The Challenge of Long-Term Cultivation of Human Precision-Cut Lung Slices

Abstract

Human precision-cut lung slices (PCLS) have proven to be an invaluable tool for numerous toxicologic, pharmacologic, and immunologic studies. Although a cultivation period of <1 week is sufficient for most studies, modeling of complex disease mechanisms and investigating effects of long-term exposure to certain substances require cultivation periods that are much longer. So far, data regarding tissue integrity of long-term cultivated PCLS are incomplete. More than 1500 human PCLS from 16 different donors were cultivated under standardized, serum-free conditions for up to 28 days and the viability, tissue integrity, and the transcriptome was assessed in great detail. Even though viability of PCLS was well preserved during long-term cultivation, a continuous loss of cells was observed. Although the bronchial epithelium was well preserved throughout cultivation, the alveolar integrity was preserved for about 2 weeks, and the vasculatory system experienced significant loss of integrity within the first week. Furthermore, ciliary beat in the small airways gradually decreased after 1 week. Interestingly, keratinizing squamous metaplasia of the alveolar epithelium with significantly increasing manifestation were found over time. Transcriptome analysis revealed a significantly increased immune response and significantly decreased metabolic activity within the first 24 hours after PCLS generation. Overall, this study provides a comprehensive overview of histomorphologic and pathologic changes during long-term cultivation of PCLS.

Figure 1

Illustration of the criteria used for histologic grading of human precision-cut lung slices (PCLS). The PCLS and the respective grading are shown broadly and with a magnification (insets) of the characteristics concomitant with the grading. A detailed description of the criteria can be found in the text (Materials and Methods). Briefly, the alveolar epithelium was scored from A0 (best) to A3 (worst), the endothelium was scored from E0 (best) to E3 (worst), and the bronchial epithelium was scored from B0 (intact) to B1/2 (detached). Scale bars: 100 μm (A0–A3 and B0–B1); 50 μm (E0–E3); 20 μm (insets). Original magnifications: ×100 (A0–A3 and B0–B1); ×200 (E0–E3); ×400 (insets).

Figure 2

Viability of long-term cultivated human precision-cut lung slices (PCLS) as measured by the water soluble tetrazolium (WST)-1 (for quantification of cell proliferation and viability) and lactate dehydrogenase (LDH) (for quantification of cell death and cell lysis) assays. PCLS were derived from immediately processed or overnight stored fibrotic explants as well as tumor lung resections. WST-1 values are shown as the corrected absorbance, and LDH values are represented as the relative amount of released LDH activity compared with the totally lysed control. Data are expressed as means ± SD. ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. n, number of PCLS; N, number of lungs.

Figure 3

Preservation of ciliary beat in human precision-cut lung slices (PCLS) during long-term cultivation. The PCLS derived from immediately processed or overnight stored fibrotic explants as well as tumor lung resections. Ciliary beat was assessed on each day of medium change, and the relative number of PCLS with ciliary beat was calculated compared with the initial number of PCLS with ciliary beat. Weighted means ± SD were used for the statistical analysis of all data points. Graphs were generated from weighted means of a limited number of data points (week 1 = day 7 ± 1, week 2 = day 14 ± 1, week 3 = day 21 ± 1, week 4 = day 28 ± 1) for illustration purposes. Data are expressed as means ± SD. ∗P < 0.05. n, number of PCLS; N, number of lungs.

Figure 4

Histomorphologic changes in long-term cultivated human precision-cut lung slices (PCLS) derived from immediately processed or overnight stored fibrotic explants as well as tumor lung resections. Morphologic changes were analyzed considering the three major structural compartments of a lung slice (alveolar epithelium, bronchial epithelium, and endothelium). The grading ranges from 0 to 3 (bronchial epithelium, 0 to 2), with 0 displaying the best and 3 the worst condition (a detailed description can be found in Materials and Methods). Each day shows the weighted mean fraction of PCLS with respective grading. n, number of PCLS, N, number of lungs.

Figure 5

Exemplary illustration of long-term cultivated human precision-cut lung slices (PCLS) with foci of squamous metaplasia predominantly found in the alveolar epithelium and rarely adjacent to bronchial epithelium. A: Low-power magnification of a PCLS severely affected by keratinizing squamous metaplasia (hollow arrows) after 28 days of cultivation. Metaplasia is found in the alveolar epithelium distal from normal-appearing bronchiole (B) and with normal-appearing alveolar septum (asterisks) in between. B: Image section of a normal-appearing bronchiole (B) next to a normal-appearing alveolar septum (asterisks) and distal to alveolar epithelium affected by squamous metaplasia (hollow arrow). C: High-power magnification of a bronchiole (B) with absent cilia and a transition to keratinizing squamous metaplasia (hollow arrow). D: Overview of a PCLS with apparent foci of squamous metaplasia and keratinization. E: High-power magnification of squamous metaplasia with prominent keratinization. F: Higher magnification with narrowed condenser aperture showing the characteristic features of squamous metaplasia with keratinization: Prominent nucleoli (arrows), keratohyaline granules (spiky arrows), corneocytes (directional arrows), stratum granulosum (single dagger), and stratum corneum (double dagger). Scale bars: 200 μm (A); 100 μm (B and D); 50 μm (C); 20 μm (E and F). Original magnifications: ×40 (A); ×100 (B and D); ×200 (C); ×400 (E and F).

Figure 6

Occurrence of squamous metaplasia and keratinization in long-term cultivated human precision-cut lung slices (PCLS) derived from immediately processed or overnight stored fibrotic explants as well as tumor lung resections. Each day shows the weighted mean fraction of PCLS in which squamous metaplasia and keratinization occurred. Keratinization was never observed without the occurrence of squamous cell metaplasia. n, number of PCLS; N, number of lungs.

Figure 7

Transcriptome analysis of long-term cultivated human precision-cut lung slices (PCLS) using RNA sequencing and pathway enrichment analysis. The PCLS used for sequencing originated from three tumor resection lungs. A: Heat map and clustering of genes calculated from the differential expression of the top 1000 genes. Every lane represents a different lung specimen and/or a different day of sampling. In addition, samples were taken from lungs before filling with agarose (d0 BF) and after filling with agarose (d0 AF). B: Enrichment analysis [Gene Ontology (GO): biological function] of the gene sets belonging to the respective cluster for the identification of overrepresented pathways (Supplemental Figure S3). The number of genes significantly associated with the respective pathways of a cluster are written in brackets. On the basis of the identified pathways, each cluster was manually assigned to a higher-order category. C: RNA integrity number (RIN) of RNA isolated from PCLS (fibrotic and tumor lung resections) at different points in time during long-term cultivation. FDR, false discovery rate; N, number of lungs.

Figure 8

Abstraction of all findings, including all data regardless of the tissue type (fibrotic or tumor resection lungs) or processing (immediate processing or overnight storage). Viability (measured by lactate dehydrogenase and water soluble tetrazolium-1 assays), tissue integrity (endothelium and bronchial and alveolar epithelium), and transcriptome analysis (RNA sequencing pathway enrichment analysis) was visualized manually. The graphs for ciliary beat and the histopathologic findings [squamous metaplasia and keratinization] were generated from weighted means, including precision-cut lung slices (PCLS) from all conditions (ciliary beat: N/day = 13 to 15, n/day = 386 to 487; histopathologic findings: N/day = 16, n/day = 249 to 340) at the specific point in time (ciliary beat: days 0, 7 ± 1, 14 ± 1, 21 ± 1, and 28 ± 1; histopathologic findings: days 1, 8, 14, 21, and 28). n, number of PCLS; N, number of lungs.