- Split View
-
Views
-
CiteCitation
H. Mordarska, J. Zakrzewska-Czerwiñska, M. Paściak, B. Szponar, S. Rowiñski; Rare, suppurative pulmonary infection caused by Nocardiopsis dassonvillei recognized by glycolipid markers, FEMS Immunology & Medical Microbiology, Volume 21, Issue 1, 1 May 1998, Pages 47–55, https://doi.org/10.1111/j.1574-695X.1998.tb01148.x
Download citation file:
© 2018 Oxford University Press
Close -
Share
Abstract
An opportunistic actinomycete was isolated as the only etiological agent of a severe, suppurative pulmonary infection. The strain was rapidly recognised as Nocardiopsis by the taxonomically important and immunologically active glycolipid markers (G1 and G2). Identification of the clinical isolate, from a group of actinomycetes mainly known as soil habitants, was definitely proved by chemotaxonomic studies (cell wall/sugar, phospholipid and fatty acid types) as well as by genomic data (GC content, DNA-DNA reassociation). The level of DNA-DNA homology of the clinical actinomycete, in comparison with other reference members of this genus, revealed the highest (88%) relatedness to Nocardiopsis dassonvillei. The results confirmed the value and generic specificity of glycolipid markers from Nocardiopsis, the first time used for rapid recognition of a clinical strain causing a nocardiosis-like disease.
1 Introduction
The natural habitat of Nocardiopsis is the soil and isolation of the microorganisms from other sources is not frequent [1]. Invasive disease in humans due to this actinomycetes is uncommon and controversial. They, like most other clinically important actinomycetes, are opportunistic rather than invasive pathogens. They have been especially implicated in cutaneous and subcutaneous and also in respiratory diseases [2, 3].
The differentiation of Nocardiopsis from other spore forming actinomycetes was difficult for a long time, because of the high morphological similarity to phylogenetically unrelated taxa e.g. Streptomyces. This could cause they were discarded in medical laboratories as contaminants [1].
A big progress in the classification of the microorganisms was achieved by development of chemotaxonomic studies. Basing on them most former species of Nocardiopsis appeared to be misclassified and have been transferred to other genera [4, 5]. Until now only seven species and one subspecies are known to have the features which justify their placement in this genus [1, 6, 7]. Two of them, Nocardiopsis dassonvillei and Nocardiopsis listeri, were isolated from human and animal material being implicated in ocular and pulmonary infections [7].
Among essential chemotaxonomic characters of Nocardiopsis the most specific seemed to be polar lipids especially phosphatidylcholine (PC) and two major compounds of glycolipid pattern, G1 and G2[8], known at present as dimonomannosyl digliceryde (DMDG) and diacylated trehalose (DAT), respectively [8, 9]. For the first time they were used in a rapid recognition of a clinical strain causing a serious pulmonary infection and identified as Nocardiopsis sp.
The results of this investigation are presented below.
2 Materials and methods
2.1 Bacteria and culture conditions
The object of the investigation was a clinical strain (isolated from a pus sample) temporarily labelled as CI-1. For comparison the following representative strains of most Nocardiopsis species and subspecies were used: N. alba subsp. alba PCM 2496 (DSM 43377™), N. alba subsp. prasina PCM 2493 (JCM 3336™), N. alborubida PCM 2490 (DSM 40465™), N. antarcticus PCM 2489 (JCM 6843™), N. dassonvillei PCM 2492 (JCM 7437™), N. listeri, PCM 2491 (DSM 40297™), N. lucentensis PCM 2497 (IFO 1584™) (PCM, Polish Collection of Microorganisms; DSM, Deutsche Sammlung von Mikroorganismen; JCM, Japan Collection of Microorganisms; IFO, Culture Collection of Institute for Fermentation Osaka).
Bacteria were grown on yeast-extract dextrose (YED) medium (pH 7.5–8) in submerged, shaken cultures at 35°C for 48 h. After checking for purity, they were killed with 1% (v/v) formalin harvested by centrifugation, washed, and a part dried, a part frozen at −20°C. The obtained cell masses were used for further studies.
2.2 Preliminary morphological observation
The preliminary macroscopic and microscopic observations of CI-1 strain were performed on mature (5–7 days) colonies grown on YED agar plate and on 48 h submerged culture incubated at 35°C. For staining of smears Gram and Kinyoun's (acid-fast) methods were used.
2.3 Glycolipid analysis
Bacterial biomass (200–300 mg) was extracted twice with 10 ml chloroform/methanol mixture (2:1, v/v) for 12 h at 37°C using a Teflon screw cap and gentle shaking.
The pooled extracts were evaporated under reduced pressure, redissolved in CHCl3/MeOH (1:1, v/v) to a final concentration of 50 mg/ml. The samples were analyzed by thin layer chromatography (TLC) according to Mordarska and Paściak [10].
2.4 Immunoserologic studies
2.4.1 Antigens and sera
G1 and G2, used as antigens for some immunoserologic tests, were obtained as described before [8]. Briefly, the crude lipid (~50 mg) extracted (as above) from wet cell mass of the typical strain of N. dassonvillei PCM 2492 (JCM 7437™) was separated on a silica gel 60, F-254 plate (layer thickness 2 mm; Merck) using the solvent system chloroform/methanol/water (65:25:4, v/v/v). The lipid bands were visualized first with UV light. The bands corresponding to G1 and G2 were scraped separately except the two narrow margins that were additionally checked for the localization of both glycolipids by spraying with orcinol or vanillin reagents and heating at 120°C for 10 min [10]. The G1 and G2 were eluted from the scraped bands with chloroform/methanol/water (1:1:0.2, v/v/v) and evaporated under stream of N2. The purification was done repeatedly until a single spot of each of the glycolipids was obtained on two-dimensional TLC (for the solvent systems see Fig. 3). The purified glycolipids were kept in CHCl3 under N2 at −20°C (they can be safely left in these conditions for a long period).
Two-dimensional thin layer chromatogram of polar lipids from CI-1 strain (TLC system, first direction as in Fig. 2; second direction, chloroform/acetic acid/methanol/water, 80:15:12:4, v/v/v/v; detection, spraying with 0.5% vanillin in ethanol with 3% H2SO4. Abbreviations: G1, G2, major glycolipids; DPG, diphosphatidylglycerol; PG, phosphatidylglycerol; PC, phosphatidylcholine.
Two-dimensional thin layer chromatogram of polar lipids from CI-1 strain (TLC system, first direction as in Fig. 2; second direction, chloroform/acetic acid/methanol/water, 80:15:12:4, v/v/v/v; detection, spraying with 0.5% vanillin in ethanol with 3% H2SO4. Abbreviations: G1, G2, major glycolipids; DPG, diphosphatidylglycerol; PG, phosphatidylglycerol; PC, phosphatidylcholine.
The sera tested in the study were obtained from the patient suffering from a pulmonary infection caused by the clinical isolate (CI-1) and from rabbits immunized with CI-1 or with N. dassonvillei (JCM 7437™) strains. Additionally some control sera were used from physically healthy human or rabbit individuals as well as from patients with tuberculosis or lung carcinoma, before treatment.
2.4.2 Immunization
Rabbits (5–6 months old) were immunized twice a week with a cell mass of actinomycetes according to the schedule of Gamian et al. [11] with a small modification. The cells were previously disintegrated in x Press (AB BIOX), lyophilised and suspended in phosphate-buffered saline (PBS). The following antigenic doses were used: 100 µg of dry cell mass/ml for the first subcutaneous injection and doubly increasing density (200, 400 to 6400 µg/ml) for the succeeding intravenous injections. The antisera were checked by conventional passive hemaglutination test (PHT) using sheep erythrocytes coated with Nocardiopsis cellular antigens prepared earlier for immunization. Because of pour antigenicity of actinomycetes the immunization was usually prolonged by two additional injections with the highest dose of cells (6400 µg/ml). After the last injection (7–10 days) the titres of antisera were verified by PHT, then rabbits were bled. The separated antisera were decomplemented (56°C, 30 min) and stored at −20°C.
2.4.3 Delayed type hypersensitivity (DTH)
DTH reaction was induced after the immunization of the rabbits. Glycolipid suspensions in PBS (1000, 100, 50 µg/ml), obtained by sonification in ultrasonic water bath (3 min, 20°C), were injected (0.1 ml) intracutaneously in the clipped paravertebral region of immunized animals. For control purpose 0.1 ml of bacterial cell suspension (1 mg/ml) and PBS were injected.
2.4.4 Dot-EIA
A cellulose nitrate membrane (MFS, USA, pore size 0.45 µm), and the method employed by Papa (called here dot-EIA) [12] were used with a small modification. Briefly, glycolipid antigens (dissolved in chloroform) were placed on nitrocellulose strips in an equal volume (1 µl) with decreasing concentration (100, 50, 10, 1 ng). They were dipped in 0.2% casein in Tris-buffered saline, pH 7.5 (TBS) for 1 h (with gentle shaking), rinsed twice with TBS followed by incubation with specific antiserum diluted in TBS for 2 h at 37°C. After the strips were washed in TBS (3×), the anti-rabbit or anti-human (respectively) IgG/HRP (horse-radish peroxidase) conjugate (DAKO) was used. The substratum was 0.05% 4-chloro-1-naphthol (Sigma). The strips were incubated in the reagent at room temperature and the reaction was stopped by washing with water.
2.5 Cell chemistry
Cell wall, whole cell sugars and polar lipids of CI-1 strain were determined by TLC chromatography [13, 14].
The fatty acid methyl ester profile was determined by gas liquid chromatography-mass spectrometry (GC-MS) [15] using Hewlett-Packard (model 5971A) apparatus, a type HP1 glass capillary column (0.2 mm by 12 m) and a temperature increasing from 50 to 270°C at a rate of 8°C/min in electron ionization (70 eV) mode.
The mycolic acid methyl ester analysis was employed by method of TLC [16].
2.6 Genotaxonomic studies
For the isolation and purification of DNA there was used the procedure described before [17]. Briefly, the organisms were grown as submerged cultures in modified Souton's broth ad 37°C for 24 h. They were harvested (1 g) suspended in 1 ml of sucrose buffer (0.3 M sucrose, 25 mM EDTA, 25 mM Tris-HCl, pH 8.0) and incubated with lysozyme (1 mg) at 37°C for 1–3 h. When the lysis occurred, SDS was added to the final concentration of 2% and the samples were then incubated at 4°C for 0.5 h. DNA was extracted using phenol saturated with 0.1 M Tris-HCl (pH 9.0). After ethanol precipitation, DNA samples were treated with RNase (100 µg/ml) at 37°C for 0.5 h. The extraction with phenol followed by ethanol precipitation was repeated. Purified DNA was dissolved in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0).
2.6.1 DNA base composition
The average guanine plus cytosine (G+C) content was determined by thermal denaturation [18].
Reference DNA was obtained from Micrococcus luteus (M. lysodeikticus) ATCC 4698.
2.6.2 DNA reassociation studies
The levels of DNA relatedness between the strain CI-1 and all typical strains of Nocardiopsis species were examined using a nitrocellulose membrane filter technique [17, 19].
3 Results
An opportunistic strain of actinomycete was isolated from a pus sample of a large pulmonary abscess of a 46 year old man suffering from chronic pneumonia. The patient was hospitalized at the Department of Tuberculosis and Lung Diseases, Military Clinical Hospital in Wrocław and was treated first with antibiotics (mainly Amikacin) whereafter with high doses of sulfonamides.
3.1 Preliminary morphological data
The basic microscopic observations of the clinical isolate showed long, straight or flexuous, moderately branched rods with a burbled wire-like appearance (Fig. 1A). They were Gram-positive and non-acid-fast. The strain produced also white, abundant, Streptomyces-like aerial mycelium on blood as well as yeast-extract dextrose agar (Fig. 1B).
Preliminary morphological observations of the clinical isolate (CI-1). A: Light micrograph of Gram-positive mycelium from 48 h submerged culture (YED medium, growth temperature 37°C; magnification, ×1000). B: Colonies with white aerial mycelium on YED agar (incubation 10 days at 37°C; magnification, ×5).
Preliminary morphological observations of the clinical isolate (CI-1). A: Light micrograph of Gram-positive mycelium from 48 h submerged culture (YED medium, growth temperature 37°C; magnification, ×1000). B: Colonies with white aerial mycelium on YED agar (incubation 10 days at 37°C; magnification, ×5).
3.2 Polar glycolipids profile
Rapid TLC analysis of polar lipids from biomass of CI-1 strain and of other Nocardiopsis spp. used for comparison revealed two major glycolipids, G1 and G2, in the crude extracts (Fig. 2Fig. 3). The results showed the generic specificity of both glycolipid markers as well as the taxonomic position of the clinical isolate within Nocardiopsis genus.
Chromatogram of glycolipids from clinical isolate (CI-1) and from typical representatives of Nocardiopsis species. Abbreviations: 1, N. alborubida; 2, N. listeri; 3, N. antarcticus; 4, CI-1; 5, N. dassonvillei; 6, N. alba subsp. alba; 7, N. alba subsp. prasina; 8, N. lucentensis; solvent system, chloroform/methanol/water 65:25:4, v/v/v; detection, spraying with 0.5% orcinol in ethanol with 3% H2SO4 followed by heating at 120°C for 10–15 min, to locate sugar containing lipids; G1, G2, major glycolipids.
Chromatogram of glycolipids from clinical isolate (CI-1) and from typical representatives of Nocardiopsis species. Abbreviations: 1, N. alborubida; 2, N. listeri; 3, N. antarcticus; 4, CI-1; 5, N. dassonvillei; 6, N. alba subsp. alba; 7, N. alba subsp. prasina; 8, N. lucentensis; solvent system, chloroform/methanol/water 65:25:4, v/v/v; detection, spraying with 0.5% orcinol in ethanol with 3% H2SO4 followed by heating at 120°C for 10–15 min, to locate sugar containing lipids; G1, G2, major glycolipids.
3.3 Immunoserological studies
The results presented above were also proved by preliminary immunoserological studies. The patient's serum as well as the rabbit antisera against CI-1 and N. dassonvillei™ strain, used as a positive control, reacted in the simple dot-EIA test with G1 and G2 antigens (Fig. 4). A strong positive signal was observed with the sera diluted 1/100 and decreasing concentration of the glycolipid markers: G1, to 100 ng for patient (Fig. 4A) or to 50 for rabbit (Fig. 4 B, C) sera and G2, to 10 ng, irrespective of the sera. Control sera (from healthy individuals and from those with tuberculosis or lung carcinoma), taken for cross reactivity, were negative.
Reactivity of glycolipid markers, G1 (lower rows) and G2 (upper rows), from Nocardiopsis dassonvillei type strain (PCM 2492™) in simple dot-EIA test. Concentration of the antigens: 100, 50, 10, 1 ng. A: Patient's serum. B: Rabbit polyclonal antisera against the clinical isolate (CI-1). C: Rabbit polyclonal antisera against N. dassonvillei (type strain) as positive control.
Reactivity of glycolipid markers, G1 (lower rows) and G2 (upper rows), from Nocardiopsis dassonvillei type strain (PCM 2492™) in simple dot-EIA test. Concentration of the antigens: 100, 50, 10, 1 ng. A: Patient's serum. B: Rabbit polyclonal antisera against the clinical isolate (CI-1). C: Rabbit polyclonal antisera against N. dassonvillei (type strain) as positive control.
Rabbit SDH reactions after immunization with CI-1 cell mass suspension (Cm), developed with both G1 or G2 glycolipids as well as with Cm and PBS (positive and negative controls, respectively) were presented in Table 1. The positive reactions were obtained only with G2 and Cm but neither with G1 antigen nor PBS.
Ability of the major glycolipids (G1, G2) from Nocardiopsis dassonvillei to induce delayed hypersensitivity in rabbits injected with the clinical isolate (CI-1)
| Antigen | Intensity of SDH reaction (mm) | ||
| dose 5 µg | dose 10 µg | dose 100 µg | |
| G1 | 2 | 0 | 0 |
| G2 | 8 | 14 | 14 |
| Cm | 10 | n.d. | n.d. |
| PBS | 0 | ||
| Antigen | Intensity of SDH reaction (mm) | ||
| dose 5 µg | dose 10 µg | dose 100 µg | |
| G1 | 2 | 0 | 0 |
| G2 | 8 | 14 | 14 |
| Cm | 10 | n.d. | n.d. |
| PBS | 0 | ||
Diameter of erythema; G1, G2, major glycolipids from Nocardiopsis dassonvillei (PCM 2492) in PBS; Cm, dried cell mass of CI-1 in PBS (positive control); PBS, negative control; n.d., not determined.
Ability of the major glycolipids (G1, G2) from Nocardiopsis dassonvillei to induce delayed hypersensitivity in rabbits injected with the clinical isolate (CI-1)
| Antigen | Intensity of SDH reaction (mm) | ||
| dose 5 µg | dose 10 µg | dose 100 µg | |
| G1 | 2 | 0 | 0 |
| G2 | 8 | 14 | 14 |
| Cm | 10 | n.d. | n.d. |
| PBS | 0 | ||
| Antigen | Intensity of SDH reaction (mm) | ||
| dose 5 µg | dose 10 µg | dose 100 µg | |
| G1 | 2 | 0 | 0 |
| G2 | 8 | 14 | 14 |
| Cm | 10 | n.d. | n.d. |
| PBS | 0 | ||
Diameter of erythema; G1, G2, major glycolipids from Nocardiopsis dassonvillei (PCM 2492) in PBS; Cm, dried cell mass of CI-1 in PBS (positive control); PBS, negative control; n.d., not determined.
3.4 General chemotaxonomic profile
The generic position of the rare clinical isolate was also proved by most other chemotaxonomic characteristics known as essential for the final classification of Nocardiopsis[1]. They were established after pure cell wall or whole cell analysis.
The diagnostic amino acid of the cell wall was meso-DAP; additionally alanine and glutamic acid also occurred. No characteristic sugars were found, only glucose was detected in the whole cell hydrolysates; i.e. the organism had the cell wall of type III/C.
The fatty acid spectrum was designated as type 3d [20] consisting mainly, apart from 18:0 (17.2%), of acid of most branched fatty acids: iso-methyl 16:0 (29.6%); iso-methyl 18:0 (10.6%); anteiso-methyl 17:0 (14.7%); and 10-methyl 18:0 (15.3%). No mycolic acids were found.
The polar lipids (Fig. 3), besides two major glycolipids G1 and G2, were of phospholipid type P-III: phosphatidylglycerol (PG), diphosphatidylglycerol (DPG), phosphatidylmethylethanolamine (PME) and phosphatidylcholine (PC) [14].
3.5 Genotaxonomic studies
The G+C mol % of the DNA from CI-1, determined by the thermal denaturation method, was 60. This value appeared to be in the same range as those found for Nocardiopsis genus (64–69) [21]. These data as well as DNA-DNA hybridization between CI-1 and two opportunistic species of Nocardiopsis genus were presented in Table 2. CI-1 showed distinctly high (88%) homology to the type strain of N. dassonvillei and low (22%) homology to the other reference organism. On this basis CI-1 was intensified as N. dassonvillei. The strain was deposited in the Polish Collection of Microorganisms (N. dassonvillei PCM 2503; clinical isolate).
Genetic relatedness between the clinical isolate (CI-1) and two opportunistic representatives of Nocardiopsis genus
| Strain | G+C (mol %) | DNA homology % with CI-1 |
| CI-1 | 69 | 100 |
| N. dassonvillei (PCM 2492™) | 69 | 88 |
| N. dassonvillei (IMRU 509™) | 69.8 | n.d. |
| N. listeri (PCM 2491™) | n.d. | 22 |
| Strain | G+C (mol %) | DNA homology % with CI-1 |
| CI-1 | 69 | 100 |
| N. dassonvillei (PCM 2492™) | 69 | 88 |
| N. dassonvillei (IMRU 509™) | 69.8 | n.d. |
| N. listeri (PCM 2491™) | n.d. | 22 |
Genetic relatedness between the clinical isolate (CI-1) and two opportunistic representatives of Nocardiopsis genus
| Strain | G+C (mol %) | DNA homology % with CI-1 |
| CI-1 | 69 | 100 |
| N. dassonvillei (PCM 2492™) | 69 | 88 |
| N. dassonvillei (IMRU 509™) | 69.8 | n.d. |
| N. listeri (PCM 2491™) | n.d. | 22 |
| Strain | G+C (mol %) | DNA homology % with CI-1 |
| CI-1 | 69 | 100 |
| N. dassonvillei (PCM 2492™) | 69 | 88 |
| N. dassonvillei (IMRU 509™) | 69.8 | n.d. |
| N. listeri (PCM 2491™) | n.d. | 22 |
4 Discussion
Two glycolipids G1 and G2, known as chemotaxonomic markers of Nocardiopsis[8, 9], were successfully used for rapid recognition of a rare nocardiosis-like disease. They were found in the cells of a clinical isolate (the only etiological agent of the serious infection) by using a simple TLC analysis for polar glycolipids. Though, the first microscopic and macroscopic observations of the isolate, especially the abundant, aerial mycelium on blood or yeast-extract dextrose agar suggested rather a Streptomyces-like organism that could be a contamination.
The preliminary recognition of CI-1 strain as Nocardiopsis sp., based on the simple TLC pattern of its polar glycolipids, was also proved by immunoserologic studies. The components were specifically active in the humoral response and one of them (G2) was additionally characterized by a strong cellular response. The results were in agreement with our earlier observation [23] and helped us to support the diagnosis of the nocardiosis-like infection that could be successfully cured with combination of sulfonamides [24] instead of antibiotics.
It is worth to mention that G1 and G2 markers represent two predominating types of bacterial glycolipids known as glycosyl glycerides and acyl sugars [25]. In the previous study G1 was preliminary recognized as mannosyl diglyceride (MDG) and G2 as acylated glucose (AG) [8]. At present, the former one appeared to be a glycolipid containing trehalose (DAT) [9] and its detailed structural analysis will be reported in a separate paper.
Meanwhile, we also made an effort to check the taxonomic position of the isolate by the chemotaxonomic and genotaxonomic studies. The chemotaxonomic profile, known as essential for the classification of Nocardiopsis genus [1], was shown also to be typical for this strain. Even, the ability to synthesize tuberculostearic acid (over 15% of the total fatty acids of the isolate) is known as shared by Nocardiopsis with the IV type cell wall of actinomycetes [26].
The genotaxonomic studies, performed to identify CI-1 on a species level, showed that DNA from this strain was characterized by the highest (88%) homology according to the reference DNA from N. dassonvillei. The DNA homology of the isolate to DNAs from other reference opportunistic organisms (N. listeri) was much lower (22%). This index appeared to be on or behind the interspecies hybridization level which is known to fluctuate within 70–100%[27]. Thus, the clinical isolate was shown to belong to N. dassonvillei. That also meant that in Nocardiopsis genus only two species, N. dassonvillei and N. listeri, could still be taken under consideration as the clinically important taxa.
To sum up, the major glycolipids from Nocardiopsis appeared to be useful not only as chemotaxonomic but also as immunodiagnostic markers. The components were earlier found in N. dassonvillei[8], the only representative of the monotyping genus of Nocardiopsis for a long time. The markers allowed us to distinguish these microorganisms from Actinomadura, known as an etiological agent of madurosis. In this study, using the simple TLC method, we could also try to compare the CI-1 glycolipid pattern with those from nearly all known up to day Nocardiopsis spp. with exception of unobtainable N. halophila[7]. The results seemed to show its generic specificity.
It is worth to stress that the procedure for preliminary detection and for later preparation of the glycolipid markers does not require sophisticated techniques. Besides easily available bacterial crude lipids and simple TLC-PLC equipment as well as reagents, it is important to use: (1) the sugar specific spray reagent (orcinol) visualizing the bacterial TLC glycolipid profile more distinctly than whole lipid detectors; (2) the highly sensitive dot-EIA test easily applicable, especially for lipid antigens better soluble in chloroform than in other organic solvents.
The results achieved by each of the techniques or by the combination of both may be the most satisfactory, since they confirm etiology and minimize a possibility of misdiagnosis of the neglected infections.
Highly sensitive and specific glycolipid antigens were also successfully used for serodiagnosis of tuberculosis [28, 29], farmer's lung disease [30, 31] and melioidosis [32]. Moreover, the authors suggested the practical advantages of glycolipids over the protein antigens. The antigenicity and specificity of some bacterial lipids had also been proved by interesting experiments on the presentation of these agents to T cell clones [33].
Further studies are required in order to determine both taxonomic as well as immunodiagnostic values of glycolipid markers and their bioactive role in the aspect of structure and function.
Acknowledgements
This work was partially supported by Committee of Scientific Research (KBN), Grant No 4 PO5A 038 09.
The authors would like to thank Dr. A. Gamian for assistance with fatty acid analysis. Special thanks are also due to B. Lis for excellent technical assistance with the genotaxonomic studies.




