This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Identification of Molecular and Functional Heterogeneity of Epithelial Progenitor Cells in the
Upper Airway
by
Monica Allison Clifford
A thesis submitted in conformity with the requirements
Identification of Molecular and Functional Heterogeneity of Epithelial Progenitor Cells in the
Upper Airway
Monica Allison Clifford
Master of Science
Medical Biophysics University of Toronto
2013
Abstract
Upper airways are lined with a pseudostratified mucociliary epithelium
maintained by basal cells. To investigate functional and phenotypic heterogeneity
within the human basal cell compartment, we used a combination of limiting dilution
assays and surface marker profiling on primary cultures of basal cells with verified
progenitor activity. The limiting dilution assay suggested functional heterogeneity in the
ability of basal cells to repopulate a filter and maintain a barrier at ALI. The frequency of
cells with this activity varied between patient strains and ranged from 0.08%-1% of basal
cells. Validation of large-scale comprehensive surface marker profiling on basal cells led
to identification of 74 antigens demarking consistent subpopulations. Preliminary
functional analyses suggest differences in differentiation potential of some
subpopulations. This work supports the idea that the basal cell compartment may be
functionally heterogeneous, and provides new molecular tools for interrogation of
human basal cells.
iii
Acknowledgments
It is a great pleasure to thank the many people who have made this thesis possible.
It is difficult to overstate my gratitude to my supervisor Dr. Nadeem Moghal. With his
enthusiasm and passion for science he inspired me and with his patience and knowledge he
taught me. Throughout my degree and thesis preparation he provided encouragement, sound
advice and lots of good ideas. I would have been lost without him.
I wish to thank my lab mates, Boram Kim and Emily van de Laar, and my many colleagues for
their support, guidance, assistance and companionship.
I am grateful to my committee Dr. Sean Egan and Dr. Norman Iscove for their wisdom and
guidance throughout my degree.
Finally, I wish to thank my family, friends and the many people who provided me with the
support I needed to succeed and who made this experience memorable.
iv
Contents
Abstract ............................................................................................................................................ ii
Acknowledgments ........................................................................................................................... iii
List of Tables .................................................................................................................................... iv
List of Figures .................................................................................................................................. vii
List of Appendices.......................................................................................................................... viii
List of Abbreviations ........................................................................................................................ ix
mouse (1:500). Samples are washed 3 times with 0.1% Triton X-100 in PBS. Tissue
sections and cytospun preparations were mounted in vectashield mounting media with
DAPI.
2.7 Enface staining
For a quantitative assessment of differentiation, enface staining technique was
established. ALI membranes are formalin fixed as described above. ALI membranes are
stained as described above; antigen retrieval is not required for βTUBIV staining. Upon
completion of staining membranes are washed three times in 0.1% Triton X-100 in PBS
then dehydrated through an alcohol series from 70% for approximately 1 minute, to
90% ETOH briefly and rinsed in 100% ETOH to remove any residual water from the
surface of the ALI. The membrane is carefully cut from the well insert with a sharp
scalpel while keeping the membrane pressed against the cutting surface. In a fume
hood, the membrane is placed in a drop of 30% Permount in xylene on a pre-cleaned
glass microscope slide, epithelial side up. More Permount-xylene solution is added on
19
top of the membrane and a cover glass is carefully placed to avoid bubbles and flatten
the membrane, the slide is dried in a fume hood overnight. Imaging of enface staining
was performed using Metamorph® tiling utility.
2.8 High- Throughput Antibody Screen
Cells were expanded under submerged conditions to 70% confluence as
P0plastic. Cells were trypsinized with 0.0125% trypsin+EDTA (Clonetics), neutralized
with TNS, pelleted and resuspended in Hanks buffered salt solution with 1% FBS. Cells
were submitted to the antibody core facility. Cells were then stained with the panel of
antibodies in a 96-well plate format; marker expression was analyzed by flow cytometry.
2.9 RNA Extraction and cDNA Synthesis
Total RNA was extracted using Ambion Micro RNA kit, RNA isolate was treated
with DNase I to remove contaminating genomic DNA. A total of 1ug of RNA was
converted into cDNA in a 20ul reaction volume using the High Capacity cDNA Reverse
Transcriptase kit (ABS). An automated thermocycler was used to provide the following
conditions: 10 minutes at 25°C, 120 minutes at 37°C, 5 seconds at 85°C and 4°C until
transferred to -20°C for long-term storage.
2.10 Real-Time PCR
cDNA stock solution was prepared by diluting cDNA to a concentration of 2 ng/µl
in autoclaved miliQ water. The SYBR Green system (Biorad) was used for qPCR
reactions. 4 µl of diluted cDNA was added to a reaction mixture containing 5 µl of SYBR
green super mix (Biorad) and 1 µl of primer mix containing 3µM of forward and reverse
20
primers (Appendix B). Reaction conditions for detection of amplification were 95°C for 3
minutes, 95°C for 10 seconds, and 60°C for 30 seconds for 40 cycles.
2.11 Statistical Analysis of Limiting Dilution Analysis Results
Stem/progenitor cell occurrence follows a Poisson distribution. Limiting dilution
analysis results are analyzed by a webtool: Extreme limiting dilution analysis [50]. The
program enables inclusion of densities that result in all failures or all successes. Results
are based on the assumption of a linear relationship between seeding density and
outcome.
3. Results
3.1 Early passage HTECs are p63-positive and retain regenerative properties
To begin to address the degree of molecular heterogeneity within the human basal cell
compartment, we used previously described methods to extract large airway progenitor
cells from adult human tracheal tissue [51-53]. Although our goal was to analyze basal
cells in as close to an in vivo state as possible, we found that the most effective means
of isolating these cells also cleaved many cell surface markers. To bypass this problem
and increase the yield of cells for high-throughput assays we expanded enzymatically
digested, P0 tracheal suspensions without passaging, in a serum free medium previously
described to maintain large airway progenitor activity [47]. After expansion, we
evaluated the purity of this P0-plastic population by immunofluorescent (IF) antibody
staining for basal and columnar cell markers. We used α-p63, α-βTUBIV, α-MUC16 and
21
α-Muc5ac to assess the presence of basal cells, ciliated cells, mucinous cells and goblet
cells respectively. Staining showed approximately 90% of cells were positive for p63
(Fig. 1a). No cells had polarized apical βTUBIV staining, a hallmark feature of ciliated
cells, though a small percentage stained weakly for βTUBIV in cytoplasmic foci (Fig.
1b) [54,55]. No cells expressed MUC16 (fig. 1c) or Muc5ac (data not shown). To
assess if p63-negative nuclei represented contaminating cell types, we examined
another basal cell marker, CD44 by FACS analysis, showing that P0plastic cells are
~100% positive for CD44 (fig. 1e) [12]. Based on CD44 staining data, we re-
examined p63 expression by IF and found a number of nuclei initially scored
negative actually had low levels of p63. Together these data indicate that P0plastic
cells represent a close to pure population of basal cells, with a low amount of
heterogeneity in p63 expression.
To verify that proliferating P0plastic basal cells retained progenitor activity,
we seeded cells derived from different donor tracheas into rat tracheal xenograft
(Strain 37)[37] and ALI culture (Strain 38 and 23)[38] systems, which support the
differentiation of basal cells into columnar cells. These cells formed well-
differentiated epithelia in xenograft and ALI cultures showing they retained
functional in vivo progenitor properties (fig. 2a, b; fig 3a). The presence of ciliated,
secretory and basal cells was confirmed in ALI-culture sections stained for βTUBIV
(fig. 2c; fig. 3b), MUC16 (fig. 2d; fig. 3c), and p63/CD44 (fig. 2e; fig. 3d). Although
other studies have reported differentiation of basal cells into Muc5ac-positive goblet
cells under ALI culture conditions, we did not detect this lineage (fig. 2f; fig. 3e)[39].
22
Instead, we found robust differentiation into MUC16-positive mucinous cells,
another abundant in vivo mucinous large airway lineage [56].
Figure 1. P0plastic HTECs are a purified population of basal cells. Cytospun
unpassaged, HTECs expanded in submerged condition on collagen coated plastic (P0p)
stain positively for (A) basal cell marker p63 (86.3 +1.2%), but do not express (B) tubIV
(0%) or (C) Muc16 (0%). (D) Percentage of cells expressing each marker. (E)
Expression of basal cell marker, CD44, was evaluated by FACS on P0p cells.
23
Figure 2. P0p cells retain the ability to differentiate in xenograft and ALI cultures.
(A) P0p cells were seeded into denuded rat tracheas and implanted into NOD/SCID
mice. After 33 days, xenografts were fixed, sectioned and stained by H&E (Str. 37). (B-
F) P0p cells were grown at ALI cultures for 21 days, after which filters were fixed,
sectioned and stained with H&E and indicated antibodies (Str. 38). All scale bars
represent 50 m.
24
Figure 3. P0p cells retain the ability to differentiate in ALI cultures (Str.23). (A-E)
P0p cells were maintained at ALI for 21 days, after which filters were fixed, sectioned
and stained with H&E and indicated antibodies (Str. 38). All scale bars represent 50 m.
3.2 Ciliogenesis can be used to evaluate culture performance at ALI.
To quantify p63-positive progenitor activity in ex vivo cultures we focused on the
ciliated cell lineage, which is a major lineage derived from p63-positive progenitors [16].
Three methods to evaluate the extent of ciliogenesis were explored: 2D sections, enface
staining and gene expression analysis.
Intracellular localization of βTUBIV at the apical surface is a definitive marker for
mature ciliated cells [54,55]. However, the distribution of βTUBIV-positive cells is not
uniform in differentiated ALI cultures (fig. 4a). This can be seen in sections from the
25
same membrane where in some sections ciliated cells are sparsely distributed (fig. 4a),
while in others they are densely distributed (fig. 4b). In contrast, basal cells, as marked
by p63, appear uniformly distributed as assessed by random sections of the same
membrane (fig. 4c and d). The impact of uniformity of cellular distribution on
estimating the mean number of cells is demonstrated by comparing standard deviations
in the percentages of ciliated cells and basal cells (fig. 4e). The percentage of total cells
having undergone ciliogenesis was 10.4 ±5.8%, compared to 25.4 ±2.7% for evenly
distributed basal cells. Thus, without extensive sectioning, random sections do not
readily allow for accurate extrapolation of the extent of differentiation for an entire
membrane, making it difficult to quantify differences in ciliogenesis between samples.
To circumvent this problem, we sought to establish a quantitative method to assess the
entire membrane for ciliated cells.
An enface staining technique was adopted as a non-biased method of
determining the extent of ciliogenesis. The membrane is stained for βTUBIV and images
of the entire membrane are captured and stitched together using Metamorph® tiling
software (fig. 4f). Software analysis of the enface image provides either: (1) a count of
ciliated cells, which is useful under conditions where limited differentiation occurs; or
(2) a percentage of reconstituted epithelium that is covered by cilia. In theory, this
approach accurately evaluates the extent of ciliated cell differentiation across the entire
culture, and should allow for quantitative comparison between cultures.
26
Figure 4: Accurate evaluation of ciliogenesis is possible by enface staining and
FoxJ1 expression, but not by 2D ALI sections. (A,B) 2D sections stained for tubIV show irregular distribution of ciliated cells, in contrast, (C,D) p63-positive cells are evenly distributed. (E)The variability in percentage and frequency of positive cells as calculated
based on staining of 2D sections gives low confidence in values for tubIV when compared to normally distributed p63. (F) Enface staining allows quantitative determination of total ciliogenesis. (G) Analysis of gene expression provides a quantitative method to compare induction of FoxJ1 between cultures, physiologic controls (tracheal cells) and baseline controls (P0plastic basal cells). Scale bars in IF
images: 50 m; enface image: 1 mm.
27
In addition to βTUBIV staining, we also evaluated FOXJ1 gene expression in
differentiated cultures (fig. 4g). FOXJ1 is a transcription factor involved in ciliogenesis in
various tissues; in the airway epithelium, its expression is restricted to ciliated cells [57].
Gene expression analysis by qPCR has the advantage of being very sensitive and, like
enface staining, uses the entire population of cells from the culture. qPCR analysis does
not permit determination of ciliated cell number, but we have found that induction of
FOXJ1 gene expression tracks with emergence of ciliated cells (fig 6g-l, and 7c). Gene
expression analysis allows for direct quantitative comparison between samples, as well
as comparison to physiological (tracheal cells) and base line control cells (P0plastic,
starting population). Our qPCR analysis of FOXJ1 in ALI cultures revealed that P0plastic
basal cells retained a robust ability to induce FOXJ1 to in vivo, tracheal levels (fig. 4g).
3.3 Kinetic analysis of mucociliary differentiation of P0-plastic cells seeded at
ALI.
In order to characterize the differentiation potential of putative subpopulations
of basal cells, we analyzed the kinetics of mucociliary differentiation from bulk P0-plastic
cultures at ALI. P0plastic progenitors isolated from three different donor tracheas were
grown at ALI, and cultures evaluated for differentiation markers at six stages of
repopulation and differentiation. The stages examined were: proliferating subconfluent
cultures, covering 50-70% of filter, cultures just achieving confluence, which is the point
when media was removed from the upper chamber and cells were exposed to air, 3
days post-confluence, 7 days post-confluence, 14 days post-confluence, and 21 days
28
post-confluence. Cultures were evaluated by IF staining on sections, gene expression
analysis, and enface staining.
As p63 progenitors differentiate, we expect to see a shift from a population of
basal cells to a heterogeneous culture containing mixed lineages. Thus, we first
examined changes in the proportion of basal cells by immunostaining sections for p63.
We found that at subconfluence and at confluence nearly all nuclei are p63-positive (fig
5a, b). As early as 3 days post-confluence, a p63-negative layer of nuclei was seen
above the basally located p63-positive nuclei, consistent with pseudostratification (fig.
5c). This staining pattern of basally located p63-positive nuclei and apical p63-negative
nuclei persisted until the cultures reached maturity at 21 days post-confluence (fig. 5d,
e, and f). To investigate the contributions of TAp63 and Np63 to immunostaining and
to quantify p63 expression, we looked at gene expression levels of both p63 isoforms.
We saw the expected downward trend in expression of Np63; however, in one patient
strain there was an initial increase leading up to confluence, prior to its eventual decline
(fig. 6a). In contrast, TAp63 initially increased in all patient strains, and declined
following confluence in 2 of 3 patient strains (fig. 6b). All progenitor strains showed the
expected decline in Np63 expression that would occur during differentiation.
However, our data also suggest there may be some strain-dependent differences in
regulation of p63 isoforms during differentiation at ALI. The significance of these
differences is not known.
29
30
Figure 5: Qualitative assessment of differentiation in ALI culture at multiple time
points by immunofluorescence. (A-F) p63, (G-L) tubIV, and (M-R) Muc16 expression is assessed at multiple time points during repopulation and differentiation of ALI cultures by P0p cells. Expression of p63 in: (A) subconfluent cultures, (B) at confluence, and (C-F) in post confluent cultures, p63 staining is seen only in basally located cells. (G) Some
cells in subconfluent cultures have cytoplasmic tubIV foci, (H-I) these are not observed at confluence or 3 days post-confluence, (J) appearance of pre-ciliated cells is seen at 7 days post-confluence, and (K-L) mature ciliated cells are present at 14, and 21 days post-confluence. (M-O) Muc16 is not seen until 3 days post confluence, (P-R) and is
maintained until 21 days post-confluence. Scale bars: 25 m.
Polymerization of βTUBIV into apically projecting microtubules of cilia is a
definitive marker of the airway ciliated cell lineage [54,55,58]. In sections of
subconfluent, confluent, and 3 day post-confluence ALI cultures, we observed some
cytoplasmic βTUBIV staining, but no apical polymerization (fig. 5g, h, and i). At 7 days
post-confluence, intracellular accumulation of high levels of βTUBIV was seen in some
cells, presumably those undergoing ciliogenesis (fig. 5j). However, only at 14- days and
21-days post-confluence did apical localization of βTUBIV appear, indicating mature
ciliated cells (fig. 5k, l).
Similar kinetics for ciliogenesis were observed by following FOXJ1 expression by
qPCR. FOXJ1 expression was first detectable at 3-7 days post-confluence (fig. 6c), and
reached over 700% of physiological levels in two of three patient samples by 21 days
post-confluence (Fig 6c). Although the extent of induction varied substantially between
patient strains, the kinetics of ciliogenesis were similar. Enface staining for βTUBIV also
showed kinetics similar to the emergence of FOXJ1 which is consistent with FOXJ1’s late
role in ciliogenesis [59]. By enface staining, 3-day post-confluent cultures also showed
intracellular βTUBIV staining, but no mature ciliated cells (data not shown). At 7 days
31
post-confluence, a limited number of mature ciliated cells were apparent, and by 14
days and 21 days post-confluence, cultures had large numbers of ciliated cells (fig. 7e-g).
Monitoring culture progression by light microscopy, we consistently observed
mucous-like secretions on the surface of differentiating cultures prior to detecting any
sign of ciliogenesis. To examine for emergence of mucinous lineages, we followed
expression of MUC16, which in vivo, marks abundant surface and glandular tracheal
cells [56]. In subconfluent cultures, and when confluence was first achieved, we saw no
reactivity with an α-MUC16 antibody (fig. 5m, n). However, as early as 3 days post-
confluence, we observed cell-specific staining at the apical surface of the epithelium (fig.
5o), which became more intense and more wide-spread by 1 week post confluence (fig.
5p). After this period it became difficult to determine which cells were unambiguously
MUC16 positive, possibly due to the antigen being clipped and spread across the
epithelial surface (fig. 5q, r). By qPCR, MUC16 gene expression was first detectable at
confluence, but at less than 1% of physiological levels (fig. 6d). At 3 days post-
confluence, expression in two of three patient samples reached over 10% of physiologic
levels, while in the third strain expression had reached less than 2% of these levels. In
subsequent weeks there was strain to strain variability in gene expression. In Str.
30277, MUC16 expression peaked at 7 days, reaching ~80% of physiologic levels but
then declined (fig. 6d). In Str. 78297, MUC16 levels rose to over 200% of physiological
expression by day 21 (fig. 6d). In Str. 39, MUC16 expression plateaued on day 14, at
25% of physiologic levels (fig. 6d). Our data indicate that for quantification of mucinous
cell numbers by antibody staining, cultures between 3 and 7 days post confluence are
32
optimal. However, for greater sensitivity, qPCR on bulk populations of cultures older
than 7 days post confluence is better suited.
33
Figure 6: Significant changes in gene expression occur during differentiation at
ALI and vary between patient samples. Gene expression changes in (A) Np63, (B)
TAp63, (C) FOXJ1 and (D) MUC16 were monitored during repopulation and
differentiation of P0p cells in ALI cultures. Enface staining for TUBIV showing (E) a
limited number of ciliated cells at 7 days post-confluence, (F, G) but substantial ciliated
cell coverage at both 14 and 21 days post-confluence.
3.4 ALI cultures may maintain a progenitor population during differentiation.
Serial passaging of HTECs on 2D plastic, under submerged culture conditions has
been reported to result in attenuation of basal cell progenitor capacity as assessed in ALI
culture [38]. Gray et al. demonstrated that NHTBE cells grown to a P3 plastic stage
retain the ability to functionally differentiate at ALI into an epithelium possessing many
properties of the in vivo epithelium including mucociliary differentiation and tight
junction formation [38]. For large scale or long-term studies it would be of great
interest to preserve progenitor activity ex vivo. The ALI culture system recapitulates an
in vivo environment by creating an air-liquid interface where cells are fed basolaterally
and exposed to air at the apical surface. This environment, combined with factors in the
media, leads to mucociliary differentiation of p63-positive basal cells, and formation of a
pseudostratified epithelial layer. It is unknown if this in vivo-like environment is better
able to maintain stem/progenitor cells than 2D-culture. Alternatively, differentiation
signals could push the stem/ progenitor cell population toward a terminally
differentiated fate, where basal cells seen in ALI cultures serve only a structural role in
epithelial anchoring.
34
35
Figure 7. Differentiation of basal cells at ALI can maintain a progenitor
population. (A) P0plastic cells were passaged on either ALI or plastic, re-equilibrated
on plastic and their ability to differentiate at ALI was re-assessed; arrows indicated
points where samples were collected. Differentiated cultures grown from either plastic-
passaged (P2pA) or ALI-passaged (PAPA) cells were assessed for expression of: (B)
FoxJ1, (C) Muc16, (D) Np63 and (E) TAp63. Enface staining for tubIV show extent of
ciliogenesis in (F) P2pA, plastic-passaged and (G) PAPA, ALI-passaged cultures.
To address this question, we compared progenitor activity of ALI-passaged and
plastic-passaged cells relative to the P0plastic progenitor population. Initially, after
expansion, P0p cells were seeded in replicate at ALI or passaged on plastic as P1 cells
(fig. 7a). After a well differentiated epithelium was generated at ALI (3 weeks), or P1
plastic cells reached 70% confluence, CD44+ basal cells were FACS purified from the
cultures (fig. 7a). To control for potential changes in progenitor activity due to culturing
in the different media associated with ALI vs. plastic culture we re-equilibrated both ALI
and plastic sorted cells in LHC-9 media on 2D plastic before seeding the P3 cells into ALI
for final analysis (fig. 7a). Both PAP and P2p cells proliferated in ALI culture wells at
comparable rates to form a confluent monolayer. After establishment of the ALI, both
populations maintained a barrier.
We next examined ciliated cell lineage potential of PAP and P2p progenitor cells.
In general, passaging under both conditions reduced ciliogenic differentiation. FOXJ1
induction was significantly reduced under both conditions in all three stains relative to
P0pA (fig. 7b). However, in Str. 30277, P2p basal cells had significantly larger induction
of FOXJ1 gene expression at ALI. These data indicate the ability to give rise to ciliated
cells is significantly impacted by prolonged culture under both conditions (fig. 7b). A
36
dramatic decline in extent of ciliogenesis from P0plastic to P2plastic cells was seen by
enface staining (fig. 7f and g); given the deterioration of P2pA cultures in 2 of 3 patient
strains, we were unable to attribute failure of PAPA cultures in these 2 strains to their
previous differentiation at ALI. In Str. 39 we observed ciliary differentiation of PAPA
cultures at levels comparable to P2pA cultures (fig. 7g, h). MUC16 was induced at levels
similar to P0pA under both treatment conditions in all strains, suggesting that MUC16
lineage potential is maintained during the culture period examined (fig. 7c). Evaluation
ofNp63 and TAp63 showed no consistent difference following treatment conditions
(fig. 7d and e). These results demonstrate that both plastic and ALI culture can maintain
MUC16 competent lineage potential. However ALI is not better than plastic for
maintaining ciliogenic potential, and may be worse.
3.5 Functional heterogeneity is suggested by limiting dilution analysis
While some work supports the concept of functional heterogeneity within the
basal cell compartment, definitive evidence for a hierarchical organization is still lacking
[28,40,45]. To examine potential functional heterogeneity in human basal cells, we
attempted to quantify the frequency of ALI repopulating cells in limiting dilution assays.
We hypothesized that cells which could: proliferate, form tight junctions, and
differentiate on ALI, would have properties equivalent to in vivo progenitors. P0plastic
cells were seeded on ALI filters at seeding densities ranging from 10 000 to less than 100
cells per membrane. Cultures were assessed on their ability to: form a barrier, to
37
maintain a barrier once exposed to air, and to differentiate as assessed by the presence
of beating cilia. We employed the ‘Extreme Limiting Dilution Analysis’ (ELDA) webtool
(http://bioinf.wehi.edu.au/software/elda/) to determine the frequency of a cell that
met these criteria [50].
In order for ELDA to provide an accurate estimate of frequency, it requires a
mixed outcome at a single seeding density on replicate filters. Some filters must show
success, while others show failure [50]. Furthermore, ELDA analysis assumes a single-hit
model, in which a single cell, the stem cell, is the only factor that affects success of the
culture; therefore the single hit hypothesis must be met before frequency of a
stem/progenitor cell can be estimated [50]. If additional factors or parameters that
affect the success of a culture can also be diluted, ELDA cannot be used to quantify the
frequency of a single progenitor population [50].
We initially used presence of functional cilia, directly assessed by light
microscopy, as the output for success and the indicator of presence of ALI-repopulating
cells. However, in experiments using basal cells derived from 11 donor tracheas, none
provided sufficient data to test the single-hit criteria (table 1) and allow estimation of
the frequency of ciliogenic progenitors. In no case did we observe a mixed response of
cultures successfully differentiating and failing to differentiate at a single seeding
density (Appendix C). Although there was strain to strain variability, we did determine
that in all cases, a seeding density of 6 000 cells/ filter was sufficient for success at ALI
(table 1). When success was re-defined to include those cultures that only reached
38
confluence and formed tight junctions, but did not undergo differentiation, 5 patient
strains gave sufficient data to test the single hit hypothesis. In all of these patient
strains, the single hit criterion was met. The estimated frequency of a repopulating cell
ranged from as high as 1 in 96 in one patient strain, to as low as 1 in 1297 in another
(Table 2), suggesting human basal cells are functionally heterogeneous in their ability to
repopulate an ALI filter (Table 3).
Table 1: LDA results are unable to support that differentiation at ALI is
dependent on a single factor. P0plastic cells were seeded in replicate filters at
multiple dilutions and grown to confluence while being fed apically and
basolaterally. If confluence was reached, cultures were exposed to air apically
and fed basolaterally until differentiation was observed or culture failed to
differentiate. Those cultures which underwent mucociliary differentiation were
scored as successes and ELDA [50] was used to determine frequency of cell that
leads to success. No experiment satisfied the criteria of the statistical test and
frequency could not be estimated. Experimental data is available in Appendix C.
Lowest density at which differentiation occurred is presented.
Patient Sample
Reject single-hit hypothesis (p<0.05)
Lowest successful seeding density
18 Insufficient data 10 94214 Insufficient data 100 78297 Insufficient data 250 79168 Insufficient data 500 39 Insufficient data 500 23 Insufficient data 1000 11111 Insufficient data 1000 38 Insufficient data 1500 29945 Insufficient data 2000 32 Insufficient data 6000 55 Insufficient data 6000
39
Table 2: LDA results show frequency of a cell capable of repopulating an
ALI culture. P0plastic cells were seeded in replicate filters at multiple dilutions
and grown to confluence while being fed apically and basolaterally. If confluence
was reached, cultures were exposed to air apically and fed basolaterally until
differentiation was observed or culture failed to differentiate. Those cultures
which reached confluence and/or underwent differentiation were scored as
successes and ELDA [50] was used to determine frequency of cell that leads to
success. Lower and Upper bounds represent 95% confidence intervals.
Patient Sample
Lower Estimated frequency
Upper Reject single-hit hypothesis (p<0.05)
78297 511 164 53 No (0.483) 79168 203 96 45 No (0.906) 29945 1030 519 261 No (0.267) 32 398 182 83 No (0.823) 55 3441 1297 489 No (0.297)
Table 3: Frequency of a cell capable of repopulating an ALI culture is
significantly different between patient strains. Pairwise comparison of
estimated repopulating cell frequency between patient samples was performed
populations (Appendix A), which might be due to technical differences in staining
between HTS and single tube format.
3.7 Differences in isolated cell populations can be determined using ALI
culture.
While comprehensive functional evaluation of all reproducible subpopulations
would be beyond the scope of this work, we chose a few candidate subpopulations for
functional studies. Bulk population cells were FACS sorted based on expression of CD54
43
(fig. 9a, 5.4%) or podoplanin (fig. 10a, 85%), and seeded in replicate at ALI (10 000 cells
per well).
Monitoring progress of CD54-positive and -negative cultures revealed a marked
difference in epithelial morphology arising immediately after confluence and persisting
until 3 weeks post confluence when cultures had fully differentiated (fig. 9b-d). CD54-
positive cultures resembled control cultures, with many gland-like structures. In
contrast, CD54-negative cells formed very smooth cultures, with few gland-like
structures. During the three week period following establishment of ALI, CD54-positive
cultures appeared to slough a significant amount of debris or cells. This process was not
observed in CD54-negative cultures. These differences did not impact the extent of
ciliogenesis by enface staining (fig. 9e-g). Although gene expression analysis for
induction of FOXJ1 showed high variability between replicate cultures of each
subpopulation, no significant difference between CD54-positive and CD54-negative
cultures and unsorted controls were detected (fig. 9h). Evaluation of MUC16 induction
also showed no reproducible difference between populations (fig. 9i).
Cells sorted for the expression of podoplanin (fig. 10a) and grown at ALI showed
no differences during culture maturation at ALI. Enface staining revealed similar levels
of ciliogenesis in podoplanin-positive, podoplanin-negative and unsorted cultures (fig.
10b-d). Similar to our analysis with CD54 subpopulations, there were large differences
in FOXJ1 expression between replicate cultures, but overall, no reproducible difference
44
Figure 9: P0p basal cells sorted for expression of CD54 resulted in morphologically distinct cultures. (A) FACS plot showing gating used to sort P0plastic cells based on podoplanin expression. Phase contrast microscopy showing morphology of: (B) CD54-positive, (C) CD54-negative, and (D) unsorted ALI cultures at 3
weeks post-confluence (scale bars: 50 m). Enface staining of: (E) CD54-positive, (F) CD54-negative, and (G) unsorted ALI cultures at 21 days post-confluence (scale bars: 1 mm). QPCR analysis showing (H) FoxJ1 and (I) Muc16 gene expression between CD54-positive and CD54-negative cultures.
45
Figure 10: P0p basal cells sorted for expression of podoplanin gave rise to
indistinguishable cultures. (A) FACS plot showing gating used to sort P0plastic cells
based on podoplanin expression. Enface staining of: (B) podoplanin-positive, (C)
podoplanin-negative, and (D) unsorted ALI cultures at 21 days post-confluence (scale
between podoplanin-positive and podoplanin-negative cultures.
46
between podoplanin-positive and –negative subpopulations (fig. 10e). Similarly, no
significant differences were observed in MUC16 induction between the podoplanin-
positive and podoplanin-negative subpopulations (fig. 10f).
Rare populations of cells are of great interest, as they may represent stem or
progenitor cells. However, these populations pose a challenge for study due to their
limiting number. To bypass this issue, we adopted the strategy of depleting the bulk
population of such rare cell populations. If the depleted cells represent an essential
population it is expected that cultures would perform poorly even at high seeding
densities. Depletion of populations expressing: CD116, CD117, CD127 or CD337 showed
no noticeable decline in performance (data not shown). These data suggest that such
subpopulations do not represent essential populations for success at ALI, or that the
culture conditions employed here do not require such populations.
4. Discussion
Under steady state conditions the airway epithelium is relatively quiescent.
However, it is regularly exposed to harmful environmental agents and pathogens, and is
the site of a number of prevalent diseases such as lung cancer, and cystic fibrosis.
Hence, a mechanism is required for normal, long term maintenance that can also
respond rapidly to injury and disease conditions. Presumably, the regenerative process
responsible for maintaining epithelial integrity involves a local stem cell population. This
population could follow a classical stem cell hierarchy as seen in several highly
proliferative tissues including the intestinal epithelium and blood, where a rare stem cell
47
gives rise to progeny cells [22]. Alternatively, the lung epithelium could follow a non-
classical stem cell hierarchy as proposed in the liver [27,60]; in this model a large
number of equipotent, facultative stem cells would act in steady state and following
injury to repair or maintain the epithelium. Arguments can be made for either model in
the lung epithelium; a classical stem cell model might provide for long term
regenerative potential to maintain a tissue over the lifespan of the organism. A non-
classical model could be sufficient for long term maintenance of a tissue with a low
turnover rate, while providing a large number of progenitor cells capable of responding
quickly to acute injury. No consensus has been reached as current evidence supports
either model existing in the lung.
Basal cells are known to possess the regenerative capacity of the tracheal
epithelium [16,40,41], but also serve to anchor the entire epithelium to the basement
membrane through their unique formation of hemidesmosomes [9,10]. In light of the
multiple known functions for basal cells, in either a classical or non-classical model, we
might expect to find heterogeneity amongst basal cells at varying stages of maturity,
lineage commitment, or functional commitment. In any case, basal cell subpopulations
could be molecularly and functionally distinct.
In this work I explored cellular heterogeneity of the human trachea progenitor
cell compartment. Some evidence from murine studies has supported the idea of
functional difference in subsets of basal cells based on expression of cytokeratins, K5
and K14 [16,44-46]. Other work suggests that the anatomical location of basal cells is a
48
predictor of progenitor capacity [40,45,46]. In humans, the existence of cellular
heterogeneity has been supported by several lines of evidence [28,40,42]. In vitro
studies suggest there are differences in the differentiation potential of human p63+
progenitor cells that reside at different locations within airways. P63+ cells isolated
from the trachea were shown to differentiate into ciliated and goblet cells at ALI, while
those from the distal airway formed monolayers with very little mucociliary
differentiation [42]. Differences in clonogenic lineage potential of tracheal basal cells
have been shown in tracheal xenografts [40]. Furthermore, 1.6% of cells isolated from
murine trachea have some clonogenic potential, and 0.05- 0.1% of human and murine
tracheal cells can give rise to large colonies in vitro [42,44]. Since basal cells represent
approximately one third of all epithelial cells in the proximal airway [2], these numbers
suggest that rare basal cells possess clonogenic potential, while even fewer possess a
large capacity for proliferation. These results support the existence of subpopulations
within the basal lineage with distinct functional properties. Based on this evidence, we
set out to discover the frequency of such a population of cells in the human trachea, and
to find molecular markers which could be used to identify and separate functionally
distinct subpopulations.
4.1 Establishment of culture systems and assays
To begin to investigate heterogeneity in the basal cell compartment we first had
to obtain functional progenitor populations and establish techniques to evaluate them.
Using previously described methods, human tracheal epithelial cells (HTECs) were
dissociated from tracheal tissue and cultured to selectively expand basal cells while not
49
supporting growth or survival of fibroblasts or columnar cells [47]. These culture-
purified basal cells, grown on plastic without passaging (P0plastic), possess the capacity
to differentiate into other functional tracheal lineages [38]. Using these methods, we
derived basal cells from a number of donor tracheas and confirmed their purity and
progenitor function.
The basal cell purity of P0plastic cells was initially assessed by FACS analysis
showing over 99% expressed CD44, an in vivo basal cell specific marker (fig. 1). Purity
was also evaluated by IF staining; nearly 90% of P0plastic cells stained positively for p63,
another basal cell-specific marker, and no cells expressed columnar cell markers MUC16
or polarized apical βTUBIV, markers for mucinous and ciliated cells respectively (fig. 1).
Given these data and our observation that with longer exposure times we could detect
low levels of p63 in “p63 negative” nuclei, we conclude that virtually all P0plastic cells
are basal cells. The reduced levels of p63 immunoreactivity observed in some cells
could be a staining artifact, or could represent cells undergoing a cellular process which
corresponds with decreased expression of p63; these changes may occur during cell
division or migration. Such processes may also account for the observed βTUBIV
cytoplasmic foci, which could represent cells in a particular stage of the cell cycle, or
cells undergoing specific morphological changes [58].
To verify progenitor activity of the P0plastic basal cells and to establish methods
to functionally interrogate basal cell subpopulations, we characterized the behaviour of
basal cells in ALI cultures. Basal cells were initially seeded subconfluently at ALI, grown
50
to confluence while being fed both basolaterally and apically, and then fed only
basolaterally for 3 weeks. Assessment of p63 immunostaining and qPCR showed that
basal cells grown at ALI retain their basal identity until cultures reach confluence. By
three days post-confluence there were many p63 negative nuclei, and secretory cells
were detected by MUC16 protein and gene expression suggesting that progenitors were
transitioning into other lineages. Emergence of ciliated cells was apparent by 7 days
post confluence as assessed by IF staining. At 14 days post confluence, the presence of
beating cilia was visible by light microscopy, which corresponds with large increases in
FOXJ1 expression. These data demonstrate that P0plastic basal cells are a pure
population of progenitors that have retained mucociliary potential.
Having purified a population of functional progenitors we sought to develop a
method to functionally evaluate progenitor cells that is: consistent between samples,
easily monitored and compared, and gives good differentiation. Using the ALI culture
system to differentiate progenitor cells, we looked at ciliogenesis as a progenitor
output. Ciliogenesis is an ideal lineage readout as: it is robust; it can be observed
microscopically in living culture by light microscopy; and it can be measured
quantitatively at the level of gene and protein expression.
Histological sections have historically been used to examine ciliogenesis. They
reveal epithelial morphology, and IF staining provides a qualitative view of cell types
present and may allow some cell types to be quantified [41,61-63]. However, we found
histological sections to have limited value for quantitative analysis when limited
51
differentiation is achieved. This is largely due to uneven distribution of ciliated cells. To
resolve this issue, we investigated two alternative methods to quantitatively assess
differentiation: qPCR and enface staining. We show that FOXJ1 gene induction tracks
well with emergence of ciliated cells as determined by IF staining for βTUBIV, (fig. 5 and
6). Furthermore, using IF staining, ciliated cells are easily delineated from non-ciliated
cells because βTUBIV is contained within membrane bound cilia projections. The enface
staining and tiling method described here allows the extent of differentiation to be
measured across the entire ALI culture, removing any variability due to uneven
distribution of ciliated cells.
We observed macroscopic secretions on the apical surface of the ALI culture
preceding appearance of functional cilia, suggesting the early emergence of secretory
cells. Surprisingly, antibody and gene expression analysis suggest these secretions are
due to MUC16 secretory cells, not due Muc5ac+ goblet cells (fig. 2, 3). Although we
found that P0plastic cells can differentiate into goblet cells in xenograft cultures, we
have not observed the emergence of goblet cells in ALI cultures, suggesting our ALI
conditions lack required signaling for this fate. Instead, our ALI conditions drive
physiologic levels of MUC16 secretory cell differentiation, another abundant secretory
lineage in the upper airway [56] (fig. 2, 3). We observed that Muc16 gene and protein
expression show similar kinetics at ALI, and are induced in all samples. However, trends
in induction and expression are not as robust or consistent between patient strains as
those observed with ciliated cell markers (fig. 5 and 6).
52
Previous work has indicated that serial passage of human basal cells on plastic
leads to significant decline in progenitor potential [38]. We asked whether serial
passage of cells on ALI culture might provide better preservation of progenitor activity.
This could be important for maintaining progenitor cells for long term experiments or
expanding limited primary cells to acquire large numbers required for high-throughput
assays. We compared the progenitor activity of plastic-passaged cells to ALI-passaged
cells by comparing their ability to differentiate at ALI (fig. 7). Consistent with previous
reports, we found that the progenitor potential of plastic passaged cells fell drastically
over 2 passages. A similar decline was observed in the activity of ALI-passaged cells.
The extent of ciliogenesis was substantially lower in P2pA and PAPA cultures than in
P0pA from all patient strains examined. However, there was little impact on MUC16
lineage potential following P2pA and PAPA cultures, which showed similar MUC16
expression to P0pA cultures. These data show that ALI passaging is not better than
submerged LHC-9 culture at preserving mucociliary progenitor activity. The differential
impact of ciliary and mucinous potential could suggest these lineages are derived from
different progenitor cell populations, which are differentially maintained during culture.
Alternatively, the ciliogenic potential may be maintained but the signaling environment
of culture may preferentially drive a mucinous fate. Further development of both ALI
and submerged culture conditions could permit better maintenance or detection of
progenitor potential.
4.2 Evidence for functional heterogeneity
53
To investigate functional heterogeneity within the basal cell compartment, we
employed a limiting dilution assay (LDA). In theory, if rare stem cells exist in our
P0plastic basal cell population, they would be diluted out as cultures were seeded at
lower densities. The distribution of these cells would follow a Poisson model, meaning
at some critical seeding density we would have a high probability of seeding one or zero
stem cells. At this density we would see some cultures succeed, while those cultures
not containing a stem cell would fail. Alternatively, we could see a similar result if the
cells were equipotent, but produced a paracrine factor required for growth, since a
minimum number of cells might be required for sufficient concentrations of that factor
to accumulate.
Using this technique, we were able to obtain sufficient data from 5 patient
strains to confirm that a single factor may be responsible for success of the cultures, as
defined by the cultures reaching confluence and maintaining a barrier (table 2). We
found that basal cells isolated from different patients performed significantly differently
in ALI cultures (table 3). The estimated frequency of a basal cell capable of
repopulating an ALI culture ranged from 1 in 96 to 1 in 1297 (0.08%-1%) (fig. 3). These
values are similar to those previously reported in human and mouse for the frequency
of cell capable of giving rise to large colonies (0.05- 0.1%) [42,44]. The slightly lower
frequency of clonogenic cell observed in these studies may be due to failure to sort
basal cells from columnar cells, differences between species, damage from the sorting
process [44], or culture conditions [42] used in these studies. Alternatively, this could
54
reflect natural variation between patients based on biological differences such as age, or
gender.
The data presented here supports the existence of subsets of ALI repopulating
cells rather than the existence of a dilutable paracrine factor essential for growth. First,
the complete failure of one replicate culture at the same seeding density that results in
successful repopulation in another replicate culture is difficult to reconcile with dilution
of a paracrine factor unless an extreme error in seeding density occurred. Second, the
sometimes large differences in induction of differentiation genes between replicate
cultures (fig. 6, 7, 10 and 11), seeded at high initial densities, are consistent with the
notion that there are rare cells that give rise to differentiated progeny, and that these
cells are variably diluted during seeding. This observation is also difficult to reconcile
with the model that equipotent cells simply produce a limiting paracrine factor,
especially when cells are seeded at high densities. Our data support a model where rare
cells, representing less than 1% of P0plastic basal cells, have the ability to undergo
massive expansion and repopulate an ALI filter. However, the mucociliary
differentiation of these cultures may be dependent on additional factors or cell types.
None of the LDA experiments yielded sufficient results to test whether a single factor
was responsible for successful mucociliary differentiation. However, the difference in
our ability to test the single hit hypothesis for repopulation but not differentiation
suggests a different factor, or number of factors, is required for each process.
4.3 Evidence for molecular heterogeneity
55
Large scale comprehensive surface marker profiling has not been reported for
lung basal cells in any species. To comprehensively investigate molecular heterogeneity
within the basal cell population we employed a high-throughput screen consisting of
339 antibodies. Here we consider a subpopulation to be any subset of P0plastic cells
that have differential expression of the examined surface marker. We initially identified
115 cell surface markers that denoted subpopulations of basal cells in 2 independent
patient strains. Of these markers, 74 were confirmed in a non-high throughput test on
cells from different patient strains, indicating reproducible surface marker
heterogeneity among human basal cells. The observed heterogeneity could be due to
the presence of hierarchical relationships between basal cells with different proliferative
and lineage potential, or result from equipotent basal cells that are in different states.
For example, these states could reflect different stages of maturation or differentiation
or stress response to in vitro culture. The cell surface marker signature of any single
subpopulation of basal cells could be comprised of single or multiple markers within the
group of 74. At present, it is unclear how many distinct populations are represented by
the 74 markers.
While the extent to which this surface marker heterogeneity reflects in vivo
populations has not been addressed in this work, our results provide tools for future
directed in vivo analyses. A previous study by Atsuta et al. looking at integrin expression
on human bronchial epithelial cells supports some of our findings. They also found
evidence for expression of 12 surface markers identified in our screen on P0plastic
bronchial epithelial cells [64]. However, they did not examine potential heterogeneity in
56
expression of these markers. Our data suggest that 9 of these 12 markers are actually
heterogeneously expressed among basal cells.
Little work has been done to identify normal lung stem cell markers. However,
one recent report by Kajstura et al. proposed the existence of rare human CD117+,
KLF4+, NANOG+, OCT3/4+ and SOX2+ stem cells in the lung. While most of these cells
are p63 negative and lack epithelial markers, some evidence was provided that
indicated a fraction of these cells could acquire a p63 positive fate [28]. In our work, we
found inconsistent expression of CD117 across different patient samples. Furthermore,
we were unable to expand viable primary unpassaged HTECs using the culture
conditions that were reported to promote expansion of the CD117+ stem cells [28].
While it is possible that our failure to consistently detect such a cell could be due to its
relative enrichment in the distal rather than primary airways, the existence of CD117+
stem cells has yet to be confirmed by other labs.
It is possible that distinct progenitor cells throughout the airways share some
surface antigens. Studies of lung progenitor cells in distal regions of the airway have led
to the identification of several surface markers. Kim et al. described a bronchioalveolar
stem cell as Sca1+ and CD34+. Sca-1 is a murine specific antigen and we did not detect
CD34 immunoreactivity in our basal cells.
In some cases, cancer stem cells share phenotypic properties with normal stem
cells [65]. Several surface markers have been identified for lung cancer stem cells, also
known as tumour initiating cells (TIC). Tirino et al. proposed that a CD133+ cell fraction
57
of non-small cell lung cancers (NSCLC) includes TICs [66]. They found a CD133+ fraction
in 72% of NSCLC, and showed that culturing these cells as non-adherent spheres
increased the proportion of CD133+ cells, which resulted in a nearly 4-fold increase in
tumour initiating capacity of these cultures in NOD/SCID mice [66]. However, CD133+
cells were not detected in our screen, which suggests that such cells may only arise in
disease states, or may reflect non-basal cell progenitors. More recent work by Zhang et
al. did not support CD133 as a marker for NSCLC TICs, but provided evidence for the
CD166+ fraction as marking the majority of TICs [67]. In our screen, all basal cells
expressed CD166, which could be consistent with some TICs arising from the p63+ basal
cell compartment.
Putative stem cell populations have been characterized in a number of other
epithelia. Notably, mammary glands are composed of a stratified epithelium where
p63+ basal cells are thought to contain the progenitor activity. CD49f, CD24 and CD29
have been identified as potential markers of mammary epithelial stem cells [68,69], and
CD49f has also been used to enrich progenitor activity in p63+ population of
keratinocytes [70]. All three of these markers were identified in our screen as being
heterogeneously expressed in lung basal cells and are promising candidates for further
study.
To investigate the existence of potential functional differences between marker-
positive and marker-negative subpopulations of P0plastic basal cells we looked at CD54
and podoplanin, markers of relatively abundant subpopulations which showed
58
promising results in preliminary experiments. Cells were sorted based on expression of
CD54 or podoplanin, and fractions were assessed at ALI for: proliferative potential,
extent of differentiation, and emergence of mature cell types. For both CD54 and
podoplanin, the positive and negative fractions had indistinguishable lineage potential
as assessed by qPCR and enface staining of 21 day post confluent cultures. However,
monitoring of CD54 subpopulations revealed marked, reproducible differences in
culture morphology during epithelial regeneration (fig. 10). CD54 expression on airway
epithelial cells has been linked to local molecular events in inflammation in response to
allergen exposure, playing a role in eosinophil recruitment and infiltration [71]. Thus,
these morphological differences between CD54+ and CD54- generated ALI epithelia
could reflect functional differences in the ability to promote specific inflammatory
processes (fig. 10) [71,72]. The failure to detect a difference between podoplanin
subpopulations could reflect stochastic changes in ex vivo expression without functional
significance, or could reflect limitations of our assays in detecting functional differences
between subpopulations of basal cells. Notably, we only examined two lineage markers
and our assays do not extensively evaluate the self-renewal potential of basal cells.
The markers we have identified can now be used to help distinguish a classical
from non-classical stem cell model for epithelial regeneration in the trachea. Candidate
subpopulations can be sorted and re-cultured, and the proliferative ability of these cells,
as well as their ability to regenerate marker positive and marker negative states can be
assessed. Sorted cells can also be interrogated in ALI models employing serial
passaging, as well as tracheal xenograft models, which may help uncover functional
59
differences between basal cell subpopulations. Additionally, antibody staining of
tracheal tissue could be used to distinguish relevant in vivo subpopulations from those
generated artifactually by ex vivo culture. Finally, multi-parametric FACS analysis of
P0plastic basal cells will be useful to determine how many subpopulations are
represented by the markers identified here.
Here I present data demonstrating human tracheal P0plastic cells are a purified
population of basal cells that retain progenitor activity (Fig. 1, 2 and 3). I provided
detailed analysis of the differentiation process of P0plastic basal cells on ALI, showing
robust induction of markers for both ciliogenic and secretory differentiation programs,
as well as optimized methods for assessing progenitor output and differentiation (Fig. 4,
5 and 6). I demonstrated that ALI culture conditions are able to maintain a progenitor
population at an extent similar to LHC-9 conditions (Fig. 7). Performing a limiting
dilution assay on basal cells with in vitro and in vivo validated progenitor activity, I
determined the frequency of clonogenic cells capable of repopulating ALI ranges from
0.08%-1%, and varies significantly between patients. Importantly, I contributed a list of
74 validated cell surface markers that are heterogeneously expressed amongst P0plastic
basal cells. Subpopulations varied in frequency from <1% to ~50%. Preliminary
functional analyses suggest there may be differences in the differentiation potential of
some of the subpopulations. This work supports the notion that the basal cell
compartment may be functionally heterogeneous, and provides a new arsenal of
molecular tools for the directed investigation of heterogeneity among human basal
cells.
60
7. References
1. Message SD, Johnston SL (2004) Host defense function of the airway epithelium in health and disease: clinical background. J Leukoc Biol 75: 5-17.
2. Knight DA, Holgate ST (2003) The airway epithelium: structural and functional properties in health and disease. Respirology 8: 432-446.
3. Wanner A, Salathe M, O'Riordan TG (1996) Mucociliary clearance in the airways. Am J Respir Crit Care Med 154: 1868-1902.
4. Puchelle E, Zahm JM, Tournier JM, Coraux C (2006) Airway epithelial repair, regeneration, and remodeling after injury in chronic obstructive pulmonary disease. Proc Am Thorac Soc 3: 726-733.
5. Crystal RG, Randell SH, Engelhardt JF, Voynow J, Sunday ME (2008) Airway epithelial cells: current concepts and challenges. Proc Am Thorac Soc 5: 772-777.
6. Spina D (1998) Epithelium smooth muscle regulation and interactions. Am J Respir Crit Care Med 158: S141-145.
7. Evans CM, Williams OW, Tuvim MJ, Nigam R, Mixides GP, et al. (2004) Mucin is produced by clara cells in the proximal airways of antigen-challenged mice. Am J Respir Cell Mol Biol 31: 382-394.
8. Boers JE, Ambergen AW, Thunnissen FB (1999) Number and proliferation of clara cells in normal human airway epithelium. Am J Respir Crit Care Med 159: 1585-1591.
10. Evans MJ, Van Winkle LS, Fanucchi MV, Plopper CG (2001) Cellular and molecular characteristics of basal cells in airway epithelium. Exp Lung Res 27: 401-415.
11. Rhodin JA (1966) The ciliated cell. Ultrastructure and function of the human tracheal mucosa. Am Rev Respir Dis 93: Suppl:1-15.
12. Lackie PM, Baker JE, Gunthert U, Holgate ST (1997) Expression of CD44 isoforms is increased in the airway epithelium of asthmatic subjects. Am J Respir Cell Mol Biol 16: 14-22.
13. Romano RA, Ortt K, Birkaya B, Smalley K, Sinha S (2009) An active role of the DeltaN isoform of p63 in regulating basal keratin genes K5 and K14 and directing epidermal cell fate. PLoS One 4: e5623.
14. Guo X, Keyes WM, Papazoglu C, Zuber J, Li W, et al. (2009) TAp63 induces senescence and suppresses tumorigenesis in vivo. Nat Cell Biol 11: 1451-1457.
15. Cole BB, Smith RW, Jenkins KM, Graham BB, Reynolds PR, et al. (2010) Tracheal Basal cells: a facultative progenitor cell pool. Am J Pathol 177: 362-376.
16. Rock JR, Onaitis MW, Rawlins EL, Lu Y, Clark CP, et al. (2009) Basal cells as stem cells of the mouse trachea and human airway epithelium. Proc Natl Acad Sci U S A 106: 12771-12775.
17. Rawlins EL, Hogan BL (2006) Epithelial stem cells of the lung: privileged few or opportunities for many? Development 133: 2455-2465.
18. Rawlins EL, Okubo T, Que J, Xue Y, Clark C, et al. (2008) Epithelial stem/progenitor cells in lung postnatal growth, maintenance, and repair. Cold Spring Harb Symp Quant Biol 73: 291-295.
19. Boers JE, den Brok JL, Koudstaal J, Arends JW, Thunnissen FB (1996) Number and proliferation of neuroendocrine cells in normal human airway epithelium. Am J Respir Crit Care Med 154: 758-763.
61
20. Jeffery PK (1983) Morphologic features of airway surface epithelial cells and glands. Am Rev Respir Dis 128: S14-20.
21. Thompson AB, Robbins RA, Romberger DJ, Sisson JH, Spurzem JR, et al. (1995) Immunological functions of the pulmonary epithelium. Eur Respir J 8: 127-149.
22. Morrison SJ, Uchida N, Weissman IL (1995) The biology of hematopoietic stem cells. Annu Rev Cell Dev Biol 11: 35-71.
23. Pinto D, Clevers H (2005) Wnt control of stem cells and differentiation in the intestinal epithelium. Exp Cell Res 306: 357-363.
24. Blenkinsopp WK (1967) Proliferation of respiratory tract epithelium in the rat. Exp Cell Res 46: 144-154.
25. Evans MJ, Cabral LJ, Stephens RJ, Freeman G (1975) Transformation of alveolar type 2 cells to type 1 cells following exposure to NO2. Exp Mol Pathol 22: 142-150.
26. Rawlins EL, Okubo T, Xue Y, Brass DM, Auten RL, et al. (2009) The role of Scgb1a1+ Clara cells in the long-term maintenance and repair of lung airway, but not alveolar, epithelium. Cell Stem Cell 4: 525-534.
27. Forbes S, Vig P, Poulsom R, Thomas H, Alison M (2002) Hepatic stem cells. J Pathol 197: 510-518.
28. Kajstura J, Rota M, Hall SR, Hosoda T, D'Amario D, et al. (2011) Evidence for human lung stem cells. N Engl J Med 364: 1795-1806.
29. Smith LJ (1985) Hyperoxic lung injury: biochemical, cellular, and morphologic characterization in the mouse. J Lab Clin Med 106: 269-278.
30. Evans MJ, Cabral-Anderson LJ, Dekker NP, Freeman G (1981) The effects of dietary antioxidants on NO2- induced injury to type 1 alveolar cells. Chest 80: 5-8.
31. Van Winkle LS, Johnson ZA, Nishio SJ, Brown CD, Plopper CG (1999) Early events in naphthalene-induced acute Clara cell toxicity: comparison of membrane permeability and ultrastructure. Am J Respir Cell Mol Biol 21: 44-53.
32. Hong KU, Reynolds SD, Giangreco A, Hurley CM, Stripp BR (2001) Clara cell secretory protein-expressing cells of the airway neuroepithelial body microenvironment include a label-retaining subset and are critical for epithelial renewal after progenitor cell depletion. Am J Respir Cell Mol Biol 24: 671-681.
33. Reynolds SD, Giangreco A, Power JH, Stripp BR (2000) Neuroepithelial bodies of pulmonary airways serve as a reservoir of progenitor cells capable of epithelial regeneration. Am J Pathol 156: 269-278.
34. Kim CF, Jackson EL, Woolfenden AE, Lawrence S, Babar I, et al. (2005) Identification of bronchioalveolar stem cells in normal lung and lung cancer. Cell 121: 823-835.
35. Jackson EL, Willis N, Mercer K, Bronson RT, Crowley D, et al. (2001) Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 15: 3243-3248.
36. Whitcutt MJ, Adler KB, Wu R (1988) A biphasic chamber system for maintaining polarity of differentiation of cultured respiratory tract epithelial cells. In Vitro Cell Dev Biol 24: 420-428.
37. Engelhardt JF, Yankaskas JR, Wilson JM (1992) In vivo retroviral gene transfer into human bronchial epithelia of xenografts. J Clin Invest 90: 2598-2607.
38. Gray TE, Guzman K, Davis CW, Abdullah LH, Nettesheim P (1996) Mucociliary differentiation of serially passaged normal human tracheobronchial epithelial cells. Am J Respir Cell Mol Biol 14: 104-112.
39. Matsui H, Randell SH, Peretti SW, Davis CW, Boucher RC (1998) Coordinated clearance of periciliary liquid and mucus from airway surfaces. J Clin Invest 102: 1125-1131.
62
40. Engelhardt JF, Schlossberg H, Yankaskas JR, Dudus L (1995) Progenitor cells of the adult human airway involved in submucosal gland development. Development 121: 2031-2046.
41. Hajj R, Baranek T, Le Naour R, Lesimple P, Puchelle E, et al. (2007) Basal cells of the human adult airway surface epithelium retain transit-amplifying cell properties. Stem Cells 25: 139-148.
42. Kumar PA, Hu Y, Yamamoto Y, Hoe NB, Wei TS, et al. (2011) Distal airway stem cells yield alveoli in vitro and during lung regeneration following H1N1 influenza infection. Cell 147: 525-538.
43. Engelhardt JF, Yankaskas JR, Ernst SA, Yang Y, Marino CR, et al. (1992) Submucosal glands are the predominant site of CFTR expression in the human bronchus. Nat Genet 2: 240-248.
44. Schoch KG, Lori A, Burns KA, Eldred T, Olsen JC, et al. (2004) A subset of mouse tracheal epithelial basal cells generates large colonies in vitro. Am J Physiol Lung Cell Mol Physiol 286: L631-642.
45. Hegab AE, Ha VL, Gilbert JL, Zhang KX, Malkoski SP, et al. (2011) Novel stem/progenitor cell population from murine tracheal submucosal gland ducts with multipotent regenerative potential. Stem Cells 29: 1283-1293.
46. Borthwick DW, Shahbazian M, Krantz QT, Dorin JR, Randell SH (2001) Evidence for stem-cell niches in the tracheal epithelium. Am J Respir Cell Mol Biol 24: 662-670.
47. John F. Lechner ML (1981) A Serum-free Method for Culturing Normal Human Bronchial Epithelial Cells at Clonal Density. Methods in Cell Science 9: 43-48.
48. Shi SR, Chaiwun B, Young L, Cote RJ, Taylor CR (1993) Antigen retrieval technique utilizing citrate buffer or urea solution for immunohistochemical demonstration of androgen receptor in formalin-fixed paraffin sections. J Histochem Cytochem 41: 1599-1604.
49. Shi SR, Liu C, Taylor CR (2007) Standardization of immunohistochemistry for formalin-fixed, paraffin-embedded tissue sections based on the antigen-retrieval technique: from experiments to hypothesis. J Histochem Cytochem 55: 105-109.
50. Hu Y, Smyth GK (2009) ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays. J Immunol Methods 347: 70-78.
51. Yamaya M, Finkbeiner WE, Chun SY, Widdicombe JH (1992) Differentiated structure and function of cultures from human tracheal epithelium. Am J Physiol 262: L713-724.
52. Karp PH, Moninger TO, Weber SP, Nesselhauf TS, Launspach JL, et al. (2002) An in vitro model of differentiated human airway epithelia. Methods for establishing primary cultures. Methods Mol Biol 188: 115-137.
53. Wu R, Zhao YH, Chang MM (1997) Growth and differentiation of conducting airway epithelial cells in culture. Eur Respir J 10: 2398-2403.
54. Renthal R, Schneider BG, Miller MM, Luduena RF (1993) Beta IV is the major beta-tubulin isotype in bovine cilia. Cell Motil Cytoskeleton 25: 19-29.
55. Jensen-Smith HC, Luduena RF, Hallworth R (2003) Requirement for the betaI and betaIV tubulin isotypes in mammalian cilia. Cell Motil Cytoskeleton 55: 213-220.
56. Davies JR, Kirkham S, Svitacheva N, Thornton DJ, Carlstedt I (2007) MUC16 is produced in tracheal surface epithelium and submucosal glands and is present in secretions from normal human airway and cultured bronchial epithelial cells. Int J Biochem Cell Biol 39: 1943-1954.
57. Hackett BP, Brody SL, Liang M, Zeitz ID, Bruns LA, et al. (1995) Primary structure of hepatocyte nuclear factor/forkhead homologue 4 and characterization of gene expression in the developing respiratory and reproductive epithelium. Proc Natl Acad Sci U S A 92: 4249-4253.
63
58. Dumontet C, Duran GE, Steger KA, Murphy GL, Sussman HH, et al. (1996) Differential expression of tubulin isotypes during the cell cycle. Cell Motil Cytoskeleton 35: 49-58.
59. Yu X, Ng CP, Habacher H, Roy S (2008) Foxj1 transcription factors are master regulators of the motile ciliogenic program. Nat Genet 40: 1445-1453.
60. Forbes SJ, Poulsom R, Wright NA (2002) Hepatic and renal differentiation from blood-borne stem cells. Gene Ther 9: 625-630.
61. You Y, Richer EJ, Huang T, Brody SL (2002) Growth and differentiation of mouse tracheal epithelial cells: selection of a proliferative population. Am J Physiol Lung Cell Mol Physiol 283: L1315-1321.
62. LeSimple P, van Seuningen I, Buisine MP, Copin MC, Hinz M, et al. (2007) Trefoil factor family 3 peptide promotes human airway epithelial ciliated cell differentiation. Am J Respir Cell Mol Biol 36: 296-303.
63. Prytherch Z, Job C, Marshall H, Oreffo V, Foster M, et al. (2011) Tissue-Specific stem cell differentiation in an in vitro airway model. Macromol Biosci 11: 1467-1477.
64. Atsuta J, Sterbinsky SA, Plitt J, Schwiebert LM, Bochner BS, et al. (1997) Phenotyping and cytokine regulation of the BEAS-2B human bronchial epithelial cell: demonstration of inducible expression of the adhesion molecules VCAM-1 and ICAM-1. Am J Respir Cell Mol Biol 17: 571-582.
65. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3: 730-737.
66. Tirino V, Camerlingo R, Franco R, Malanga D, La Rocca A, et al. (2009) The role of CD133 in the identification and characterisation of tumour-initiating cells in non-small-cell lung cancer. European Journal of Cardio-Thoracic Surgery 36: 446-453.
67. Zhang WC, Shyh-Chang N, Yang H, Rai A, Umashankar S, et al. (2012) Glycine decarboxylase activity drives non-small cell lung cancer tumor-initiating cells and tumorigenesis. Cell 148: 259-272.
68. Shackleton M, Vaillant F, Simpson KJ, Stingl J, Smyth GK, et al. (2006) Generation of a functional mammary gland from a single stem cell. Nature 439: 84-88.
69. Stingl J, Eirew P, Ricketson I, Shackleton M, Vaillant F, et al. (2006) Purification and unique properties of mammary epithelial stem cells. Nature 439: 993-997.
70. Li A, Simmons PJ, Kaur P (1998) Identification and isolation of candidate human keratinocyte stem cells based on cell surface phenotype. Proc Natl Acad Sci U S A 95: 3902-3907.
71. Ciprandi G, Pronzato C, Ricca V, Passalacqua G, Bagnasco M, et al. (1994) Allergen-specific challenge induces intercellular adhesion molecule 1 (ICAM-1 or CD54) on nasal epithelial cells in allergic subjects. Relationships with early and late inflammatory phenomena. Am J Respir Crit Care Med 150: 1653-1659.
72. Striz I, Mio T, Adachi Y, Heires P, Robbins RA, et al. (1999) IL-4 induces ICAM-1 expression in human bronchial epithelial cells and potentiates TNF-alpha. Am J Physiol 277: L58-64.
64
Appendices
Appendix A
Antibody Str. 38 Str. 23
Single
tube
CD50 0.0367 2.55E-03 0
CD123 0.18 0.139 0.054
CD180 0.182 0.0448 0
CD195 0.283 1.36 0
CD249 0.332 0.829 0
CD33 0.399 0.64 0.017
CD22 0.441 0.548 0.055
CD122 0.502 0.231 0.46
CD62L 0.511 1.11 0
CD312 0.519 1.25 0
CD181 0.533 0.555 0.48
CDW93 0.567 0.241 0
CD62P 0.609 0.4 0
CD275 0.808 0.173 0.49
CD116 0.948 0.618 0.89
CD266 0.978 0.991 0.071
CD213a2 1.05 0.589 0
CD159c 1.07 1.5 0
CD182 1.07 0.312 0.12
VBTA8 1.07 1.49 0.88
CD32 1.09 0.364 0
CA9 1.11 1.64 0.036
CD16 1.13 0.461 0.085
CD82 1.22 1.35 0.045
CD6 1.26 1.31 0
CD132 1.42 0.383 0.87
CD7 1.65 1.15 0.11
HLA DM 1.69 0.953 0
CD193 1.83 1.95 0.37
CD247 0.207 2.22 0
CD138 0.697 5.36 0
CD85g 1.41 2.35 0
CD56 1.71 2.32 0
CD164 2.01 1.34 0
CD158e2 2.07 1.28 0.65
CD99 2.09 3.36 0
CD227 2.09 2.43 1.48
CD38 2.61 5.97 0
CD117 2.66 0.338 6.35
CD2 2.68 3.38 0
CD87 3.09 0.757 0
65
CD97 3.09 3.37 0
SSEA-4 3.11 0.632 0.58
CD175s 3.17 8.88 0
FOXP3 3.9 0.0561 0
VDTA2 4.16 2.38 1.31
CD267 4.42 1.17 1.01
CD125 4.52 1 0
CD157 4.89 13 0.015
CD268 4.91 1.31 1.96
CD13 5.37 9.3 31.6
CD90 5.54 1.93 4.42
CD49D 5.55 8.4 1.85
CD51 5.76 1.34 0
GMA DTA 5.86 0.805 0.76
CD36L1 6.18 5.53 1.22
CD91 7.46 6.05 0
CD127 8.14 0.707 0.61
CD61 3.17 17.7 0
CD170 4.57 18.8 0
CD295 6 13.9 0.43
CD314 14.8 2.55 1.86
CD337 17.8 4.26 0.72
CDH3 18.1 49.3 0
CD158A 19.2 0.821 0.015
CD253 19.3 2.37 0.057
CDW218a 20 24.9 0.48
SSEA-3 20.6 3.16 0.017
CD245 21 46.2 22.8
CD252 23.2 2.16 0.31
CD10 27.7 9.02 2.79
CD102 29.3 21.9 3.18
CD263 30.8 2.81 0.63
CD49A 38.7 37.4 0
6D12 39.9 20.7 30.5
CD66C 45 17.5 17.5
CD271 21.9 87.7 2.28
CD205 41.2 61.3 0
CD201 32.8 55.7 17.8
CD148 46.3 56.4 1.51
CD151 58 1.82 7.28
CD130 61.5 14.2 1.41
CD54 67 43.4 31.2
CD108 72.4 33.3 41.9
CD105 76.9 68 0
CD71 82.9 28.7 0.91
CD264 85.8 36.9 42.4
66
CD65s 86.9 76.1 15.9
CD63 87 77.8 6.84
Podoplanin 89.7 71.3 58.6
CD261 82 92.1 0.36
CD262 82.1 90.3 21.7
CD223 57.4 94.1 44.6
CD44 66.4 94.2 N/A
EGF-R 91.3 96.6 5.42
CD66 95.7 76.3 39
CD9 96.9 96.7 73.1
CD340 97 99.1 75.6
CD221 98.2 99 1.45
CD142 98.2 96.2 95.1
CD47 98.5 99.9 3.49
CD326 98.5 98.7 65.7
CD81 98.5 93.9 75.9
CD24 98.7 74.9 62.3
CD172a 98.8 50.8 47.4
CD29 98.8 99.8 100
CD119 99.1 97.7 1.34
CD58 99.1 99.5 99.42
CD276 99.4 99.7 100
CD49F 99.5 99.9 99.1
CD147 99.7 99.5 100
TNFR RLP 85.1 99.3 1.17
67
Appendix B
Sequence of primers
Gene Forward Primer Reverse Primer MUC16
Np63 TAp63 FOXJ1 TBP
TGC GGT GTC CTG GTG ACC ACC CGC GGA AAC AAT GCC CAG ACT CAA TGT ATC CGC ATG CAG GAC T CAC CTG AGC CGA GCC GGG ACT TAG CGG TGT GCA CAG GAG CCA AGA GT
CAC CGG CAA GTT CCA GTC ATT GC TGC GCG TGG TCT GTG TTA CTG TGT TAT AGG GAC TGG TGG AC CTC CCG TTA CAC GGC CTC CCG ATT TTC TTG CTG CCA GTC TGG