Sanchez-Ferrero, PhD; Lucas Veitel; George R. Washko, MD; Carla G. Wilson, MS;

PFT QA Center, Salt Lake City, UT: Robert Jensen, PhD

Data Coordinating Center and Biostatistics, National Jewish Health, Denver, CO: Douglas Everett, PhD; Jim Crooks, PhD; Katherine Pratte, PhD; Matt Strand, PhD; Carla G. Wilson, MS

Epidemiology Core, University of Colorado Anschutz Medical Campus, Aurora, CO: John E.Hokanson, MPH, PhD; Erin Austin, PhD; Gregory Kinney, MPH, PhD; Sharon M. Lutz, PhD;Kendra A. Young, PhD

Mortality Adjudication Core: Surya P. Bhatt, MD; Jessica Bon, MD; Alejandro A. Diaz, MD, MPH; MeiLan K. Han, MD, MS; Barry Make, MD; Susan Murray, ScD; Elizabeth Regan, MD;Xavier Soler, MD; Carla G. Wilson, MS

Biomarker Core: Russell P. Bowler, MD, PhD; Katerina Kechris, PhD; Farnoush Banaei-Kashani, PhD

COPDGene® Investigators – Clinical Centers Ann Arbor VA: Jeffrey L. Curtis, MD; Perry G. Pernicano, MD Baylor College of Medicine, Houston, TX: Nicola Hanania, MD, MS; Mustafa Atik, MD; Aladin Boriek, PhD; Kalpatha Guntupalli, MD; Elizabeth Guy, MD; Amit Parulekar, MD;

Brigham and Women’s Hospital, Boston, MA: Dawn L. DeMeo, MD, MPH; Craig Hersh, MD, MPH; Francine L. Jacobson, MD, MPH; George Washko, MD

Columbia University, New York, NY: R. Graham Barr, MD, DrPH; John Austin, MD; Belinda D’Souza, MD; Byron Thomashow, MD

Duke University Medical Center, Durham, NC: Neil MacIntyre, Jr., MD; H. Page McAdams, MD; Lacey Washington, MD

HealthPartners Research Institute, Minneapolis, MN: Charlene McEvoy, MD, MPH; Joseph Tashjian, MD

Johns Hopkins University, Baltimore, MD: Robert Wise, MD; Robert Brown, MD; Nadia N.Hansel, MD, MPH; Karen Horton, MD; Allison Lambert, MD, MHS; Nirupama Putcha, MD, MHS

Lundquist Institute for Biomedical Innovation at Harbor UCLA Medical Center, Torrance, CA:Richard Casaburi, PhD, MD; Alessandra Adami, PhD; Matthew Budoff, MD; Hans Fischer, MD; Janos Porszasz, MD, PhD; Harry Rossiter, PhD; William Stringer, MD Michael E. DeBakey VAMC, Houston, TX: Amir Sharafkhaneh, MD, PhD; Charlie Lan, DO

Minneapolis VA: Christine Wendt, MD; Brian Bell, MD; Ken M. Kunisaki, MD, MS

Morehouse School of Medicine, Atlanta, GA: Eric L. Flenaugh, MD; Hirut Gebrekristos, PhD;Mario Ponce, MD; Silanath Terpenning, MD; Gloria Westney, MD, MS

National Jewish Health, Denver, CO: Russell Bowler, MD, PhD; David A. Lynch, MB

Reliant Medical Group, Worcester, MA: Richard Rosiello, MD; David Pace, MD

Temple University, Philadelphia, PA: Gerard Criner, MD; David Ciccolella, MD; Francis Cordova, MD; Chandra Dass, MD; Gilbert D’Alonzo, DO; Parag Desai, MD; Michael Jacobs, PharmD; Steven Kelsen, MD, PhD; Victor Kim, MD; A. James Mamary, MD; Nathaniel Marchetti, DO; Aditi Satti, MD; Kartik Shenoy, MD; Robert M. Steiner, MD; Alex Swift, MD;Irene Swift, MD; Maria Elena Vega-Sanchez, MD

University of Alabama, Birmingham, AL: Mark Dransfield, MD; William Bailey, MD; Surya P.Bhatt, MD; Anand Iyer, MD; Hrudaya Nath, MD; J. Michael Wells, MD

University of California, San Diego, CA: Douglas Conrad, MD; Xavier Soler, MD, PhD; Andrew

Yen, MD

University of Iowa, Iowa City, IA: Alejandro P. Comellas, MD; Karin F. Hoth, PhD; John Newell, Jr., MD; Brad Thompson, MD

University of Michigan, Ann Arbor, MI: MeiLan K. Han, MD MS; Ella Kazerooni, MD MS;Wassim Labaki, MD MS; Craig Galban, PhD; Dharshan Vummidi, MD

University of Minnesota, Minneapolis, MN: Joanne Billings, MD; Abbie Begnaud, MD; Tadashi

Allen, MD

University of Pittsburgh, Pittsburgh, PA: Frank Sciurba, MD; Jessica Bon, MD; Divay Chandra,

MD, MSc; Joel Weissfeld, MD, MPH

University of Texas Health, San Antonio, San Antonio, TX: Antonio Anzueto, MD; Sandra Adams, MD; Diego Maselli-Caceres, MD; Mario E. Ruiz, MD; Harjinder Singh