Proteomic analysis of sputum in patients with active pulmonary tuberculosis


  • Y. R. Fu,

    1. ) Department of Medical Microbiology of Weifang Medical University
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Z. J. Yi,

    1. ) Clinical Faculty (Affiliated Hospital), Department of Laboratory Medicine, Key Laboratory of Clinical Laboratory Diagnostics in Universities of Shandong, Weifang Medical University, Weifang
    Search for more papers by this author
    • These authors contributed equally to this work.

  • S. Z. Guan,

    1. ) Clinical Faculty (Affiliated Hospital), Department of Laboratory Medicine, Key Laboratory of Clinical Laboratory Diagnostics in Universities of Shandong, Weifang Medical University, Weifang
    Search for more papers by this author
  • S. Y. Zhang,

    1. ) Department of Laboratory Medicine, Chest Speciality Hospital of Weifang, Weifang, China
    Search for more papers by this author
  • M. Li

    1. ) Clinical Faculty (Affiliated Hospital), Department of Laboratory Medicine, Key Laboratory of Clinical Laboratory Diagnostics in Universities of Shandong, Weifang Medical University, Weifang
    Search for more papers by this author

Corresponding author: Z. J. Yi, No. 465, Yuhe Road, Kuiwen District, Faculty of Laboratory Medicine of Affiliated Hospital of Weifang Medical University, Weifang 261031, Shandong Province, China


Clin Microbiol Infect 2012; 18: 1241–1247


The protein composition of sputum most faithfully reflects the state of the lungs. The aim of this study was to determine whether relative qualitative and quantitative differences in protein expression of sputum could be related to active pulmonary tuberculosis. Sputum samples were collected from 65 patients with active pulmonary tuberculosis and 38 healthy controls. Comprehensive proteomic approaches were used to profile the proteome changes of host sputum in response to Mycobacterium tuberculosis infection using two-dimensional electrophoresis in combination with matrix-assisted laser desorption ionization time-of-flight/time-of-flight mass spectrometry. Mascot software was used to identify proteins from protein databases. Enzyme-linked immunosorbent assay was used to confirm the proteomic results. A total of 62 differentially expressed proteins were identified, among which, 15 proteins were up-regulated and 47 proteins were down-regulated in the tuberculosis sputum compared with the controls. Bacterial protein UqhC was the most increased protein, whereas serum albumin was the most decreased protein in the tuberculosis sputum compared with the controls. The enzyme-linked immunosorbent assay analysis was consistent with proteomic data. Bioinformatics analysis suggested that multiple host cell pathways were involved in the tuberculosis infection processes, including acute phase response, signal transduction, cytoskeleton structure, immune response and so on. In all, for the first time, our results revealed that a number of proteins were differentially expressed during active pulmonary tuberculosis infection. These data will provide valuable clues for further investigation of tuberculosis pathogenesis and biomarkers for detection of active pulmonary tuberculosis infection.


In developing countries, tuberculosis (TB) is still a common disease [1]. Investigations into the molecular pathogenesis of TB have not provided adequate answers to the basic question concerning the mechanisms of this disorder. Newly available proteomic procedures are promising methods for TB research, because specific patterns of protein expression occurring in biofluids and tissue may be of great potential relevance to pathogenesis and biomarkers of disease [2,3].

The protein composition of sputum most faithfully reflects the state of the lungs and sputum can be easily collected by non-invasive means [4,5]. However, up to now, little is known about the effect of pulmonary TB infection on sputum protein composition. Considering the important role of proteins in assessment of lung diseases, we proposed that protein expression was altered in cell-free sputum of patients with active pulmonary TB. Therefore, the study attempted to reveal protein profiles in sputum associated with active pulmonary TB, and to explore the potential biological functions of identified candidate proteins.


Eligibility for the 65 patients who entered into the study included typical symptoms of pulmonary TB: fibrocavitary lung infiltrate on chest radiograph, at least one sputum specimen staining positive and/or sputum culture positive. Patients were excluded from this study if they had any other coexisting acute or chronic illnesses. Also, 32 healthy age and sex-matched subjects were recruited as controls (Table 1).

Table 1.   Characteristics of participants
CharacteristicsTB (n = 65)Control (n = 38)
  1. Thirty-nine patients were M. tuberculosis positive for both sputum smear and culture, and 26 were sputum smear negative but culture positive. All patients had the clinical signs and symptoms of active pulmonary TB, comprising, 82.8% cough, 60.7% fever, 58.2% weight loss, 40.4% haemoptysis and 38.9% night sweats. Healthy controls were free of TB infection and any clinical symptoms of any infectious disease.

Age, mean (range) years46.11 ± 23.98 (15–60)42.53 ± 19.16 (17–55)

The study was performed with the approval of Weifang Medical University local ethics committee and carried out in compliance with the Helsinki Declaration. Informed consents were obtained from all the subjects before the commencement of the study.

For sputum collection, early morning sputum was collected before starting chemotherapy, as described in previous reports [6,7]. Sputum was homogenized with an equal volume of dithiothreitol (DTT, final concentration 10 mM) at 4°C and centrifuged to yield cell-free supernatant and a cell pellet within 1 h of collection. To control for saliva contamination, the cell pellet was stained with Wright stain and used for differential cell analysis. A sputum sample was considered adequate for further analysis if it contained fewer than 80% squamous cells (Fig. 1). For adequate sputum, cell-free supernatant was aliquoted and stored immediately in liquid nitrogen until analysis.

Figure 1.

 Protein patterns of sputum. (a) A representative TB example is shown. (b) A representative control sample is shown. The resulting two samples containing 150 μg protein were added to pH 3–10, non-linear DryStrips (24 cm long; Bio-Rad, Hercules, CA, USA). The p values for these spots in three independent experiments were <0.05.

For protein preparation, the supernatant samples of sputum were desalted by Desalting Columns (Pierce, Appleton, WI, USA) and were resuspended in solubilization buffer (5 M urea, 2 M thiourea, 2% CHAPS, 65 mM DTT, 0.5% pH 3–10 ampholyte). Protein concentration was determined using Bradford reagent (Bio-Rad, Hercules, CA, USA). The protein samples were pooled into TB group and control group, and adjusted to a final concentration of 1 mg/mL by mixing equal amounts of protein from individual TB samples and controls, respectively. The resulting two samples were further analysed.

Gel electrophoresis and image analysis were performed as described elsewhere [8] with some modifications. Each sample, containing 150 μg protein with the final volume of 410 μL solubilization buffer, was loaded onto 24-cm immobiline DryStrips (pH 3–10, non-linear) on an IEF system (Bio-Rad). For isoelectric focusing (IEF), an active rehydration at 50 V for 12 h at 20°C was performed before the focusing programme of a total of about 70 kVh as follows: 100 V for 0.5 h, 300 V for 0.5 h, 600 V for 0.5 h, 1000 V for 0.5 h, increased gradient voltage to 3000 V for 2 h, and 9000 V for 3 h. Once IEF was completed, the strips were equilibrated in buffer (6 M urea, 30% glycerol, 2% SDS and 0.01% Bromphenol blue) with addition of 1% (w/v) DTT for 15 min and 4% (w/v) iodoacetamide for 15 min. The IEF strips were then run on 12% SDS-PAGE using the Protean II xi system (Bio-Rad). The gels were stained with a Bio-Rad Silver Stain and scanned with a Bio-Rad GS-800 scanner. Gel images were analysed using PDQuest software 7.1 (Bio-Rad). To provide consistency, as proposed by Plymoth et al. [9], each pooled sample was analysed in triplicate.

Mass spectrometry analyses were performed as described elsewhere [10] with some modifications. Briefly, protein spots of interest were destained, dehydrated, and then incubated with modified trypsin (10 μg/mL in 25 mM ammonium bicarbonate). Followed by double extraction, the extracts were dried and then solubilized with matrix solution (5 μg/μL α-cyano-4-hydroxycinnamic acid) directly onto the matrix-assisted laser desorption/ionization (MALDI) target. The mass spectrometry (MS) scans were acquired in positive reflector ion mode and the spectra were accumulated until sufficient intensity was achieved (on average 4800 laser shots per spot). The first ten precursor ions with the highest intensity were selected for further MS/MS analysis.

The resulting peptide mass lists were used to search the NCBInr database. Database searches were performed using the mascot search engine (Matrix science, Boston, MA, USA). The taxonomy used was human or bacteria. The following parameters were set up for the database search: peptide mass tolerance, 0.2 Da; MS/MS mass tolerance, 0.3 Da; enzyme, trypsin; maximum number of missed cleavages, 1; fixed modification, carbamidomethyl; variable modifications, methionine oxidation. Protein scores >65, confidence score(s) of MS and MS/MS spectra >95% were regarded as positive identifications.

Enzyme-linked immunosorbent assay (ELISA) was used to confirm the proteomics results. Interleukin-25 (IL-25) level was measured with an IL-25 ELISA kit according to the instructions (BioSource, Nivelles, Belgium). Optical density at 450 nm was measured using a spectrophotometer. All samples were assayed in duplicates.

Related results were presented as the mean ± standard deviation (SD) from three independent experiments. Mean values were statistically compared using ANOVA tests or Student’s t-test. Statistical significance was defined as p < 0.05.


Image analysis of two-dimensional gels

Studying sputum fluid is clinically important, and greater knowledge concerning its components holds potential value for measuring health and disease. To get an insight into the protein profiles induced by TB infection, a comparative proteomic analysis of cell-free sputum from TB patients and healthy controls was performed.

Pooling of the samples facilitated the initial identification of differentially regulated proteins in the patients with active pulmonary tuberculosis and healthy controls. Pooling also reduced the experimental variations in the data and minimized the data files subjected to computer-intensive comparative analysis. Many published research findings were from pooled samples [8,11]. Our pretest data indicated that the profiles for the pooled samples could generally mirror the protein profiles of individuals. So, in our study, pooled samples were used for proteomics analysis as described in other studies [12,13].

Representative 2D gel images are shown in Fig. 1(a,b). A mean of 900 protein spots per gel were recognized and the volume of all spots was evaluated. Of 185 differentially expressed protein spots, 150 were decreased and 35 were increased in the TB sample compared with the controls. Compared with the controls, levels of proteins in the TB sample were increased on average by 2.2- to 6-fold, but decreased on average by 2.1- to 100-fold.

Protein identification and differentially expressed proteins

All differentially expressed proteins are summarized in Table 2. The result of matrix-assisted laser desorption/ionization-time of flight/time of flight-mass spectrometry MALDI-TOF/TOF-MS analysis of IL-25 is shown in Fig. 2 as an example. Out of the 62 proteins, 15 were up-regulated, whereas 47 were down-regulated in the TB samples compared with the controls. Among the 62 identified proteins, 55 are from humans and seven are from bacteria, respectively.

Table 2.   Functional classification of proteins isolated from two-dimensional electrophoresis gels of sputum and identification of 62 significantly changed proteins
CategoryProtein Mra pIaAccession no.Expression
  1. The pI and mass values were obtained from the MASCOT database (gi, Gene bank ID).

  2. aCalculated value.

Acute phase proteinSerum albumin69348.66.13 gi|23307793 Down-regulated
Acute phase proteinAlbumin-like52047.85.69 gi|763431 Down-regulated
Acute phase proteinAlloalbumin venezia69181.45.99 gi|178345 Up-regulated
Acute phase proteinChain A, crystal structure of65178.25.57 gi|55669910 Down-regulated
the Ga module complexed
with human serum albumin
Acute phase proteinChain A, crystal structure66428.95.67 gi|3212456 Down-regulated
of human serum albumin
Acute phase proteinChain A, human serum66409.95.66 gi|31615330 Down-regulated
albumin mutant R218h
complexed with thyroxine
Acute phase proteinChain A, human serum66369.95.62 gi|31615331 Down-regulated
albumin mutant R218p
complexed with thyroxine
Acute phase proteinPRO267532553.46.14 gi|7770217 Down-regulated
Acute phase proteinPRO204429228.86.97 gi|6650826 Down-regulated
Acute phase proteinPRO170832150.26.6 gi|7959791 Down-regulated
Acute phase proteinPRO261956745.25.96 gi|11493459 Down-regulated
Acute phase proteinAlbumin preproprotein69321.55.92 gi|4502027 Down-regulated
Acute phase proteinTransferrin76999.66.81 gi|4557871 Down-regulated
Acute phase proteinLactoferrin783468.56 gi|186833 Up-regulated
Acute phase proteinAmylase alpha57758.96.47 gi|224980 Down-regulated
Acute phase proteinChain B, crystal structure37624.75.84 gi|2781208 Down-regulated
of fibrinogen fragment D
Acute phase proteinHaptoglobin Hp241716.96.23 gi|223976 Down-regulated
Acute phase proteinAlpha-2-macroglobulin70750.85.47 gi|177872 Down-regulated
Signal proteinLIM18295.99.54 gi|45504687 Down-regulated
Signal proteinSH362220.27.67 gi|1843392 Down-regulated
Signal proteinNuclear receptor subfamily 246125.88.62 gi|5032173 Down-regulated
Signal proteinLymphoid-restricted membrane protein56178.25.44 gi|42789729 Down-regulated
Signal proteinGTP-binding RAS-like 122314.48.94 gi|21553323 Down-regulated
Signal proteinEssential meiotic endonuclease 1 homolog 248320.98.78 gi|58197552 Down-regulated
Signal proteinKIAA1893 protein81050.78.74 gi|15620845 Up-regulated
Signal proteinKIAA1949 protein73019.25.46 gi|18916755 Down-regulated
Signal proteinST622438.710.06 gi|55958259 Down-regulated
Signal proteinRAD50 protein64848.98.7 gi|38511824 Down-regulated
Signal proteinEnolase 147139.37.01 gi|4503571 Down-regulated
Cytoskeleton structureACTG1 protein29392.75.5 gi|40226101 Down-regulated
Cytoskeleton structureFELIC67597.95.99 gi|15011298 Down-regulated
Cytoskeleton structureARP2 actin-related protein 2 homolog44732.26.3 gi|15778930 Up-regulated
Cytoskeleton structureTropomyosin isoform28402.64.89 gi|854189 Up-regulated
Cytoskeleton structureBeta actin41709.75.29 gi|4501885 Down-regulated
Cytoskeleton structureActin-related protein 2 isoform b44732.26.3 gi|5031571 Down-regulated
Immune responseIL-25185258.73 gi|18034676 Up-regulated
Immune responsePoly-Ig receptor75473.75.38 gi|514366 Down-regulated
Immune responseAnti-HBs antibody heavy chain51243.78.56 gi|41059927 Down-regulated
Immune responseImmunoglobulin heavy chain variable region9572.56.4 gi|27650530 Up-regulated
Immune responseIg A1 Bur73330.99.24 gi|229585 Up-regulated
Bacteriostatic proteinCysteine-rich protein 322455.89.2 gi|116517305 Down-regulated
Bacteriostatic proteinS100A810827.76.51 gi|21614544 Down-regulated
Bacteriostatic proteinChain A, Neutrophil Gelatinase-Associated Lipocalin20662.69.1 gi|29726386 Down-regulated
Bacteriostatic proteinLysozyme precursor16544.29.38 gi|307141 Down-regulated
Bacterial proteinMT387618414.611.35 gi|15843389 Up-regulated
Bacterial protein50S ribosomal protein L323372.310.27 gi|50955566 Up-regulated
Bacterial proteinUqhC34551.98.26 gi|28875479 Up-regulated
Bacterial proteinAspartyl/glutamyl-trna amidotransferase subunit b54580.84.84 gi|38233683 Up-regulated
Unnamed bacterial proteinHypothetical protein43842.28.5 gi|10764643 Up-regulated
Unnamed bacterial proteinHypothetical protein70651.98.85 gi|21739648 Up-regulated
Unnamed bacterial proteinHypothetical protein86940.15.29 gi|10438291 Up-regulated
Unnamed proteinHypothetical protein52689.88.83 gi|34526069 Down-regulated
Unnamed proteinHypothetical protein17712.75.2 gi|16306948 Down-regulated
Unnamed proteinHypothetical protein70651.98.85 gi|21739648 Down-regulated
Unnamed proteinHypothetical protein782888.56 gi|467237 Down-regulated
Unnamed proteinHypothetical protein48610.59.02 gi|34527430 Down-regulated
Unnamed proteinHypothetical protein2212.29.79 gi|553734 Down-regulated
Unnamed proteinHypothetical protein2212.29.79 gi|553734 Down-regulated
Unnamed proteinHypothetical protein23693.65.17 gi|106482 Down-regulated
Unnamed proteinHypothetical protein52536.87.29 gi|190570174 Down-regulated
Unnamed proteinHypothetical protein48610.59.02 gi|34527430 Down-regulated
Unnamed proteinHypothetical protein2212.29.79 gi|553734 Down-regulated
Figure 2.

 The result of the matrix-assisted laser desorption/ionization-time of flight/time of flight-mass spectrometry MALDI-TOF/TOF-MS analysis of the protein IL-25. It was identified to be human IL-25 (NCBI accession number: 18034676) by protein database search. (a) Peptide mass fingerprint. (b) MS/MS profile of the peptide with a mass of 1045.56 Da; amino acids indicated as ASEDGPLNSR; confidence score 99.783%.

Confirmation of IL-25 expression by ELISA

Because proteomics results were derived from the pooled sputum samples, to validate our proteomics results, ELISA analysis was further performed for the detection of individual samples from all the cases. To confirm the proteomic results, IL-25 was selected and subjected to analysis. The data showed that IL-25 had an increased abundance in the TB group compared with the controls (Fig. 3). The result was consistent with the proteomic data.

Figure 3.

 Confirmation of IL-25 expression by ELISA. The data showed that IL-25 had an increased abundance in the TB group compared with the control group. #The p value for IL-25 was <0.05 in the active TB group compared with the controls.

Bioinformatics analysis of identified proteins from 2DE

To further understand cellular pathways involved in response to TB infection, the identified proteins could be clustered into several distinct groups involved in acute phase response, signal transduction, cytoskeleton structure, immune response and other functions (Table 2).


Sputum fluid contains a wide variety of proteins that are released either locally by epithelial and inflammatory cells or through plasma exudation. Because of the diverse origin of sputum proteins, analysis of sputum fluid may reveal important pathological mediators and may enable more accurate characterization of many lung diseases at the molecular level.

Differential-display proteomics offers the opportunity to understand disease mechanisms and to develop new biomarkers for early diagnosis. In this report, we present the first differential proteomic analysis of sputum from patients with active pulmonary tuberculosis and healthy controls.

In the present study, bioinformatics analysis suggested the 62 identified proteins are involved in acute phase response, signal transduction, cytoskeleton structure, immune response and so on. Major biological functions of the acute phase proteins are to restore homeostasis and to improve survival. Albumin, as a nutritional factor and an acute phase protein, was significantly reduced in the serum of pulmonary TB patients. In the present study, levels of albumin and its derivatives were obviously decreased in the sputum of patients with pulmonary TB. One of the possible causes of the low sputum albumin was considered to be low nutritional status of patients with active pulmonary TB. Iron (Fe) acquisition is critical to the metabolism and growth of mycobacterium tuberculosis (MTB), which can acquire Fe from extracellular transferrin and lactoferrin. Transferrin was a negative acute phase reactant and this may explain a decrease in sputum transferrin, which was in agreement with a reduction in serum transferrin [14]. Lactoferrin had antimicrobial activity and was part of the innate defence, mainly at mucoses [15]. Limited data are available on the relationship between lactoferrin and pulmonary TB. Our results showed that lactoferrin was increased in the TB sputum, which was identical to the increase in plasma lactoferrin [16]. In the present study, alpha-2-macroglobulin and haptoglobin Hp2 had obviously declined in the TB sputum, which was in contrast to the findings of Adedapo et al. [17]. The detailed mechanism is unknown.

Eleven signal-related proteins were identified in the present study, among which, only KIAA1893 was up-regulated. LIM and SH3 decreased in the TB sputum, and were involved in signal transduction and organization of the actin filaments during various cellular processes [18]. The exact cellular functions of LIM and SH3 are as yet unknown. Nuclear receptor subfamily 2 was involved in signalling pathways and had been shown to regulate pathways in embryonic development, as well as in maintenance of proper cell function in adults [19]. Lymphoid-restricted membrane protein was expressed in a developmentally regulated fashion in both B- and T cell-lineages, which suggested an important role for Lrmp in the IP3-Ca2+ signal cascade [20]. GTP-binding RAS-like 1 belonged to a distinct branch of the functionally diverse Ras superfamily of monomeric GTPases [21]. KIAA1893 (G protein-regulated inducer of neurite outgrowth 1) was a critical subunit of N-methyl-d-aspartate receptors, members of the glutamate receptor channel superfamily, which were heteromeric protein complexes with multiple subunits arranged to form a ligand-gated ion channel [22]. Little is known about the function of KIAA1949. ST6 could catalyze the transfer of sialic acid to GalNAc, and showed a highly restricted pattern of expression in normal adult tissues, being largely limited to the gastrointestinal tract and absent in mammary glands [23]. RAD50 was involved in several critical cellular functions, including the repair of damaged DNA [24]. Enolase-1 may act as a cell surface plasminogen receptor during inflammatory cell invasion [25]. Up to now, the relationships between the above-mentioned 11 signal-related proteins and TB are little known.

Six cytoskeleton-related proteins were identified in our study, among which, only ARP2 actin-related protein 2 homolog and tropomyosin isoform were up-regulated, and the others were down-regulated. The cytoskeleton consists of dynamic filamentous structures. Actin proteins are important for cell movement and maintaining the cytoskeleton, which is the structural framework that determines cell shape and organizes cell contents. ACTG1 protein, ARP2 actin-related protein 2 homolog, beta actin and actin-related protein 2 isoform b are parts of the actin protein family. Tropomyosin can regulate actin-myosin interaction and actin filament function in the cytoskeleton [26]. FELIC, a novel protein, contains a domain homologous to the cytoskeletal protein ezrin, and forms a Lyn interaction with the GTPase Cdc42. The detailed function of FELIC is unknown. The above-mentioned findings suggested that after TB infection, levels of some cytoskeleton-related proteins were changed.

Five immune response-related proteins were identified in the study, of which, three proteins were increased and the other two were decreased. Antibody-related proteins included anti-HBs antibody heavy chain, immunoglobulin heavy chain variable region and Ig A1 Bur. We were more interested in both immunoglobulin heavy chain variable region and Ig A1 Bur, because the two antibodies might be specific ones against antigens of Mycobacterium tuberculosis. If so, the two antibodies might be used as biomarkers for diagnosis of active pulmonary TB. IL-25 is a recently identified member of the IL-17 cytokine family. Unlike the other members of this family, IL-25 promotes T helper (Th) 2 responses by driving the expression of IL-4, IL-5 and IL-13 [27]. TB infection itself is characterized by relatively high levels of Th2 cytokines such as IL-4, IL-5 and IL-13, which down-regulate Th1 responses, subsequently subvert adequate protective immunity, and ultimately impair bactericidal function. We found that IL-25 level was increased in the TB sputum compared with the controls and this may partly explain why tuberculosis infection is associated with Th2 response and impaired bactericidal function.

Four bacteriostatic proteins were identified. Human cysteine-rich protein, found in exocrine secretions and secretory granules of neutrophilic granulocytes, is believed to play a role in innate immunity. S100A8 is crucial for resistance against invasion by pathogenic bacteria [28]. Chain A, Neutrophil Gelatinase-Associated Lipocalin acted as a potent bacteriostatic agent because of its ability to capture and deplete siderophores and constituted a critical component of the innate immunity against bacterial infection [29]. Lysozyme is responsible for breaking down the polysaccharide walls of many kinds of bacteria and thus it provides some protection against infection. It was expected that these bacteriostatic proteins would display increased expression in the TB sputum; however, it was a big surprise to note that the four bacteriostatic proteins were significantly reduced in the TB sputum. The detailed mechanism is unknown; however, at some point, it may be confirmed that patients with TB infection are easily infected by other pathogens because of these low-level bacteriostatic proteins.

In addition to the identified 11 hypothetical human proteins, it is noteworthy that seven bacterial proteins were identified, including three unnamed bacterial proteins. Among the bacterial proteins, we were especially interested in MT3876. MT3876 is a hypothetical protein from MTB and function of the protein remains unknown. MT3876 may be a biomarker of active pulmonary TB infection.

In all, for the first time, our results revealed that a number of proteins were differentially expressed during active pulmonary tuberculosis infection. However, the relationship between TB infection and protein expression is just beginning to be explored and further investigation is needed to provide researchers and clinicians with a better understanding of tuberculosis pathogenesis and biomarkers for detection of active pulmonary tuberculosis infection.


We thank the researchers in the Beijing Proteome Research Center for their technical support.


This study was supported by grants from the National Science Foundation of China (No. 30972639) and the Project of Shandong Province Higher Educational Science and Technology Program of China (No. J09LF20).

Transparency Declarations

The authors declare no conflict of interests.