Ultrastructural analysis of the intracellular surfactant in lungs of healthy and ovalbumin sensitized and challenged Brown Norway rats

Abstract

Introduction: In human and experimentally induced asthma, a dysfunction of the intra-alveolar-surface active agent (surfactant) has been demonstrated. Type II alveolar epithelial cells (AEII) synthesize, secrete and recycle surfactant. Prior to secretion, intracellular surfactant is stored in specific secretory organelles of AEII. The lamellar bodies (Lb) represent its ultrastructural correlate. The aim of this study was to investigate whether disturbances of the intra-alveolar surfactant are accompanied by alterations in the intracellular surfactant.Material and Methods: Brown-Norway rats were sensitized twice with ovalbumin (OVA) and heat killed Bordetella pertussis bacilli. During airway challenge, an aerosol of 5% ovalbumin/saline solution (0.25 l/min) was nebulized. 24 h after airway challenge, lungs were fixed by vascular perfusion. AEII and their Lb were characterized stereologically by light and electron microscopy.Results: In both groups, AEII were structurally intact. The number of AEII per lung and their number-weighted mean volume did not differ (controls: 49 × 10⁶, 393 µm³; asthmatics: 44 × 10⁶, 390 µm³). A mean of 90 Lb in AEII of asthmatics and of 93 Lb in AEII of controls were evaluated. The Lb mean total volume was 59 µm in asthmatics and 68 µm in controls. Values of both parameters did not reach significance. Also, the size distribution and mean volume of Lb was not influenced by asthma induction, because the volume weighted mean volume of Lb (2.18 µm in asthmatics compared to 1.87 µm in controls) and the numerical weighted mean volume (0.96 µm in asthmatics and 0.75 µm in controls) were comparable in both groups.Conclusion: The obtained results suggest that asthma-induced surfactant dysfunction is not related to disturbances in the intracellular surfactant´s ultrastructural correlates.

Keywords:

• Allergy
• lamellar bodies
• stereology
• surfactant
• type II alveolar epithelial cell

Introduction

Pulmonary surfactant (surface active agent) stabilizes alveolar walls and prevents alveolar collapse during exhalation by a surface area depending reduction of alveolar surface tension.^1–3 Surfactant can be subdivided in an intra-alveolar pool and an intracellular compartment.^4,5 The intra-alveolar pool is represented by the active and inactive surfactant subtypes in the alveolar hypophase of the lining layer covering the alveolar epithelium.^6–10 The intracellular surfactant is found in type II alveolar epithelial cells (AEII), which synthesize, store, secrete and recycle surfactant.¹¹ The phospholipids are synthesized in the endoplasmic reticulum and stored in so-called immature lamellar bodies (Lb).¹² Mature Lb are formed by fuzing immature Lb with SP-B and -C containing multivesicular bodies (mvb).^13,14 SP-B and SP-C containing Lb are secreted by controlled exocytosis, SP-A and SP-D containing mvb by constitutive exocytosis.^15,16

Human bronchial asthma is one of the most common diseases.¹⁷ Animal models of asthma are widely used.^18–23 Drawing conclusions from animal models to human asthmatics depends highly on the animal species selected.^18,21 Asthma induction with ovalbumin in rats leads to cellular infiltration of the lung, antigen-specific IgE production, and a predominant Th2 response.^21,24 Furthermore, parenchymal infiltration was observed.^25–31 Different research groups could verify that surfactant dysfunction was associated with a reduction in minimum surface tension in asthma-induced animals^10,32,33 as well as in patients with asthma.^20,34–36 It could be shown that the different proteins passed or secreted into the alveolar hypophase,^33,37 particularly the eosinophilic cationic protein (ECP) secreted from eosinophils,^38–41 are jointly responsible for surfactant dysfunction.

However, there is no information whether allergic inflammation affects the structural correlate of the intracellular surfactant pool. Therefore, we tested the hypothesis that asthma induced alterations also occur in the AEII and their storage organelles in the asthma model of the Brown Norway (BN) rat.

Material and methods

Male Brown Norway (BN) rats were kept under specific pathogen-free conditions. They had access to food and water ad libitum. Experiments were approved by the local government authority and met the NIH Guidelines for the Care and Use of Laboratory Animals (NIH Publication No. 85, reprint 2002).

Asthma model

Experimental asthma was induced according to a standard protocol for sensitization and challenge.^10,42 Rats of the challenge group (n = 7) were sensitized twice both with 1 mg ovalbumin (OVA) (Sigma, Taufkirchen, Germany) and 200 mg Al(OH) 3 (Sigma, Taufkirchen, Germany) in 1 ml 0.9% NaCl solution by subcutaneous application and with 5 × 10⁹ heat-killed Bordetella pertussis bacilli in 0.4 ml 0.9% NaCl solution (kindly donated by Chiron Behring, Marburg, Germany) given intraperitoneally. The sensitization was retried 6-7 days later. Airway challenge (provocation) was carried out another 6-7 days later by nebulization with 5% OVA in 0.9% NaCl as described earlier.¹⁰ Rats without any treatment served as controls (home cage controls, n = 5).

Fixation, sampling and tissue processing

Animals were anesthetized and ventilated with a tidal volume of 5 ml and a respiration rate between 30 and 40 breaths per minute as described earlier.¹⁰ After thoracotomy and opening of the pericardial cavity a cannula was engaged in the pulmonary artery via the right ventricle. Preperfusion with Krebs-Henseleit solution containing 0.25 g/1900 ml procaine-HCl and in situ perfusion fixation with 1.5 glutaraldehyde and 1.5% paraformaldehyde in 0.15 M HEPES buffer followed immediately after stopping ventilation and exsanguination as described earlier.¹⁰ After perfusion fixation, the lungs were fixed by immersion for 24 h. After the determination of lung volume by the Archimedes principle,⁴³ lungs were embedded in 2% aqueous agar, and each organ was cut from apical to caudal into 3 mm slices using a tissue slicer as described in more detail earlier.¹⁰ All further tissue processing of the specimens followed a routine protocol.^10,44,45

Stereology

Stereological analyses were carried out according to the guidelines for quantitative assessment of lung structure.⁴⁶ Subcellular volume densities (Vv/AEII) of intracellular surfactant component profiles such as Lb, composite bodies (cb), and multivesicular bodies (mvb) as well as of additional subcellular organelle profiles like nucleus, mitochondria, and cytoplasm were evaluated according to the point counting method.⁴⁷

We characterized AEII by determining their number, their mean volume and their subcellular composition. Furthermore, the Lb as surfactant storage organelles were characterized by their number and their mean volume. For sampling test fields in light and electron microscopical sections we used the systematic uniform random sampling method that covers all regions with equal probability.⁴⁶ We used the following parameters to verify possible asthma induced alterations of AEII in asthmatics compared to controls, which are described in detail in reviews of Hsia⁴⁶ and others.^48,49

Using the disector principle, the number of AEII per lung was determined light microscopically and the number of Lb per AEII electron microscopically. Comparing paired sections from the same tissue block with known distance in z direction, AEII were counted if their nucleolus was visible in one section but not in the other section (Figure 1). Lb were counted if Lb appeared in one section but not in the other (Figure 2). The height (h) of the light microscopic disector was 2 µm, h of the electron microscopic disector 100 nm.

Figure 1. Disector method, counting of AEII. The number of AEII per lung was determined light microscopically using the disector principle. Comparing paired sections from the same tissue block with known distance in z direction (sampling section for counting, look-up section for comparison), AEII were counted if their nucleolus was visible in the sampling section but not in the look-up section. The section pairs were used for counting in both directions, i.e. using each section once for counting and once for comparison. The height (h) of the light microscopic disector was 2 µm. Scale bar = 25µm.

Figure 2. Disector method, counting of Lb. Cellular Lb were counted on paired sections if Lb appeared in one section but not in the other. As for counting of AEII, each section was used once for counting and once for comparison (i.e. counting in both directions). The height (h) of the electron microscopic disector was 100 nm. Scale bar = 2 µm.

At first, the numerical density of AEII (N_V (AEII/par)) and Lb (N_V(Lb/AEII)) were evaluated according to the following formula: $${\mathrm{N}}_{\mathrm{V}}\left(\text{AEII}/\text{par}\right)=\text{ 1}/2\times \text{ a}\left(\mathrm{f}\right)\times \text{ h }\times \Sigma {\text{ Q}}^{-}\text{AEII}/\Sigma \text{ frames}/\text{par}$$ $${\mathrm{N}}_{\mathrm{V}}\left(\text{Lb}/\text{AEII}\right)=\text{ 1}/2\times \text{ a}\left(\mathrm{p}\right)\times \text{ h }\times \Sigma {\text{ Q}}^{-}\text{Lb}/\Sigma \text{ AEII}$$ where: h = disector height, a(f) = area per test frame, a(p) = area per test point, Q^- = counting events

The following additional parameters were calculated:^48,50 $$\text{Number\hspace{0.17em}of\hspace{0.17em}AEII\hspace{0.17em}per\hspace{0.17em}lung\hspace{0.33em}}:\text{\hspace{0.17em}N}\left(\text{AEII},\text{lung}\right)={\overline{\nu }}_{\text{N}}\left(\text{AEII}\right)\times \text{\hspace{0.17em}V}\left(\text{lung}\right)$$ $$\text{Number of Lb per AEII}:\text{ N}\left(\text{Lb}/\text{AEII}\right)={\text{ N}}_{\mathrm{V}}\text{Lb }\times {\overline{\nu }}_{\mathrm{N}}\left(\text{AEII}\right)$$ $$\text{Number\hspace{0.17em}of\hspace{0.17em}Lb\hspace{0.17em}per\hspace{0.17em}lung\hspace{0.33em}}:\text{\hspace{0.17em}N}\left(\text{Lb},\text{lung}\right)={\mathrm{N}}_{\mathrm{V}}\text{Lb\hspace{0.17em}}\times {\overline{\nu }}_{\text{N}}\left(\text{AEII}\right)\times \text{\hspace{0.17em}V}\left(\text{lung}\right)$$ $$\text{Volume of Lb per AEII}:{\text{ V}}_{\mathrm{V}}\left(\text{Lb}/\text{AEII}\right)\text{ x V}\left(\text{AEII}\right)$$ $$\begin{array}{l}\text{Total\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}Lb\hspace{0.17em}per\hspace{0.17em}lung:}\\ \text{V}\left(\text{Lb},\text{lung}\right)={\mathrm{V}}_{\mathrm{V}}\left(\text{Lb}/\text{AEII}\right)\times \text{\hspace{0.17em}V}\left(\text{AEII},\text{lung}\right)\times \text{\hspace{0.17em}N}\left(\text{AEII},\text{lung}\right)\end{array}$$

The number and the number-weighted mean volume (${\overline{\nu }}_{\text{N}}$, AEII) of AEII were analyzed using an Axioscope light microscope (Zeiss, Oberkochen, Germany) equipped with a computer-assisted stereology tool box (Cast; Olympus, Ballerup, Denmark). The number-weighted mean volume was evaluated during counting of AEII using the rotator method. The subcellular volume densities of organelles as well as the surface and numerical densities in AEII were evaluated by means of an EM 10 (Zeiss, Oberkochen, Germany) or EM 900 (LEO, Oberkochen, Germany). The volume-weighted mean volume of Lb was evaluated using the point sampled intercepts and a ruler.⁵¹ The numberweighted mean volume of Lb was determined by dividing V_V(Lb/AEII) by N_VLb.⁵⁰

Statistics

The entire graphics were created with the computer program “GraphPad Prism 6”. In the figures and tables, all data were plotted as arithmetic mean ± standard deviation (SD) if not indicated otherwise. The calculations were done with the statistics program “GraphPad InStat 6”. Stereologic analyses were carried out as evidence of statistical differences, starting at a p < 0.05. We used the non–parametric Mann-Whitney test because the data were not Gaussian distributed.

Results

We fixed the lungs by perfusion fixation in order to examine the distribution of the intra-alveolar and intracellular surfactant without changing their micro-organization. Therefore, we timely performed additional experiments using also the established asthma model of the Brown Norway rat under the same conditions in order to collect functional data and to carry out BAL to determine surfactant activity and numbers of inflammatory cells. The data of these experiments are already published. We summarized these data in the following to verify the allergic inflammation.

Sensitization and challenge with OVA lead to a significant increase of cell numbers and eosinophils in the BALF compared to controls.^42,52 We found also a significant higher infiltration of eosinphils in lung parenchyma.^52,53 Measuring lung function in intubated and spontaneously breathing animals show a significant increase of lung resistance by 500% in OVA challenged rats compared to controls.¹⁰ The minimum surface tension increases significantly by 95% after OVA sensitization and challenge.¹⁰

Moreover, sensitization and challenge with OVA led, on the structural level, to a significantly higher alveolar epithelial fraction covered with alveolar fluid as well as to a significantly increased fraction of inactive surfactant compared to controls.¹⁰ These results were obtained by evaluating the same perfusion fixed lungs, which are also used for investigation of the intracellular surfactant in the present paper.

Animal weight and lung volume (Table 1)

There were no asthma dependent alterations in body weight and lung volume as shown in Table 1.

Table 1. Body weight and lung volume.

Parameter	Controls (n = 5)	OVA challenge (n = 7)	P value control vs OVA challenge
Body weight	264 ± 62 g	226 ± 13 g	0.65 (n. s.) not significant
Lung volume, both lungs	5.34 ± 1.68 ml	4.68 ± 1.81 ml	0.50 (n. s.)
Lung volume, right lung	3.59 ± 0.95 ml	3.09 ± 1.11 ml	0.53 (n. s.)
Lung volume, left lung	1.87 ± 0.97 ml	1.65 ± 0.77 ml	0.99 (n. s.)

Table 2. Volume and number of alveolar epithelial cells type II (AEII).

Parameter	Control n = 3	OVA challenge n = 3	P value control vs OVA challenge
Cell number(N(AEII, lung)	49 ± 2 × 10^6	44 ± 5 × 10^6	0.18 (n. s.) not significant
AEII numerical weighted mean cell volume	393 ± 29 µm^3	390 ± 29 µm^3	0.99 (n. s.)

Structural preservation

The alveolar surface was covered by more edematous fluid and more inactive components of the intra-alveolar surfactant in the asthmatics as described earlier¹⁰ and documented in Figure 3a–d.

Figure 3. Alveolar space, alveolar lining layer with hypophase. (a) Control lung: surface active subtypes in the hypophase (tubular myelin, tm) covered by the partly seen surface film, visible also partly in asthma induced lungs. Scale bar = 1µm. (b) Controls: surface active subtypes (tm) and surface inactive subtypes (unilamellar vesicles, ulv (asterisk) and multilamellar vesicles, mlv, arrow), visible also partly in asthmatic lungs. Scale bar = 1µm. (c) OVA sensitized and challenged animals: surface active subtypes and surface film (arrows), which are lifted off by intra-alveolar protein rich fluid accumulation (asterisks), visible only in asthma induced lungs. Scale bar = 1µm. (d) OVA sensitized and challenged animals: Extensive accumulation of inactive surfactant components (ulv and mlv, arrows), visible only in asthma induced lungs. Scale bar = 1µm.

AEII appeared structurally intact in controls and asthma-induced rats (Figure 4a, b). Subcellular components appeared normally distributed and showed no signs of inflammatory dependent damage. No observable enlarged or very small Lb as possible signs of disturbances in secretion or synthesis were visible.

Figure 4. Type II alveolar epithelial cells (AEII). (a) Control lung: AEII, structurally intact, with intact organelles. Lb appear normal in number and size. Scale bar = 2µm. (b) OVA sensitized and challenged lung: structurally intact AEII. Distribution and size of Lb profiles are qualitatively comparable with controls. No giant or very small Lb profiles or unnatural accumulation as signs of disturbances of intracellular surfactant synthesis are seen. Scale bar = 2 µm.

Number and volume of AEII were not affected in the BN rat asthma model (Table 2)

The total number of AEII did not differ in controls and OVA sensitized and challenged animals. The number-weighted mean volume of AEII was in the same range in both groups. Thus, a determination of subcellular densities in AEII was possible, without the risk of the reference trap leading to too high or too low densities.⁵⁰ Looking at the subcellular volume fractions of compartments presented in all animal cells (nucleus, cytoplasm, mitochondria), no alterations between both groups were evaluated (Table 3). Thus, quantitative data characterizing AEII were not influenced by induction of experimental asthma.

Table 3. Volume densities of subcellular organells of AEII.

Parameter	Control (n = 5) %	OVA Challenge (n = 7) %	P value control vs OVA challenge
Volume density of nuclei (VV(nc/AEII))	23.22 ± 2.57	23.35 ± 2.49	0.88 (n. s.) not significant
Volume density of mitochondiria ((VV(mito/AEII))	9.12 ± 2.47	7.01 ± 0.85	0.11 (n. s.)
Volume density of cytoplasm (VV(cyto/AEII))	49.38 ± 2.76	53.65 ± 1.97	0.06 (n. s.)

Size distribution, volume and number of Lb in AEII of OVA sensitized and challenged rats are compared to controls (Tables 4 and 5)

Table 4. Characterization of intracellular surfactant.

Parameter	Control (n = 5)	OVA challenge (n = 7)	P value control vs OVA challenge
Volume density of Lb ((VV(Lb/AEII))	15.97 ± 2.22 %	14.74 ± 1.15 %	0.15 (n. s.) not significant
Volume density of Cb ((VV(Cb/AEII))	1.30 ± 0.33 %	0.92 ± 0.16 %	0.07 (n. s.)
Volume density of Mvb ((VV(Mvb/AEII))	0.63 ± 0.12 %	0.63 ± 0.11 %	0.99 (n. s.)
Numerical density of Lb (NV(Lb,AE II))	0.22 ± 0.04	0.17 ± 0.06	0.46 (n. s.)
Lb, volume weighted mean volume	1.87 ± 0.19 µm^3	2.18 ± 0.56 µm^3	0.34 (n. s.)
Lb, number weighted mean volume	0.75 ± 0.12 µm^3	0.96 ± 0.36 µm^3	0.34 (n. s.)

Table 5. Total number and volume of Lb.

Parameter	Control (n = 3)	OVA challenge (n = 3)	P value control vs OVA challenge
Lb number (N(Lb, AEII))	93 ± 15	90 ± 5	0.8 (n. s.) not significant
Total Lb volume per AEII (Lb, AEII)	68 ± 2 µm^3	59 ± 5 µm^3	0.1 (n. s.)
Total Lb volume per lung (Lb, lung)	3 ± 0.6 × 10^6	2.5 ± 0.6 × 10^6	0.3 (n. s.)

To check whether there are possible disturbances in Lb, we determined different parameters characterizing their densities, size, volume and number.

Because the influence of the reference space is negligible, we listed the evaluated V_V. The V_V of profiles of intracellular surfactant storage organelles as well as the numerical density (N_V(Lb,AE II)) of Lb did not differ (Table 4).

The volume-weighted mean volume and the numerically weighted mean volume of Lb were not influenced by OVA sensitization and challenge (Table 4). Multiplying the N_V (Lb,AEII) with the volume of AEII of the same animal, the Lb number in AEII was determined (Table 5). AEII of controls contain 93 and of OVA sensitized and challenged rats 90 Lb without reaching significance. Values for total Lb volume and the total Lb surface did also not reach significance.

Discussion

Quantitative ultrastructural investigations by means of stereology of AEII and their Lb after allergen challenge, as presented here, have not been performed until now. Our results show that the allergic inflammation does not influence the intracellular surfactant pool, as assessed by morphological means. Numerically weighted mean cell volume and the number of AEII are com­parable in controls and OVA sensitized and challenged animals. The lamellar bodies, as storage organelles of surfactant, represent the morphometric correlate of the intracellular surfactant pool per lung. The total volume of Lb as well as the number of Lb exhibit no differences in OVA sensitized and challenged animals when compared with controls. Furthermore, we found no increase in OVA sensitized and challenged animals of the volume-weighted mean volume of Lb, indicating that the there was no increased variability in Lb volume. The comparable values for the numerically weighted mean volume of Lb also indicate no alterations in the mean volume of single lamellar bodies, independent of the asthma induction. Thus, OVA challenge neither influences the mean variability nor the mean particle volume of Lb. The size distribution of Lb was comparable in both groups. Because of our data, we suggest that neither more or less Lb were synthesized at the smooth ER, nor increased or less amounts of Lb were secreted by exocytosis.

Our results show that there are no clearly visible structural changes of the intracellular surfactant in the asthma model of the Brown Norway rat, which may be responsible for the surfactant dysfunction, as documented by the increase of surface tension and airway resistance in the same asthma model.¹⁰ Therefore, we suppose that the inflammatory reaction after allergen challenge induce mainly alterations in the intra-alveolar surfactant pool, as demonstrated by us and others.^{10,33,34,37,39}

It is already accepted that asthma induced inflammation spreads not only into small airways but also into the gas-exchanging alveolar region of lungs.^25,27,29,42 Up to now, numerous authors verified that the inflammation is accompanied by an surfactant dysfunction using animal models of asthma or BALF of patients suffering from asthma. They proved an increase of surface tension and an increase of inactive surfactant components in BALF harvested from asthma induced lungs found as one responsible cause.^{10,34,35,37,54,55}

Surfactant dysfunction as a result of augmented inactivation of intra-alveolar surfactant in OVA sensitized and challenged rats may be caused by 1) an IL-4-driven SP-B deficiency after allergen provocation,⁵⁴ 2) secretion products of invaded eosinophils, such as ECP,³⁹ 3) inflammatory proteins and or plasma-derived, such as albumin and fibrinogen,^37,56,57 4) and by developing inflammatory dependent alveolar edema.^58,59 The accumulation of infiltrating mediators of different origins and their interaction with aggregates of alveolar surfactant⁶⁰ may accelerate the conversion from active to inactive surfactant components and increase the surfactant dysfunction.³²

Alterations in the composition of phospholipids, which are not visible microscopically, may also influence the intra-alveolar surfactant activity. e.g., a diminution in phosphatidylglycerol (PG) correlated with increased surface tension combined with an increase of inactive surfactant components.³³ Furthermore, it was suggested that the loss of PG by, for instance. increased phospholipase activity, may contribute more to surfactant dysfunction than alterations of surfactant proteins, loss of posphatidylcholine or increased levels of lysophospholipids, or the presence of increased numbers of eosinophils.⁶¹ In asthmatic animals compared to controls, decreased surfactant pool sizes and saturated phosphatidylcholine to protein ratios in BALF may predominantly be induced by leaked serum proteins.³² Summarizing the results of other authors, asthma induced inflammation predominantly disturbs the intra-alveolar surfactant pool in different ways.

Lb secretion is regulated by chemical stimuli, such as β-agonists, cholinergic agonists and adenosine, or physical stimuli, such as ventilation.^16,62 Furthermore, components of the allergic inflammatory metabolism, such as IL-1 or TNF-α,^30,34,63 are also able to influence the secretion of surfactant.¹⁶ Thus, disturbances of synthesis, storage and secretion are conceivable. Liu et al. proved a decreased synthesis, storage and activity of phospholipids, measured on isolated AEII.⁶⁴ They found a significant decrease in cellular phospholipids and phosphatidylcholine synthesis in AEII isolated from immunized and challenged guinea-pigs. The phospholipid content was 33% of control values and the rate of phosphatidylcholine synthesis less than 35%.⁶⁴ However, these biochemical alterations are not accompanied by clearly visible changes in the morphological correlates of the intracellular surfactant pool.

Taken our results together, the present study has shown that in the asthma model of the Brown Norway rat neither the AEII nor the storage organelles of surfactant exhibit any damage or significant quantitative alterations, as determined by stereological methods. However, a limiting factor of this study is that we used only male rats for our experiments.

Author contributions

AS designed the study, performed the light microscopical evaluation and wrote the manuscript, SF performed the electron microscopical evaluation, TT performed the instillation fixation and JMH performed the experiments. All authors reviewed the manuscript.

Acknowledgements

The authors gratefully acknowledge the expert technical assistance of A. Gerken, S. Freese, H. Hühn (Göttingen), M. Hanke (Hannover), M. Fathollahy (Hannover) and S. Elki (Hannover) and thank S. Schmiedl (M.Ed.) for the linguistic review of the manuscript. Finally, the authors thank Prof. Dr. M. Ochs (Berlin) very much for his support in sample processing as well as for critical reading and for his constructive suggestions improving the manuscript.

Disclosure statement

In accordance with Taylor & Francis policy and our ethical obligation as researchers, the authors declare that they have no known competing financial interests or business interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Funding

Part of this work was supported by the Deutsche Forschungsgemeinschaft, SFB 587.