Introduction
E-cigarettes (e-cigs) are often considered as a safer alternative to combustible cigarettes, as well as a method for quitting traditional cigarette smoking [
1,
2]. The e-cig delivery system is based on tanks or cartridges that are loaded with e-cig liquid (e-liquid), which is then aerosolized, with inhalation delivering the aerosol to the lungs. Usually, e-liquids are composed of propylene glycol (PG) and/or vegetable glycerin (VG), with varying concentrations of nicotine (up to 100 mg/mL). Flavoring chemicals are added as additives in e-liquids and pods that enhance the taste and reduce throat hit, however they tend to attract the younger generation unexpectedly [
3,
4]. While PG and VG are considered safe in food-grade products, adverse health effects have been reported from these substances when they serve as e-cig vehicles [
5]. Previous studies have shown that the amount of nicotine delivered by e-cigs is much higher than the amount delivered via cigarette smoke [
6,
7]. We have shown that inflammation and extracellular matrix (ECM) remodeling/dysregulated repair are altered by acute exposure to e-cigs, with or without nicotine [
8]. Considering that e-cig vaping is often a long-term habit, research on the chronic effects of e-cig aerosol exposure is essential to understand the mechanism resulting in augmented inflammatory responses and ECM remodeling, which are fundamental changes that occur during early stages of most chronic lung diseases, such as idiopathic pulmonary fibrosis (IPF) and chronic obstructive pulmonary disease (COPD) [
9,
10].
In our previous studies, acute exposure to e-cig aerosol containing flavoring chemicals was shown to cause lung inflammation, oxidative stress, and dysregulated repair [
8,
11]. Further, nicotine exposure could cause suppression of immune response resulting in augmented lung inflammation and injury following viral infection [
12]. Nicotinic Acetylcholine Receptors (nAChRs) are largely responsible for activation of acetylcholine neurotransmitter signaling pathways in the central nervous system (CNS) [
13,
14]. It has been postulated that angiotensin-converting enzyme 2(ACE2), a Covid-19 receptor, is regulated by nAChRs. The nAChRs are widely distributed in the CNS, so they can easily be activated by nicotine, initiating and reinforcing a rewarding feedback loop which might induce nicotine addiction [
15,
16]. Interestingly, lung nAChR activation could help inhibit the inflammation caused by lipopolysaccharides (LPSs) and the receptor knockouts (KOs) that promote inflammatory processes [
17,
18]. However, as one of the nAChR-agonists, nicotine could initiate receptor-related pathways, and nicotine aerosol inhalation might also induce inflammation [
11] and regulate ACE2 receptor via nAChR. Other chemicals may induce activation of nAChRs, and they may also trigger feedback-loops similar to flavoring chemicals [
19‐
21]. However, there are limited studies that directly describe the role of dysregulated repair induced by e-cig vapors with or without nicotine in a nAChRα7-dependent manner.
A common vehicle used for nicotine delivery in e-cigs is PG and/or VG. PG (C
3H
8O
2) is a colorless and odorless organic solvent that is an FDA-approved, food-grade chemical generally recognized as safe (GRAS). This is one of the reasons why people originally thought to use PG as a nicotine carrier in e-liquids. Previous study has shown that PG aerosols produce different byproducts, some of which might lead to cancer [
22]. Many e-cigs with or without nicotine are commercially available, but the understanding of respiratory health risks of e-cig inhalation are poorly investigated.
In this study, we hypothesized that e-cig aerosol induces lung inflammation and dysregulated repair/ECM remodeling in a nAChR α7-dependent manner. Our results suggest that lung inflammatory responses and dysregulated repair/ECM remodeling induced by e-cig aerosol containing nicotine could potentially be related to the nAChR α7 signaling pathway. On the other hand, PG-induced lung dysregulated repair and inflammatory response occurs in a nAChR α7-independent manner.
Methods
Animals
Adult C57BL/6 J (WT) mice and α7 nicotinic acetylcholine receptor knockout (nAChR α7 KO) mice were both purchased from Jackson Laboratory, weighing 25–35 g and aged 3–4 months old. The nAChR α7 CreCC10 mice (clara/club-cell-specific nAChR α7 deletion) were generated by crossing nAChR α7 floxed mice (nAChR α7 floxed mutant) from Jackson Laboratory (donated by Dr. Jerry Yakel, NIEHS/NIH) with mice having the Cre recombinase transgene controlled by the CC10 promoter (C57BL/6 J; from TJ Mariani, University of Rochester, Rochester, NY). Prior to e-cig exposure, mice were housed in the inhalation core facility at the University of Rochester for one week. All experiments performed in this study were in compliance with the standards set by the United States Animal Welfare Act. The Animal Research Committee (UCAR) approved the animal protocol at the University of Rochester Medical Center, Rochester, NY.
Blood gas and exercise ability measurement
Blood gas, including pH, pressure of CO
2 and O
2, concentration of HCO
3, TCO
2, glucose (Glu) and hemoglobin (Hb), percentage of O
2 and hematocrit (Hct) were analyzed by i-STAT system (i-STAT CG8+ cartridge, Cat# 03P88–25; Vet scan i-STAT 1 analyzer; Abaxis Global Diagnostics), as described previously [
23]. Exercise capacity measurement was done using a motorized animal treadmill (Columbus Instruments) as described previously [
24]. Running distance (meters) and running time (minutes) were recorded to present exercise tolerance. Mice stayed on the treadmill for 5 min before start of measurement. The treadmill began at 8.5 m/min with 0° incline for 9 min. Then, the speed was increased to 10 m/min with 5° incline for 3 min. After that, the speed was increased to 2.5 m/min every 3 min, and incline was increased to 5° every 9 min until the mouse became exhausted, judged by observation of failure of running and continuous contact with the electric grid. Following this exercise, blood was collected by submandibular venipuncture for blood gas and cotinine measurement.
E-cig device and e-liquid
The e-cig device used in this study was purchased from the Joytech VTC mini (SciReq, Montreal, Canada). The atomizer/coil (0.15 Ω) used for aerosolization of e-liquid was purchased from Kanger technology (Shenzhen, China). All other components of the e-cig exposure chamber were purchased from SCIREQ, and all the components were cleaned after exposure every day. The atomizer/coil was replaced on a weekly basis to avoid overheating and generation of carbon monoxide. The e-liquids used in this study that contain PG alone and PG with nicotine (25 mg/mL), were purchased from
xtremevaping.com.
E-cig exposure
The e-cig exposure performed here has been described in our previous studies [
25]. Briefly, the in vivo e-cig exposure was setup inside a fume hood and based on the SCIREQ InExpose e-cig extension smoking system. The e-cig exposure puffing profile used was based on realistic topographical data from e-cig users, with 3.3 s/puff, 2 puffs/min and a 70 mL puff volume [
26]. The aerosolization of e-liquid was performed by using a 3rd generation e-cig device (Joytech eVIC VTCmini), which was controlled by the SCIREQ flexiware software (V8.0). The whole-body exposure was done for a total of 2 h/day, 5 days/week, for 30 days. During e-cig exposures, temperature, humidity, oxygen and carbon dioxide percentages were monitored along with carbon monoxide levels in the chamber. The e-cig aerosol generated was passed through the condensing chamber and pumped into the mixing chamber with a flowrate of 1.0 L/min. The vapor was diluted with air in the mixing chamber and then delivered into the whole-body exposure chamber where mice were separated by dividers. Simultaneously, the e-cig aerosol in the exposure chamber was exhausted by another pump with a flowrate of 2.0 L/min. Both the pumps were calibrated and adjusted each time before exposure. Pumps were cleaned at the end of each exposure to minimize the effects of nicotine residues. Mice were divided into air (control), PG, and PG with nicotine groups, each with an equal number of males and females, for both the WT and nAChR α7 KO conditions. Air group mice stayed in the inhalation facility in a similar environment during the 30 day exposure. Serum cotinine levels measured by ELISA (Calbiotech) were ~ 500 ng/mL for the PG with nicotine group, and ~ 20 ng/mL for the PG only group, due to nicotine residue in the pumps and exposure chambers which was difficult to fully clean
(Additional File
1: Figure S1). As expected, the air group showed no cotinine.
Bronchoalveolar lavage fluid (BALF)
Mice were euthanized with Ketamine/Xylazine 24 h after the final exposure. Their tracheas were cannulated and their lungs were lavaged three times with 0.6 mL saline with 1% FBS (1.8 mL total). The recovered fluids were collected and spun down at 1000 x g for 10 min at 4 °C for harvesting of BALF cells. The supernatant was stored at − 80 °C for future analysis. The BALF cells were re-suspended in 1.0 mL saline with 1% FBS and stained with acridine orange propidium iodide (AO/PI). Total cell counts per mL were measured from AO/PI stained cells via cellometer.
Inflammatory cell count
The resuspended BALF cells were used for immune-inflammatory cell counts with cell-type-specific labelled monoclonal antibodies. Total 1.0 × 105 BALF cells were used for antibody labeling. Before antibody staining, all cells were blocked with purified anti-mouse CD16/32 (Cat# 50–163-432, Fisher Scientific) to prevent non-specific binding, and washed with PBS once. The cells were stained with F4/80 PE-conjugated antibody for macrophages (Cat# 123109, BioLegend), LY6B.2 Alexa fluor488-conjugated antibody for neutrophils (Cat# NBP213077AF488, Novus Biologicals), PE-Cyanine7 antibody for CD4a+ T-lymphocytes (Cat# 25–0041-82, Fisher Scientific), and APC conjugated Monoclonal Antibody for CD8a+ T-Lymphocytes (Cat# 17–0081-82, Fisher Scientific). The absolute cell numbers of macrophages, neutrophils, and CD4a+/CD8a+ T-lymphocytes were determined by multiplying the percentage of cells by the total cell counts. Flow cytometry was performed using the Guava® easyCyte™ flow cytometer (Millipore Sigma) and analyzed using Guava® InCyte™ software.
Measurement of pro-inflammatory cytokines by Luminex in BALF
To measure the pro-inflammatory cytokines present in BALF, a Bio-Plex Pro mouse cytokine 23-plex immunoassay kit (Cat#: M60009RDPD, BioRad) was used, according to the manufacturer’s instructions. Briefly, the diluted magnetic beads were placed into the assay plate and rinsed with wash buffer. The BALF samples and standards were then added into the wells, and shaken at 850 rpm at room temperature for 30 min. After sample incubation, the plates were washed with wash buffer 3 times, then the detection antibody was added and the plates incubated for 30 min at room temperature, shaking at 850 rpm. Next, plates were washed with wash buffer 3 times, and SA-PE was added for 10 min at room temperature, shaking at 850 rpm. After this step, the plates were washed 3 times with wash buffer, and the beads were resuspended in assay buffer for reading. Results are determined via the Luminex flexmap 3d (Luminex Corp).
Protein isolation
Lung tissues harvested from the animals during sacrifice were stored at − 80 °C for future analysis. Frozen lungs were homogenized mechanically in RIPA buffer with protease inhibitor (Cat#: 78440, ThermoFisher Scientific). After homogenization, lysates were kept on ice for 45 min, followed by centrifugation at 15,000 x g for 30 min at 4 °C. The supernatant was collected and protein concentration was quantified using the Pierce BCA Protein Assay Kit (Cat#: 23227, ThermoFisher Scientific), based on the manufacturer’s protocol.
Western blot
The protein samples from lung homogenates were running through SDS-polyacrylamide electrophoresis gels (SDS-PAGE: 10%) with 20 μg of protein in each lane. After electrophoresis, the gel was transferred onto a nitrocellulose membrane (Cat# 1620112, BioRad). The membrane was washed with tris-buffered saline containing 0.1% Tween 20 (TBS-T) for 15 min, then blocked with 5% non-fat dry milk for 1 h at room temperature. Then, the membranes were probed with primary antibodies 16 h at 4 °C: anti-MMP2 (1:1000, ab92536, Abcam); anti-MMP9 (1:1000, ab38898, Abcam); anti-MMP8 (1:1000, ab81286, Abcam); anti-MMP12 (1:1000, NBP2–67344, Novus Biologicals); anti-Collagen 1α1 (1:1000, NBP1–30054, Novus Biologicals); anti-Collagen 1α2 (1:500, NBP1–57987, Novus Biologicals); anti-fibronectin (1:2000, ab45688, Abcam); and anti-PAI-1 (1:1000, ab182973, Abcam),anti-ACE2 (1:1000, ab15348, Abcam), and the appropriate secondary antibody (Goat-Anti-Rabbit, 1:10000, Cat# 1706515, BioRad) for 1 h at room temperature. After each antibody incubation, membrane was washed 4 times in TBS-T at room temperature, 15 min per wash. The luminescence signals were developed using chemiluminescence substrate (Perkin Elmer, Waltham, MA). The membrane exposure and band intensities were detected via the Bio-Rad ChemiDoc MP imaging system (Bio-Rad Laboratories, Hercules, CA, USA). Protein quantification was done by densitometry analysis, and β-actin (1:2500, ab20272, Abcam) was used as housekeeping control.
RNA isolation and Nanostring quantification
Frozen lung tissues from − 80 °C were used for RNA isolation. Lung tissue (~ 100 mg per sample was homogenized in Trizol for 20s. The RNA was isolated using the Direct-zol™ RNA Miniprep Plus assay kit (Cat# R2073, Zymo Research). Isolated RNA was quantified and checked for quality by spectrophotometer (ND-1000, NanoDrop Technologies, Wilmington, DE, USA). RNA samples were aliquoted and sent out to Nanostring facility for analysis of genes using nCounter Mouse Myeloid Innate Immunity v2 Panel (NanoString Technologies, Inc. Cat# XT-CSO-MMII2–12).
Statistical analysis
The statistical analysis was done by either one-way ANOVA or student’s t-test, followed by Tukey’s multiple comparisons in GraphPad Prism software (Version 8.0, La Jolla, CA). Results were presented as the mean ± SEM. P < 0.05 was considered statistically significant.
Discussion
The popularity of e-cig vaping has been rising recently in the United States, and hence health concerns about vaping have recently attracted public attention [
27]. Our previous studies have shown that acute exposure to e-cigarettes could cause inflammatory responses, oxidative stress, and ECM remodeling [
8,
11,
28]. Although in the past we have shown the health risks caused by acute e-cig exposures in mice [
8,
11], sub-chronic or chronic effects of e-cig exposure-induced toxicity and respiratory health effects remain elusive. Based on our knowledge, no study has yet shown that long-term exposure of e-cig aerosol can result in persistent dysregulated repair/ECM remodeling and inflammatory responses implicating the role of the nAChRα7 receptor in the lung.
Similar to our acute exposure results, e-cig aerosol with nicotine induced inflammatory cell counts in BALF compared to air control, especially in macrophages and CD4a
+/CD8a
+ T-lymphocytes. Previous report has demonstrated that two weeks of e-cig (with nicotine) exposure increased the number of macrophages but not neutrophils, which supports our findings [
29]. Another study reported that acute exposure of e-cig aerosol increased total cell and macrophage counts; and even the vehicle exposure alone showed a trend towards increase in total cells and macrophages [
30]. Based on our results, e-cig exposure with or without nicotine causes increased inflammatory cellular influx into the lungs (BALF). PG alone (vehicle) exposure was considered safe by e-cig users, but our data reveals that sub-chronic e-cig exposure with PG (without nicotine) induced lung inflammation. The genetic ablation of nAChRα7 protects against e-cig exposure induced increase in macrophages and CD4a
+/CD8a
+ T-lymphocytes. Probably, e-cig exposure containing PG with nicotine causes lung inflammation/remodeling response via the nAChRα7 receptor mediated signaling pathway, and blocking the receptor could attenuate nicotine-dependent effects in the lungs. Prior reports show that activation of nAChR can also inhibit inflammation [
31], and we have observed increased neutrophils and CD4a
+/CD8a
+ T-lymphocytes when exposed to PG alone in nAChRα7 KO mice. The nAChR deletion may promote inflammation independent of nicotine exposure in the lungs. The chemicals derived from e-cig aerosols containing PG, such as formaldehyde, acetaldehyde, and methylglyoxal, in nicotine-free aerosol [
32] may result in increased inflammation. As reported previously, both e-cig with or without nicotine caused infiltration of inflammatory cells and pro-inflammatory mediators that are key players for causing lung inflammatory response [
8,
11,
30]. As an agonist of nAChRα7, nicotine might be both pro-inflammatory and anti-inflammatory roles, and further research is needed to study the relationship between nicotine and nicotinic receptors in the context of sub-chronic e-cig exposure.
Apart from the inflammatory cells, some pro-inflammatory mediators in BALF were significantly increased following e-cig exposure with nicotine. Specific macrophage-mediated pro-inflammatory cytokines such as IL-1α, MCP-1, and GM-CSF, were increased in BALF after PG with nicotine exposure in a nAChRα7-dependent manner. However, a few other macrophage-driven cytokines, IL-1β, TNF-α, and MIP-1β, did not show nAChRα7 dependency. Our results paralleled reports that e-cig aerosol containing nicotine induced cytotoxicity in alveolar macrophage and significant release of inflammatory mediators (TNF-α, CXCL-8, MCP-1, and IL-6) [
33]. Additional studies have shown specific T-lymphocyte-related cytokines/chemokines (IL-2, IL-9, IFN-γ, and RANTES) and their role in inflammatory signaling [
34‐
37], and our results indicated that these altered cytokine responses were nAChR-dependent in e-cig exposed mouse lungs. Mice exposed to PG with nicotine showed increased IL-5 release in a nAChR independent manner. Another report shows increased IL-4, IL-5, and IL-13 levels following e-cig containing nicotine when challenged to ovalbumin treated mice resulting in exacerbation of allergic airway inflammation [
38]. The increased IFN-γ production has been related to cytotoxicity and infiltration of CD8a
+ T-cells [
39,
40]. This altered pro-inflammatory cytokine response and cellular influx in the lung are inter-connected to each other partly due to nAChRα7 mediated signaling. It is possible that nAChRα7 deletion could attenuate the inflammatory responses induced by e-cig aerosol contained nicotine in the lungs. However, PG-induced increase in CD4a
+/CD8a
+ T-lymphocytes does not corroborate with the cytokine levels. Future studies will address immune cell-type specific role in pulmonary toxicity of e-cig with and without nicotine during long-term exposure in mice.
We measured the mRNA expression using the mouse myeloid innate immunity panel from Nanostring technology to screen the potentially affected target genes. Our results showed altered expression of SKIL and LDLR in PG with nicotine-exposed WT mice, but the same genes remain unaffected in nAChRα7 KO exposed groups. These target genes have been shown to play vital roles in TGFβ/SMAD canonical signaling and TGFβ-induced epithelial–mesenchymal transition (EMT) may be inhibited by SKIL binding to SMAD [
41,
42] In this study, we found increased protein abundance of PAI-1 in WT mice exposed to PG with nicotine, but PAI-1 was further reduced in nAChRα7 KO mice. Our results suggest that e-cig exposure of PG with nicotine may affect SKIL-TGFβ-PAI-1 axis in the lung possibly through nAChRα7 activation. The other important target significantly altered by PG with nicotine is LDLR gene. Overexpression of LDLR is capable of promoting macrophage differentiation and exaggerated inflammatory response, and silencing LDLR results in reduced inflammatory responses in THP-1 cells [
43]. In our study, LDLR was reduced in PG with nicotine exposed WT mice, while LDLR transcript levels remain unchanged in both PG with and without nicotine exposed nAChRα7 KO groups. These findings suggest that nAChRα7 may be an important gateway that affects specific inflammatory response/ECM remodeling target genes during sub-chronic e-cig exposure with nicotine.
To further study the inflammatory responses, we measured the protein abundance of NF-κB (p50/p105) in both female and male mice. The activation of NF-κB is well documented in different lung injury models [
44] including cigarette smoke-induced inflammation, wheras research on e-cig aerosol induced NF-κB activation remains less explored [
45‐
48]. Although the anti-inflammatory effects (i.e. inhibition of NF-κB signaling) via activation of nAChRs is well-known and has been studied previously, nicotine has been shown induce inflammation through activation of the NF-κB signaling pathway [
31,
49]. Our findings suggest that e-cig aerosol containing nicotine increased p50 and p105 protein abundance in female mice in a nAChRα7-dependent manner. We believe nicotine is a potent inducer of altered lung inflammation via activated nAChRα7 compared to the anti-inflammatory effects regulated by nAChR-related signaling. We found lack of nAChRα7 showing attenuation of inflammatory response following e-cig exposure with nicotine. However, nicotine-free e-cig exposure increased p105 protein levels, which suggests this subunit may act through an alternative pathway that may be independent of nAChRα7 signaling. We for the first time show that lung inflammation caused by e-cig aerosol containing nicotine is nAChRα7-dependent, probably in a sex-specific manner. However, more detailed studies are needed to further illustrate the exact mechanism by which sex-differences in lung inflammatory responses occur following e-cig exposure through cell-type specific nAChRα7-mediated signaling.
As we recently reported, acute exposure to e-cig aerosol with or without nicotine resulted in dysregulated repair and ECM remodeling [
8]. We noticed similar effects in our sub-chronic exposure model as well. E-cig aerosol with or without nicotine alters MMPs and ECM remodeling proteins in a sex-dependent manner, which are nAChRα7 receptor independent. It is known that nicotine exposure causes harmful respiratory effects, such as airway remodeling and airspace enlargement [
50,
51], which are key features of emphysema and COPD [
51‐
53]. However, dysregulation of MMPs serves not only as a factor for ECM remodeling, but also as a marker of inflammation mediated by dysregulated macrophages. Based on our results, nicotine-free e-cig aerosol induced augmented MMPs dysregulation more so than e-cig aerosol with nicotine. Increased MMP2/MMP12, along with decreased MMP9/MMP8, might suggest a compensatory feedback loop [
54]. We observed that ACE2 is increased by e-cig with nicotine exposure, whereas the level is decreased in nAChRα7-deficient mice, suggesting that ACE2 upregulation by nicotine is mediated by nAChRα7. It is thought that ACE2 is upregulated by smoking, and that its expression level is increased in smokers and COPD patients [
55,
56], as well as serving as the entry gate for the novel coronavirus (SARS-CoV-2) [
57]. Hence, e-cig with nicotine inhalation would promote SARS-CoV-2 infection, and nAChRα7 deletion and downregulation of ACE2 may have some ramifications on viral infection.
One of the major substrates of MMPs is ECM related proteins. From our previous study, mice exposed to e-cig aerosols for 3 days showed dysregulated repair/ECM remodeling both at the mRNA and protein levels in a sex-dependent manner [
8]. The current study shows that sub-chronic e-cig exposure with or without nicotine affects ECM remodeling to a certain extent dependent on nAChRα7 with some sex-specific phenotypes. We have observed increased PAI-1 only in WT male mice exposed to PG with nicotine in a nAChRα7-dependent manner, and PAI-1 is a primary ECM regulator and pro-fibrotic marker [
58]. Therefore, increased PAI-1 following inhalation of e-cig aerosol with nicotine indicates that e-cig vapor containing nicotine could increase risk of chronic fibrotic disease, implicating the crucial role of nAChRα7-related signaling. Some other ECM proteins, such as type 1 collagen and fibronectin, were altered after exposure of e-cig aerosol with or without nicotine. ECM proteins affected by e-cig exposure are associated with dysregulated wound healing and fibrotic disease [
59]. The observed increase in COL1A2 following e-cig with PG alone exposure in both male and female mice demonstrates the significant health risk imposed by PG alone, in addition to the risks associated with PG with nicotine. As mentioned above, nicotine might not be the only target that can activate nAChRα7, and PG alone via other indirect mechanisms could activate nicotinic receptors, which may be in agreement with our previous studies [
8,
19]. It is well known that collagen and fibronectin are required during the wound healing process and delayed wound healing is observed when epithelial cells were unable to synthesize fibronectin [
60]. We found down-regulation of collagen and fibronectin following e-cig exposure, which will make lungs more vulnerable later in life when exposed to other noxious gases and environmental toxicants. Further studies are required to understand the mechanisms underlying how e-cig exposures may induce damaging responses, leading to lung disease and injury.
Based on our findings, long-term studies are needed to provide further insights on how nAChRα7 dependency plays an important cell-type specific role in the lung (epithelial and fibroblast) during e-cig exposure, and its involvement in ACE2 regulation. We know that several factors, such as duration of exposure, route of exposure, dose, e-liquid composition, setting used for delivering the e-cig aerosols, concentration of PG mixture and nicotine, as well as mouse strain, background, and sex of mice may equally contribute to the difference and phenotype observed in the measure outcomes. Chronic e-cig exposures are underway that will help us to understand how these signaling mechanisms are altered by e-cig exposure with and without nicotine, and how these changes may contribute to the respiratory toxicity observed in mice, which could translate to clinical relevance in human e-cig users. Ongoing human studies are being carried out to identify potential targets and toxicity biomarkers that may relate to human respiratory health effects of e-cigarettes.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.