Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

CHARACTERIZATION AND IDENTIFICATION OF ENDOGENOUS
FACTOR(S) THAT ENHANCE INSULIN TRANSPORT ACROSS PRIMARY
RAT ALVEOLAR EPITHELIAL CELL MONOLAYERS

by

Rana Bahhady

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)

December 2007

Copyright 2007 Rana Bahhady
ii

Dedication

This work is dedicated to my father, whose persistence and wisdom, guided me, and
to my mother whose love and support, reassured me.
iii
Acknowledgements
My sincere gratitude and appreciation extends to my advisor, Dr. Shen. I was
first accepted into Dr. Shen’s lab as a Master’s student, followed by a “couple” more
years as a PhD student. Throughout the years, his guidance, kindness, and
unfathomable generosity made the knowingly tough pursuit of a doctoral degree
pleasant.
My thanks extend to Dr. Kim who always supported me and backed me up
when I was “in trouble”. I will always remember the red ink on my manuscripts as a
sign of “don’t give up and don’t get discouraged!!” Also, I would like to thank the
rest of my committee members, Dr. Okamoto, and Dr. Mircheff for their support,
constructive suggestions, and scientific discussions.
My warm feelings go to Dr. Shen’s numerous lab members, and to Daisy
whose cheerfulness and enthusiasm kept us upbeat. Most certainly are my
acknowledgments to the lovely friends I have made: Wendy, Tinten, Yun, Leena,
and Liyun, who together we shared beautiful moments talking about life,
experiences, and, undoubtedly, a lot of science!!!
Last but not least, if it were not for the years spent in the PhD program at
USC Department of Pharmaceutical Sciences, I would never have had the fortune of
meeting my beloved husband, Ron Marchelletta. Thank you, my dear, for the
wonderful days spent together!!!

iv

Table of Contents

Dedication

ii
Acknowledgements

iii
List of Tables

vii
List of Figures

viii
Abbreviations

x
Abstract

xii
Chapter One: Introduction 1
1. Lung as a route for systemic drug delivery 1
2. Alveolar epithelial cells: Type I versus type II 2
3. Mechanisms of transport across the lung epithelium 4
4. Alveolar epithelial lining fluid components 6
4.1. Surfactant protein A 10
4.2. Alpha-1 inhibitor III 13
5. Methods of enhancing drug transport across the lung epithelium 14
6. Primary rat alveolar epithelial cell culture as in vitro model 17

Chapter Two: Objectives 18
Chapter Three: Materials and Methods 19
1. Cell culture 19
1.1 Rat alveolar epithelial cell monolayers culture 19
1.2. MDCK cell culture 20
1.3. Caco-2 cell culture 20
2. Transport of insulin 21
3. Conditioned medium collection and characterization 22
3.1. Molecular sizing 22
3.2. Effect of serum, KGF, conditioned medium from RAECM-II (CMII)
and conditioned medium from RAECM-I (CMI) on insulin transport
22
3.3. Characterization of conditioned medium 23
3.4. Transport of paracellular versus transcellular markers 23
4. Bronchoalveolar lavage fluid (BALF) collection and characterization 24
4.1. Rat lung lavage fluid collection 24
4.2. Properties of the insulin transport-enhancing activity of BALF 25
4.3. Transport of paracellular versus transcellular markers 26
v
4.4. Column chromatography of BALF 27
4.5. LC-MS/MS and database search 27
5. Rat plasma collection and purification 29
6. Western blot analysis 29
6.1. Detection of SP-A in BALF 29
6.2. Detection of megalin in mouse kidney homogenate and RAECM-II 30
7. Immunofluorescence and tannic acid staining of RAECM-II 30
8. Mannosylation of albumin 32
9. Statistical analyses 32

Chapter Four: Results 33
I. Conditioned medium: Impact on insulin transport 33
1. Transport of insulin in CMII 33
2. Biochemical and physical characterization of CMII 37
3. Effects of putative factor(s) in CMII for enhancing insulin transport on
paracellular versus transcellular pathways
39
4. Effect of CMII on insulin degradation 42
5. Involvement of surfactant protein in the transport enhancement of labeled
protein
44
5.1. Primary alveolar epithelial cells cultured on T-75 as source of SP-A 44
5.2. Mannosylated albumin as a marker for mechanistic investigation 46
II. Bronchoalveolar lavage fluid: Impact on insulin transport 49
1. Transport of insulin across RAECM-II and -I, MDCK, and Caco-2 49
2. Effects of putative factor(s) in BALF on transcytosis pathways 53
7.1. Paracellular versus transcellular makers 53
7.2. Effects of temperature and endocytosis inhibitor 54
3. Apical-to-basolateral versus basolateral-to-apical transport of insulin 56
4. Characterization of insulin transport-enhancing factor(s) in BALF 57
4.1. Molecular sizing 57
4.2. Temperature and trypsin treatment 58
5. Surfactant protein A (SP-A) as a “putative” enhancing factor for both
CMII and BALF
60
5.1. Western blot of SP-A in BALF 60
5.2. Effect of native SP-A on insulin transport across RAECM-II 61
6. Purification of BALF and effect on insulin transport 63
6.1. Using ultracentrifugation and lyophilization 63
6.2. Using tangential fluid filtration 67
7. Effect of protein concentration of BALF fraction #22 on insulin
transport
71
8. Effect of BALF fraction #22 on insulin transport across RAECM-I and -
II
72

vi
9. Comparison of rat plasma to BALF 74
9.1. Purification of rat plasma 74
9.2. Effect of rat plasma on transport of insulin 76
III. Macroglobulin family: Impact on insulin transport 78
1. Inhibition of BALF-mediated enhancement of insulin using anti-
megalin antibody
78
1.1. Western blot of megalin 78
1.2. Competition with anti-megalin 79
2. Comparative studies of insulin transport in BALF versus alpha-2
macroglobulin
81
2.1. Effect of alpha-2 macroglobulin on insulin transport 81
2.2. Mechanism of alpha-2 macroglobulin and insulin association 81
3. Inhibition of BALF-mediated enhancement of insulin using activated
alpha-2 macroglobulin
84
3.1. Receptor binding of alpha-2 macroglobulin with RAECM-II 84
3.2. Competition with activated alpha-2 macroglobulin complex 86

Chapter Five: Discussion 87
1. Effect of CMII factor(s) on insulin transport 87
2. Mechanistic investigation of the effect of SP-A on insulin transport 91
3. Effect of BALF factor(s) on insulin transport 92
4. Mechanistic investigation of the effect of alpha-1-inhibitor III ( α
1
I
3
) and
alpha-2 macroglobulin ( α
2
M) on insulin transport
97

Chapter Six: Conclusion 104
Chapter Seven: Future Perspectives 108
References 112
vii
List of Tables
1. Bioavailabilities of some therapeutic peptides and proteins delivered via the
lung.

2
2. Agents used to enhance pulmonary protein absorption.

17
3. Characterization of CMII.

37
4. Effect of the 7.5x BALF on the paracellular and transcellular
pathways in RAECM-II.

53
5. Apical-to-basolateral (a-to-b) and basolateral-to-apical (b-to-a) transport of
insulin dosed in Kreb’s Ringer’s Phosphate (KRP) solution and 7.5x BALF
across RAECM-II.

57
6. Effect of different treatments of 7.5x BALF retentate on enhancing insulin
transport across RAECM-II.

58
7. Analysis and identification of protein bands, obtained from SDS- PAGE of
rat BALF (fraction #22), using LC-MS/MS and Mascot Search Engine.

69
8. Effect of protein concentration in BALF fraction #22 on enhancing insulin
transport across RAECM-II.

71
viii
List of Figures
1. Schematic representation of the protein transport pathways across the
alveolar epithelial cell barrier.

6
2. 2-D electrophoretic gel of human epithelium lining fluid.

8
3. Effects of various media on apical-to-basolateral transport of insulin across
RAECM-II.

34
4. Comparison of insulin transport across RAECM-I and RAECM-II
(a-to- b) and (b-to-a).

35
5. Comparison of insulin transport across RAECM-I and RAECM-II
(SM, CMI, and CMII).

36
6. Effect of culture medium containing 1% or 10% newborn bovine serum
(SM) on insulin transport across RAECM-II.

38
7. Effect of CMII on paracellular transport and fluid-phase transcytosis across
RAECM-II.

40
8. Temperature dependence of CMII-enhanced insulin transport across
RAECM-II.

41
9. Effect of proteinase inhibitors (PI) on insulin transport across RAECM-II.

43
10. Tannic acid staining of RAECM-II.

45
11. Immunofluorescence staining of RAECM-II.

45
12. Size exclusion gel chromatography of Man-Alb.

48
13. Effect of BALF on insulin transport across RAECM-I and -II.

51
14. Transport of insulin across RAECM-II, MDCK and Caco-2.

52
15. Effects of temperature and monensin on enhanced insulin transport across
RAECM-II.

55
16. Polyacrylamide gel electrophoresis of BALF retentate.

59
17. Western blot of SP-A present in rat BALF.

60
ix
18. Effect of human SP-A and anti-SP-A antibody on insulin transport across
RAECM-II.

62
19. Transport of insulin across RAECM-II dosed in lyophilized BALF.

64
20. Salting out and partial purification of BALF.

65
21. Relative transport activity of partially purified BALF fractions.

66
22. Elution profile and SDS-PAGE of BALF proteins.

68
23. Effect of purified BALF on insulin transport across RAECM-I and -II.

73
24. Purification of rat plasma.

75
25. SDS-PAGE of purified rat plasma and BALF fractions.

75
26. Comparison of the effect of rat plasma and BALF on insulin transport
across RAECM-II.

77
27. Western blot of megalin from RAECM-II and mouse kidney homogenate.

78
28. Inhibitory effects of anti-megalin antibody (1H2) on the enhanced insulin
transport RAECM-II.

80
29. Comparison of insulin transport in BALF versus alpha 2-macroglobulin.

82
30. Elution profile of α
2
M and
125
I-insulin using S-300 gel chromatography.

83
31. Transport of α
2
M·
125
I-insulin complex across RAECM-II.

85
32. Inhibitory effect of activated α
2
M on the enhancement of insulin transport
in BALF across RAECM-II.

86
33. Macroglobulin family and low density lipoprotein receptor family
(LDLR).

100
34. Schematic of postulated interaction between activated α
2
-M and α
1
I
3

complex with LRP.

101
35. Summary of findings. 111

x
Abbreviations
BALF Bronchoalveolar lavage fluid
BSA Bovine serum albumin
Caco-2 Human colon carcinoma
CMI Conditioned medium for RAECM-I
CMII Conditioned medium for RAECM-II
DMEM Dulbecco’s modified minimum essential medium
ELF Epithelial lining fluid
KGF Keratinocyte growth factor
KRP Kreb’s Ringer’s phosphate solution
LRP Low density lipoprotein receptor-related protein
MDCK Madin-Darby canine kidney
MEM Eagle’s minimum essential medium
NBS Newborn bovine serum
PBS Phosphate buffer saline
RAECM-I Rat alveolar epithelial cell monolayers type I cell-like
RAECM-II Rat alveolar epithelial cell monolayers type II cell-like
SDS-PAGE Sodium-dodecyl-sulfate polyacrylamide gel electrophoresis
SM Serum medium
SP-A Surfactant protein-A
TCA Trichloroacetic acid
TEER Transepithelial electrical resistance
xi
TFF Tangential flow filtration
α
1
I
3
alpha-1 inhibitor 3
α
2
M alpha-2 macroglobulin

xii
Abstract

Pulmonary delivery of peptides and proteins has been studied for decades and
the recent approval of Exubera® by the US and European markets, has made non-
invasive delivery of these drug macromolecules through the lungs attainable.
Despite the success of several proteins and peptides in achieving appreciable
systemic bioavailability relative to other non-invasive routes of administration such
as nasal and oral, the underlying mechanisms of absorption remain ambiguous.
Certainly the large surface area, rich blood supply, and minimal proteolytic activity
compared to the gastrointestinal tract, may explain the lung’s efficiency in providing
a suitable environment for drug absorption. However, it is by understanding the
intricate cellular machinery provided by the alveolar epithelium (proposed site of
systemic absorption), that further mechanisms of drug absorption may be unraveled.
In this dissertation, we investigated the effects of conditioned medium obtained from
cultured primary rat alveolar epithelial cell monolayers (CMII), and rat
bronchoalveolar lavage fluid (BALF) on the transport of insulin across the alveolar
epithelium. Our findings indicated that protein factor(s) present in both CMII and
BALF, significantly enhanced the transport of insulin across rat alveolar epithelial
cell monolayers (RAECM). The mechanism of enhancement was not due to
paracellular leakage or fluid-phase endocytosis, but involved transcellular
pathway(s), possibly receptor-mediated endocytosis. Upon partial purification of
BALF, the factor(s) with most prominent effect for enhancing insulin transport
across RAECM resided in the fraction containing high molecular weight proteins
xiii
(>200 kDa). The proteins, which were identified by LC-MS/MS, belonged to the
macroglobulin family. The mechanism afforded by macroglobulin partially
explained the insulin transport-enhancing effect observed with BALF and CMII
across RAECM suggesting that other factors may be involved. The observation that
macroglobulin may be a participating factor in enhancing insulin transport across
RAECM is a novel finding, which if extended to in vivo implications, may provide
an additional inference of the efficiency of inhaled therapeutics that are systemically
absorbed via the lung.

1
Chapter One: Introduction
1. Lung as a route for systemic drug delivery
Systemic delivery of peptides and proteins via pulmonary routes has been of
interest over several decades and clinical trials of drug molecules such as insulin
have proven the potential and efficiency of inhalation as a method of systemic drug
delivery (Patton and Byron 2007). Although the lung provides a tight barrier
compared to the gastrointestinal and nasal epithelium, the high pulmonary
bioavailability of certain peptides and proteins relative to other routes of
administration suggests more complex absorption pathways and underscores the
importance of the lung as a route for non-invasive drug delivery (Table 1). Several
reasons may explain this: 1) large surface area of the human lung (>100 m
2
) and thin
layer of epithelial cells that form the alveolus, which is the site of conventional
systemic drug absorption, although recently the upper airways have also been
implicated in the absorption of a substantial amount of inhaled protein drug (Bitonti
and Dumont 2006); 2) lack of peristaltic motility and thin layer of epithelial lining
fluid which allow for a longer presence of inhaled protein in a more concentrated
manner than that seen with orally administered macromolecules; 3) lower
amounts/activity of metabolizing enzymes compared to the gastrointestinal tract, as
well as the presence of proteinase inhibitors in the epithelial surface fluid such as α
1
-
antitrypsin, α
2
-macroglobulin, among others, which may protect inhaled proteins
against degradation (Heinemann et al. 2001; Tronde et al. 2003).

2

Table 1. Bioavailabilities of some therapeutic peptides and proteins delivered via the
lung. This table is from ref. (Patton et al. 2004).

2. Alveolar epithelial cells: type I versus type II
Type I pneumocytes are thin, squamous cells that comprise about 95% of the
total surface area of the alveolar epithelium. Type II pneumocytes, which are the
progenitors of type I pneumocytes and are more in number, are cuboidal and only
cover about 5% of the total alveolar surface. Phenotypic variability between these
two cell types reflects the difference in physiological function of each. Although the
function of type I cells is mainly gaseous exchange, subcellular microscopy has
revealed the presence of decorative vesicles called caveolae, which surround the
apical membrane (Gumbleton 2001) and shuttle in a vertical direction across the cell.
Type I pneumocytes also express a number of ion transporters and receptors all of
which emphasize the role of these cells in maintenance of normal lung homeostasis,
development and remodeling (Williams 2003).
3
Functions of alveolar type II cells include: 1) synthesis and secretion of
surfactant; 2) transepithelial movement of solutes; and 3) regeneration of the alveolar
epithelium following lung injury (Castranova et al. 1988) (Haagsman and Diemel
2001). The major surfactant protein is SP-A which is responsible for regulating the
homeostasis of surfactant phospholipids and facilitating the lowering of surface
tension in the alveolus (Khubchandani and Snyder 2001). Another protein that is
secreted by type II cells is polymeric IgA via a receptor, pIgR, which is a well-
described

integral membrane glycoprotein that binds polymeric IgA at the basolateral

epithelial cell surface (Luton and Mostov 1999). This complex is transcytosed and
polymeric IgA is finally released into the mucosal lumen. pIgR was detected in
isolated type-II but not type-I cells (Gonzalez et al. 2005) and was also involved in
the transcytosis of “secretory component” in cultured type I cell-like monolayers
(Kim and Malik 2003). Type II cells have also been implicated in the secretion of
various chemokines, either constitutively or in response to external stimuli (Thorley
et al. 2005). Parathyroid hormone-related protein was also shown to be expressed by
adult rat lung and by cultured rat alveolar type II epithelial cells and to act as an
autocrine regulator of differentiation for these cells (Hastings et al. 1997; Hastings et
al. 1994).
As mentioned earlier, most of the alveolar epithelium is comprised of type I
cells (>95% of surface) and may appear more relevant in studying drug delivery. In
my thesis work, however, I focused mainly on type II cells for several reasons: 1)
presence of dynamic secretory/reabsorption mechanisms in type II pneumocytes; 2)
4
reliable and efficient extraction of the alveolar cells is limited to type II pneumocytes
which then transdifferentiate into type I-like pneumocytes in vitro; 3) absence of any
indicative evidence that type I pneumocytes are involved in any major
macromolecular trafficking as compared to the type II pneumocytes.

3. Mechanisms of transport across the lung epithelium
Absorption mechanisms of peptides and proteins across the alveolar
epithelium are complex. Proteins and peptides with molecular weight less than 40
kDa tend to escape paracellularly through the junctional complex present between
two or three cells with rates that are inversely proportional to the molecular weight
(Patton 1996). Macromolecules greater than 40 kDa are thought not to pass through
these pores and are likely to be transported transcellularly by mechanisms such as:
fluid-phase endocytosis, macropinocytosis, caveolae- or clathrin-mediated
endocytosis (Fig. 1). The impact of each pathway is very difficult to determine in
whole lung experiments because the transport probably involves more than one
venue (e.g., airways versus alveoli), necessitating studies in simpler models such as
cultured alveolar epithelial cells. Caveolae-mediated endocytosis has been
demonstrated mainly in type I cells where for example albumin is taken up through
the gp-60 receptor (John et al. 2001). Scientists have speculated the use of caveolae
as potential sites for drug transport; however, many studies concluded that the
transport appeared to be relatively slow and that no strong evidence (even with the
use of electron microscopy) revealed unique involvement of caveolae (Gumbleton et
5
al. 2003). Clathrin-mediated endocytosis exists in both type I and type II
pneumocytes and include receptors for molecules such as SP-A (Wissel et al. 2001)
and megalin (Moestrup 1994) found exclusively on type II cells, or IgG (Kim et al.
2004), and transferrin (Matsukawa et al. 2000), (Widera et al. 2003), found on either
cell types. The role of these receptors in transcytosis has been proven in several in
vitro studies (Matsukawa et al. 2000), (Spiekermann et al. 2002), (Deshpande et al.
1994) indicating their use as potential pathways for systemic drug delivery.
Adsorptive endocytosis, which mainly depends on electrostatic interaction between
cell membrane and exogenous protein, can occur via both cell types although some
studies have shown that cationic ferritin adsorbed preferentially onto type II cells
rather than type I cells (Williams 1984), (Simionescu and Simionescu 1983) due to a
higher density distribution of negative moieties on type II cells. Our lab has also
recently reported that when insulin was chemically conjugated to polyarginine, the
transport of this conjugate was significantly higher across rat type II cell-like
alveolar epithelial monolayers as compared to type I cell-like alveolar epithelial
monolayers (Patel 2007), underscoring the importance of cell surface charge in this
delivery system.

6

Figure 1. Schematic representation of the protein transport pathways across the
alveolar epithelial cell barrier. This figure is from ref. (Kim and Malik 2003).

4. Alveolar epithelial lining fluid components and their effect on lung
microenvirons
The epithelial lining fluid (ELF) of the lungs consists of a complex mixture
of proteins, lipids, and mucopolysaccharides, which is involved in the homeostasis
and protection of the lung. It acts as a reservoir for lipid surfactants and is rich in
proteins such as those produced by the type II pneumocytes (e.g., surfactant proteins
and cyto/chemokines) and alveolar macrophages (e.g., cytokines/chemokines and
growth factors) or those of the plasma origin (e.g., IgG, IgA, transferrin,
macroglobulins, alpha 1-antitrypsin) (Wattiez and Falmagne 2005) (Wattiez et al.
1999). The volume of ELF is ~7 mL in adult humans and ~0.4 mL in rats (Watkins
and Rannels 1979).
7
The closest representation of this ELF microenvironment is the bronchoalveolar
lavage fluid (BALF), where a buffer solution (e.g., phosphate buffered saline) is
instilled into the lungs and washed out (i.e., withdrawn). It has been reported that
BALF yields an accurate reflection of ELF protein composition (Wattiez et al. 1999).
BALF has been utilized to determine some parameters of the lung’s condition or
health states. For example, BALF has been useful in identifying disease biomarkers
and thought to reflect physiological relevance of cellular and non-cellular
components (Reynolds 2000), (Mason et al. 1994). The 2-dimensional
electrophoresis gel (Fig. 2), illustrates the presence of more than 1500 spots
representing different proteins of the human ELF (Wattiez et al. 1999).
8

Figure 2. 2-D electrophoretic gel of human epithelium lining fluid. This figure is
from ref. (Wattiez et al. 1999).

However, in context of drug delivery, the ELF may comprise one of several
barriers to the air-to-blood transport of inhaled macromolecules (Patton 1996). How
do these soluble factors modulate the behavior of the epithelial cells, and if such an
interaction does exist, is it possible that it may influence the transport of the inhaled
macromolecule? How do endogenous lung proteins concentrated in the small
volume of the ELF affect the shape, charge, and absorption of the administered
peptide or protein drug? An example of interplay between cells of alveolus is
depicted by Mason et al. who showed that hepatocyte growth factor present in BALF
caused proliferation of alveolar type-II cells in vitro (Leslie et al. 1985). This factor
is predominantly secreted by alveolar macrophages (Leslie et al. 1985) and becomes
9
most relevant in cases of lung injury when proliferation of type II cells is necessary
for the restoration of damaged type I cells. Cytokines also produced by either
alveolar macrophages or type II cells affect the behavior and homeostasis of the
alveolar microenvironment as in the case of the critical role of GM-CSF in regulating
surfactant production (Trapnell and Whitsett 2002) or when IL-8 is expressed by
bronchial epithelial cells triggered by the release of defensins from neutrophils
present in the lung as a response to innate host defense (Sakamoto et al. 2005). More
recently, an interesting study by Lee et al. found that alveolar fluid from patients
with acute lung injury caused a decrease in protein levels of ion transporters
( αENaC, α1NaKATPas and CFTR) present in type II pneumocytes and thus caused a
reduction in protein clearance in these patients (Lee et al. 2007). On the contrary,
Gandhi et al. showed that exposure of type II pneumocytes to cardiogenic pulmonary
edema fluid increased expression of Na
+
permeant channels due to globulin-like
factor(s) (Gandhi et al. 2007). Expression of gp330 on type II pneumocytes is
another interesting example of the coexistence of supportive alveolar machinery
whereby lipoprotein lipase (Khoo et al. 1984), produced by the alveolar
macrophages, is taken up by gp330 to help in fatty acid processing and surfactant
synthesis, a hallmark of type II pneumocytes in vivo.
There is, therefore, an intricate system of interactions and signaling among
the molecular and cellular components of the lung that boost its innate response and
maintain a balanced microenvironment. The latter phenomenon is illustrated by the
mechanisms involved in the clearance of proteins that infiltrate the loose
10
endothelium into the lung cavities through “permeable” regions of the broncho-
alveolar epithelium or otherwise that result from pathological conditions such as
alveolar proteinosis. Mechanisms of protein clearance are performed by either 1)
removal due to mucocilliary action; 2) phagocytosis by alveolar macrophages; or 3)
absorption across the alveolar and bronchial epithelium. To be transported, however,
the endogenous protein appears to have a cognate receptor expressed on the
pulmonary epithelium. Such is the case for albumin (John et al. 2001), (Kim et al.
2003), immunoglobulin G (Kim et al. 2004), SP-A (Hastings et al. 1994), among
others which have shown saturation kinetics in either in vitro alveolar epithelial cell
culture or in whole lung models. Though the studies are inconclusive about the role
of transcytosis in alveolar protein clearance, most agree that under low protein
concentration i.e., under normal physiological situation, endocytosis via gp-60 for
albumin, FcRn for IgG, or surfactant protein receptor for SP-A may be a contributing
mechanism for the maintenance of a low protein concentration in the epithelial lining
fluid (Hastings et al. 2004). In our current investigation, we focused on two
candidate proteins in the ELF, namely surfactant protein A and alpha-1 inhibitor III.

4.1. Surfactant protein A
Given that the infallible absorption of macromolecules via the nasal, buccal
or oral routes necessitates the use of penetration enhancers, while the lung seems
permeable to a number of peptides and proteins, raises the question of whether the
presence of surfactant lipids and proteins concentrated in a very little volume of lung
11
lining fluid actually assists in the documented high bioavailability seen for such
macromolecules. SP-A structure resembles that of collectins which consist of
oligomers with C-terminal carbohydrate recognition domains (CRD) in association
with N-terminal collagen-like domains (Haagsman and Diemel 2001). The unique
structure of SP-A allows it to associate with various ligands such as surfactant lipids
as well as viruses, bacteria, allergens, DNA, and most recently to IgG (Wright et al.
1987) (Haagsman 1998) (Sidobre et al. 2000) (Ross et al. 1995) (Lin and Wright
2006). This association enhances the uptake of SP-A-ligand complex by either
alveolar macrophages or alveolar type-II cells. The interaction of SP-A with the
ligand is shown to be either calcium-dependent when the carbohydrate recognition
domains were involved in binding to sugar molecules such as mannose, fucose,
galactose or glucose (Haagsman et al. 1987), or calcium-independent when SP-A is
incubated with liposomes or mycobacterial lipoglycans (Wright et al. 1987) (Sidobre
et al. 2000). The interaction in the absence of calcium is mediated through
hydrophobic binding between the fatty acids present in the lipoglycans and the
“neck” segment flanking the CRD, which is characterized by a short stretch of
hydrophobic residues (Ross et al. 1986). However, when calcium ions are increased
above a particular concentration, the presence of SP-A induces aggregation of the
lipid surfactants and results in enhanced uptake of the SP-A-lipid complex by type II
pneumocytes (Benson et al. 1984).

12
In addition to the in vivo secretion of surfactants, SP-A and SP-D are both
produced in vitro (Mason et al. 2002) (Rooney et al. 1993) by freshly isolated
alveolar type-II cells which express a high-affinity SP-A receptor responsible for
uptake and recycling of surfactant components (Kuroki et al. 1988). Subsequent to
the uptake of the SP-A-lipid complex through clathrin-mediated endocytosis, the
complex is transported to early endosomes but then takes different intracellular
pathways where SP-A is recycled rapidly toward the cell surface, and the lipid is
transported to the lamellar bodies (Wissel et al. 2001). Also recently, Mason et al.
have shown that type II pneumocytes, when are cultured on a substrata made of rat-
tail collagen and Engelberth-Holm-Swarm tumor basement-membrane mixture and
supplemented with specific growth factors, secrete high levels of SP-A and SP-D
into the culture medium (Mason et al. 2002).

13
4.2. Alpha-1 inhibitor III
Another protein which has not been studied much in the context of drug
delivery but which is also found in ELF of rat lungs is the plasma protein, alpha-1
inhibitor III ( α
1
I
3
). It is a member of the macroglobulin family that binds to and
inhibits the action of proteinases belonging to all four mechanistic classes (i.e.,
serine, cysteine, aspartate, and metalloproteinase) (Pizzo 2001). A homologue of rat
α
1
I
3
called alpha-2 macroglobulin ( α
2
M) exists in human plasma (2-4 mg/mL)
(Sottrup-Jensen et al. 1989). Both macroglobulins share a similar mechanism of
proteinase inhibition and are cleared through a common low density lipoprotein-
receptor-related protein (LRP) (Gliemann and Sottrup-Jensen 1987). In addition,
α
2
M forms complexes with numerous growth factors, cytokines, and hormones either
through hydrophobic interactions or covalent bonding (Liu et al. 2001), (Gron and
Pizzo 1998) (Mathew et al. 2003). Some studies have utilized α
2
M for delivery of
insulin to hepatoma cells or alpha-galactosidase to fibroblasts through either
disulfide bond conjugation or complexation in the presence of trypsin, respectively
(Ito et al. 1984) (Osada et al. 1987).
As mentioned earlier, activated α
1
I
3
has been shown to bind to LRP/ α
2
MR.
This receptor, as defined by immunohistochemical labeling, exists on bronchial and
alveolar epithelium of human tissue (Moestrup et al. 1992). However, there is
conflicting data regarding its presence on rat alveolar epithelium (Zheng et al. 1994)
(Stefansson et al. 1995). In fact, a closely related receptor, gp330 (also known as
megalin), which is also a member of the low density lipoprotein receptor family,
14
shows a high distribution in rat type II pneumocytes (Zheng et al. 1994). Gp330 is
expressed on the apical surface of type II pneumocytes and is involved in clathrin-
mediated endocytosis of various ligands (Moestrup 1994). It has been reported that
megalin is involved in the internalization of a number of plasma protein carriers such
as folate binding protein (Birn et al. 2005), retinol binding protein (Christensen et al.
1999), and vitamin D binding protein (Nykjaer et al. 1999). The endocytotic fate of
gp330 ligand usually meets the lysosome where the delivered ligand is digested, and
the carrier is recycled to the apical membrane. However, there have been situations
where megalin-mediated transcytosis of thyroglobulin or retinol-binding protein in
thyrocytes or in renal proximal tubule cells, respectively, escape the lysosome and
are able to trancytose intact (Marino et al. 2000) (Marino et al. 2001).
To date, no such studies have been performed to investigate the incorporation
of proteins into rat α
1
I
3
, but we hypothesize that since rat α
1
I
3
is a monomer of α
2
M,
similar mechanisms with different kinetics and affinities may be involved in
enhancing peptide absorption across alveolar epithelium.

5. Methods of enhancing drug transport across the lung epithelium
Cellular and whole lung studies of pulmonary insulin transport have indicated
predominance of a paracellular pathway (Patton 1996) (Yamahara et al. 1994a). The
use of various chemical agents for enhancing protein delivery through the lung, has
met with appreciable success (Table 2). Amphipathic agents such as bile salts
(Yamamoto et al. 1994) (Komada et al. 1994) (Johansson et al. 2002), and polymers
15
such as carbopol (Li et al. 2006) among many other conventional tight junction
modulators, have been utilized for paracellular enhancement of insulin and other
macromolecule transport. Peptidase and proteinase inhibitors have also been studied
for their effects on enhancement of pulmonary peptide/protein absorption and have
been shown to improve the stability and bioavailability of insulin (Fukuda et al.
1995) (Shen et al. 1999) (Kobayashi et al. 1994) although disruption of tight
junctions in the lung epithelial barrier by these agents poses a potential threat of lung
edema. More recent strategies for improving pulmonary delivery of peptide/protein
have utilized the dynamic properties of the lung epithelium and endogenous proteins
present in the epithelial microenvironment. For example, erythropoietin-Fc fusion
protein in vitro (Bitonti et al. 2004), or in vivo (Bitonti and Dumont 2006), has
shown enhanced absorption presumably through the Fc receptor expressed by the
pulmonary epithelial cells. Transferrin conjugated to granulocyte colony stimulating
factor (GCSF) or horseradish peroxidase (HRP), has also led to increased transport
of proteins across the alveolar epithelium, in vitro, via receptor-mediated endocytosis
for transferrin (Widera et al. 2003) (Deshpande et al. 1994). Also interesting is the
recent use of a fusion protein consisting of alpha 1-antitrypsin and a single-chain Fv
(targets the extracellular portion of pIgR) which facilitates the delivery of the anti-
proteinase to the apical surface of human tracheal xenografts to a greater degree
compared to that of anti-proteinase alone (Ferkol et al. 2003).

16
To date, no investigation of endogenous protein factor(s) for increasing
insulin absorption in the lungs has been reported. Data from our laboratories and
others regarding transport of small proteins such as insulin and GCSF across rat
alveolar epithelial cell monolayers, shows low rates of transepithelial passive
restricted diffusion (Widera et al. 2003) (Matsukawa et al. 2000). However, the
presence of certain entities in the surrounding microenvirons of the alveolar
epithelium has a direct effect on uptake of drugs. For example, surfactant lipids,
which are endogenous in the lungs and secreted by type II pneumocytes, have been
implicated in insulin absorption enhancement (Mitra et al. 2001).
17
Table 2. Agents used to enhance pulmonary protein absorption.

Enhancer Mechanism Example Ref.
surface-active
agents
modulation of tight junctions
and/or fluidization of cell
membrane
Na-glycocholate
Span 85
(Yamamoto et al.
1994)
protease
inhibitor
inhibition of exopeptidase
and/or endopeptidase activity
Na-glycocholate
aprotinin
bacitracin
actinonin
camostat
mesylate
(Fukuda et al. 1995)
(Shen et al. 1999)
(Morimoto et al.
1993)
(Yamahara et al.
1994b)
carrier-
mediated
transcytosis through receptor-
mediated endocytosis
Fc-erythropoietin
Tf-GCSF
(Bitonti et al. 2004)
(Widera et al. 2003)

6. Primary rat alveolar epithelial cell culture as in vitro model
Primary culture of rat alveolar epithelial cell monolayers has been widely
used as a reliable in vitro model for mechanistic studies of peptide and
macromolecule drug delivery via the distal respiratory epithelial tract of the lung.
The in vitro model used in this project resembles in vivo alveolar epithelium in
morphology and phenotypic characteristics, and has been successfully used to study
transport of various proteins and peptides (Brandes and Finkelstein 1989)
(Matsukawa et al. 2000) (Widera et al. 2003). When plated on tissue culture-treated
polycarbonate filters, the isolated type II pneumocytes begin to transdifferentiate by
day three and onward into type I cell-like pneumocytes, acquiring thin cytoplasmic
processes and losing their characteristic lamellar bodies and surfactant protein
production. The presence of keratinocyte growth factor (KGF) in the culture
medium has been shown to inhibit or reverse such transition (Borok et al. 1998),
allowing the cultured cells to retain type II cell-like phenotype and morphological
traits.
18
Chapter Two: Objectives
The objective of my dissertation was to investigate the effect of endogenous
cellular factor(s) on insulin transport across primary cultured rat alveolar epithelial
cell monolayers (RAECM). The specific aims of my research were:
1) To study the effect of CMII on insulin transport across RAECM.
a) To characterize CMII factor(s) involved in insulin transport enhancement.
b) To elucidate the mechanism of insulin transport enhancement afforded by
CMII factor(s)
2) To study the effect of bronchoalveolar lavage fluid (BALF) on insulin transport
across RAECM.
a) To characterize BALF factor(s) involved in insulin transport enhancement.
b) To elucidate the transport-enhancing mechanism of BALF factor(s)
c) To purify and identify BALF factor(s)
3) To compare the effect of alpha-2 macroglobulin ( α
2
M) and alpha-1 inhibitor III
( α
1
I
3
)

on insulin transport across RAECM-II.
a) To elucidate the effect of α
2
M on insulin transport across RAECM-II
b) To determine the pathway of enhanced insulin transport through competition
assays
The findings in these studies will provide further insight of the effect of the
alveolar microenvironment on insulin transport, and understanding of possible
mechanisms involved between endogenous cellular proteins of the lung and inhaled
protein drug.
19
Chapter Three: Materials and Methods
1. Cell culture
1.1. Rat alveolar epithelial cell monolayers culture
Primary rat alveolar epithelial cells were obtained and purified as previously
described. (Borok et al. 1998). The partially purified alveolar type II cells (>90%
purity and >90% viability) were plated onto tissue culture-treated polycarbonate
filters (12 mm Transwells, 0.4 µm pore size, Costar Corning, Cambridge, MA, USA)
at 10
6
cells/cm
2
and cultured for 6 days at 37
o
C in a humidified atmosphere of 5%
CO
2
and 95% air. Culture medium consisted of a defined serum-free medium
(MDSF, a 1:1 mixture of Dulbecco’s modified minimum essential medium (DMEM)
and Ham’s F-12, Sigma Chemical, St. Louis, MO, USA) supplemented with 10%
newborn bovine serum, 1.25 mg/mL bovine albumin, 100 U/mL penicillin and 100
ng/mL streptomycin. Cells were fed from day 3 and every other day onward. These
cells transdifferentiate into cells bearing type I cell-like morphology and phenotype
and are designated as RAECM-I. In order to maintain the type II cell-like phenotype
and morphology, keratinocyte growth factor (KGF, 10 ng/ml) was added to the
bathing media from day 0 and also supplemented to the fresh culture medium for
feeding cells on day 3 and onward. This latter group of cultured cells was designated
as RAECM-II. The transepithelial resistance (TEER) and potential difference (PD)
of these RAECM-I and -II, when measured on day 6 using an epithelial volt-ohm
meter (EVOM, World Precision Instruments, Sarasota, FL, USA) are > 2,000
ohm·cm
2
and > 10 mV (apical side negative), respectively.
20
1.2. MDCK cell culture
Madin-Darby canine kidney (MDCK) cells (2.2 x 10
4
cells/cm
2
), a strain we
originally obtained from Upjohn (Kalamzoo, MI), were cultured on Transwells using
Eagle’s Minimum Essential Medium (MEM) supplemented with 10% (v/v) fetal
bovine serum (FBS), 0.5% (v/v) penicillin (10,000 units/mL)/streptomycin (10
mg/mL) solution (Sigma), 1% (v/v) L-glutamine (200 mM, GIBCO, Grand Island,
NY), and 1% (v/v) non-essential amino acid solution (100 mM, Sigma). On day 3,
MDCK cell monolayers exhibited a TEER of 1,000 ohm·cm
2
and were replenished
with the same culture medium but containing only 2.5% FBS. The concentration of
FBS was reduced further to 1% from day 5 and onward till a final TEER of 2,500
ohm·cm
2
was achieved (usually by 7-8 days in culture).

1.3. Caco-2 cell culture
Human colon carcinoma (Caco-2) cells (4.2 x 10
4
cells/cm
2
) obtained from
ATCC (Manassas, VA) were cultured on Transwells using DMEM supplemented as
described above for MEM with 10% FBS which was used throughout the culture
period. Caco-2 cells were fed every 3 days and confluence was reached within 7-8
days. The monolayers were maintained an additional 7-13 days whereby a TEER of
400 ohm·cm
2
was obtained.
21
2. Transport of insulin
Human insulin (Sigma) (Mwt. 56 kDa, IP 5.5) was iodinated using the
chloramine T method (McConahey and Dixon 1980).
125
I-insulin had a specific
activity of 3 x 10
9
cpm/mg. On day 5 or 6, RAECM-II were washed once with
MDSF and incubated for 30 min at 37
o
C. Apical medium (0.5 mL) was aspirated
and replaced with medium, conditioned medium, BALF or saline buffer (either KRP,
or PBS) containing
125
I-insulin. At 2 h post-dosing, basolateral fluid was collected
and total radioactivity was measured using a Packard gamma counter. Basolateral
samples were then treated with 15% trichloroacetic acid (TCA) on ice for 15 min,
followed by centrifugation at 2,235 x g for 15 min at 4
o
C. After aspiration of
supernatants, the radioactivity associated with pellets was measured again in the
gamma counter to determine intact insulin-associated radioactivity. In some
experiments, apical-to-basolateral transport of
125
I-insulin across RAECM-II was
compared with tranport across RAECM-I, MDCK, and Caco-2 using the same
procedure described above except that MDCK and Caco-2 cell monolayers were
washed once with serum-free MEM and DMEM, respectively, before transport
experiments.
22
3. Conditioned medium collection and characterization
3.1. Molecular sizing
On day 6, conditioned media are collected from the apical (and basolateral)
bathing media of either RAECM-I or RAECM-II and designated as aCMI (bCMI) or
aCMII (bCMII), respectively. Of these various conditioned media, we focused on
characterizing aCMII (hereafter designated as CMII for clarity), as others did not
show much of activities in enhancing insulin transport (see details in results below).
CMII was concentrated in the Centriplus YM-50 (molecular weight cutoff of 50 kDa,
Millipore, Amicon, USA), where both retentate and filtrate were used to test the
activity of CMII in modulation of peptide transport across RAECM-II, by dissolving
radiolabeled insulin in CMII retentate and filtrate and transport study was initiated as
described above.

3.2. Effect of serum, KGF, conditioned medium for RAECM-II (CMII) and
conditioned medium for RAECM-I (CMI) on insulin transport
Radiolabeled insulin was dissolved in defined serum-free medium (SFM),
SFM supplemented with 1 or 10% newborn bovine serum (SM), or SM
supplemented further with KGF (10ng/mL). Conditioned media, CMII and CMI,
from rat AEC cultures with and without KGF, respectively, were collected and used
to dose radiolabeled insulin for comparison using RAECM-I and -II. In some
experiments, 1% serum was used in lieu of 10% serum as culture medium to
examine if the extent of transport enhancing factor(s) in CMII would be affected.
23
3.3. Characterization of conditioned medium
To test the heat lability of the putative transport enhancing factor(s), an
aliquot of CMII was heated at 80
o
C for 15 min, cooled, centrifuged to remove
coagulated proteins and used for transport studies of radiolabeled insulin.
Ammonium sulphate precipitation was used to assess the nature of the factor(s)
present in CMII where a supersaturated 4M solution of (NH
4
)
2
SO
4
was added slowly
to an aliquot of CMII on ice. After centrifugation, the supernatant was removed and
the precipitate was re-suspended to original volume with water and dialyzed twice
against phosphate-buffered saline (PBS). The resulting protein-rich CMII was used
to dose radiolabeled insulin for transport studies, using PBS as vehicle control.

3.4. Transport of paracellular versus transcellular markers
To determine whether CMII was affecting the paracellular pathway of drug
transport,
14
C-mannitol (1 µCi/mL), a paracellular marker, was dosed in SM, CMI
and CMII for transport studies across RAECM-II at 37
o
C for 2 h. The basolateral
medium was then collected and measured for radioactivity using a liquid scintillation
counter (Beckman LS1801). Furthermore, to assess CMII effect on fluid phase
endocytosis, horseradish peroxidase (1 mg/mL) was dosed in a similar manner as
mannitol transport experiments and the amount transported was measured using
enzymatic assay (Herzog and Fahimi 1973). Transport of
125
I-insulin (4 µg/mL)
dosed in SM or CMII was studied across RAECM-II at 37, 16 and 4
o
C for 1 h to
determine the contribution of transcellular transport of insulin versus paracellular
24
transport by assessing the temperature-dependency of insulin transport across the
alveolar epithelial barrier. Finally, proteinase inhibitor cocktail (Sigma) added to the
dosing vehicle was used to study the effect of CMII versus SM on
125
I-insulin
degradation pathway.

4. Bronchoalveolar lavage fluid (BALF) collection and characterization
4.1. Rat lung lavage fluid collection
BALF was collected from normal Sprague-Dawley rats by lavaging the lungs
of anaesthetized rats three times (10 mL per lavage) with warm (37ºC), phosphate-
buffered Kreb’s Ringer’s phosphate solution (KRP, pH 7.4) comprised of 135 mM
sodium chloride, 5 mM potassium chloride, 24 mM dibasic sodium phosphate
septahydrate, 6 mM monobasic sodium phosphate monohydrate, 10 mM 4-(2-
hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), and 1 mg/mL glucose. The
rat lung lavage fluid was first centrifuged at 230 x g for 10 min at 4
o
C to remove
lung resident cells (e.g., macrophages) and then at 3685 x g for 15 min at 4
o
C to
remove further suspended debris present in the lavage fluid. The combined BALF
supernatant was concentrated 7.5-fold in Centricon YM-10 (Millipore Corp,
Bedford, MA) unless stated otherwise. The retentate from the 7.5x concentrate
(designated as 7.5x BALF hereafter) contained approximately ~0.5-1 mg/mL total
protein and was stored at 4
o
C for use within 5 days, or otherwise lyophilized
(Labconco, Freezone plus 6) and was kept at -20
o
C for use within no later than 2
months. In some instances, larger volume of BALF (500 mL) was concentrated 20-
25
fold using a tangential flow filtration system (TFF) which was fixed directly to a
Pellicon XL filter (50 kDa molecular weight cutoff). This yielded 20x BALF with a
higher protein concentration (2.5 mg/mL) which was necessary for factor(s)
identification.

4.2. Characterization of the insulin transport-enhancing factor(s) in BALF
To determine the molecular weight range that retains the insulin transport-
enhancing activity, rat bronchoalveolar lavage fluid was centrifuged using Centricon
devices with different molecular weight cutoffs (10, 30, and 50 kDa; or YM-10, -30
and -50). Radiolabeled insulin was added to each of the retentates and filtrates to
assess their effect on insulin transport across RAECM-II, as described above. Two
aliquots of the retentate obtained from BALF concentrated with a Centricon YM-10
were treated either by heating at 80
o
C for 15 min, cooling and centrifuging to remove
precipitated material; or by digesting with trypsin (10 µM), incubating for 30 min at
37
o
C and quenching trypsin activity with soybean trypsin inhibitor (20 µM) (Sigma).
The effect of heat lability or trypsin digestion on the putative transport-enhancing
factor(s) was assayed by measuring the apical-to-basolateral transport of
125
I-insulin
across RAECM-II.

26
4.3. Transport of paracellular versus transcellular markers
The influence of BALF factor(s) on the paracellular drug transport was
determined using the paracellular markers:
3
H-mannitol (1 µCi/mL), and sodium
fluorescein (1 mg/mL) which were dosed in BALF or saline buffer (representing
buffer used for lung lavage) for transport studies across RAECM-II at 37
o
C. Apical-
to-basolateral transport of
3
H-mannitol and sodium fluorescein after 2 h incubation
was measured in the basolateral receiver fluid using a liquid scintillation counter
(Beckman LS1801) and a fluorometer (Hitachi F-200), respectively. Furthermore, to
assess the effect of BALF factor(s) on fluid phase endocytosis, FITC-dextran 70 kDa
(0.5 mg/mL) was apically dosed in a similar manner as mannitol transport
experiments and the basolateral receiver fluid was assayed for fluorescence
To determine temperature-dependency, apical-to-basolateral transport of
125
I-
insulin dosed in 7.5x BALF or KRP was measured across RAECM-II at 37 and 4
o
C
after 1 h of apical dosing. Effects of an endocytosis inhibitor, monensin, were
determined by first pre-incubating RAECM-II with serum-free MDSF containing
monensin (60 µM) in both apical and basolateral compartments at 37
o
C for 0.5 h.
The apical medium was aspirated and replaced with
125
I-insulin dosed in 7.5x BALF
or KRP with or without monensin (60 µM) and the apical-to-basolateral transport of
125
I-insulin was measured at 37
o
C after 1 h of apical dosing as described above.

27
4.4. Column chromatography of BALF
The concentrated BALF (approximately 20 mL containing 0.5-1 mg/mL)
from a Centricon device YM-10 was lyophilized to reduce its volume and facilitate
its purification using gel filtration chromatography. The lyophilized BALF was
reconstituted with ~2 mL water and desalted using a Sephadex G-50 column (1 cm x
40 cm) that was pre-equilibrated with phosphate-buffered saline PBS, pH 7.4. The
absorbance of fractions (1 mL) was measured at 280 nm. Fractions with peak
protein absorbance were pooled and concentrated in a Centricon device (3.5 kDa
molecular weight cutoff). The retentate obtained from the 3.5 kDa cutoff device was
loaded onto a Sephacryl S-200 column (1 cm x 40 cm) that was pre-equilibrated with
PBS, pH 7.4 and fractions (1 mL) were collected and measured at 280 nm. Selected
fractions from the Sephacryl S-200 chromatography were pooled and concentrated
again in a Centricon device (3.5 kDa molecular weight cutoff) and tested for their
effect on insulin transport across RAECM-II.

4.5. LC-MS/MS and database search
Aliquots (35 µl) of sample fractions obtained from Sephacryl S-200
chromatography were analyzed using sodium-dodecyl-sulfate polyacrylamide gel
(7.5%) electrophoresis (SDS-PAGE). Coomassie stained protein bands in the gel
were excised and destained with 50% acetonitrile in 50 mM ammonium carbonate.
In-gel tryptic digest was carried out according to Gallaher et al. (Gallaher et al.
2006). Briefly, samples were reduced with 10 mM dithiothreitol (in 50 mM
28
ammonium carbonate) for 60 min at 56°C and alkylated with 55 mM iodoacetamide
(in 50 mM ammonium carbonate) for 45 min in the dark at room temperature. The
samples were digested overnight at 37°C using reductively methylated trypsin.
Digestion products were extracted from the gel twice with a 5% formic acid/50%
acetonitrile solution, followed by extraction once with 100% acetonitrile. The
solvent was evaporated using an APD SpeedVac (ThermoSavant, Holbrook, NY)
and samples were resuspended in 10 µL of 60% acetic acid, injected via autosampler
(Surveyor, ThermoFinnigan, San Jose, CA) and subjected to reverse-phase liquid
chromatography. Mass analysis was done using a LCQ Deca XP Plus ion trap mass
spectrometer equipped with a nano-spray ion source (ThermoFinnigan). The column
was equilibrated for 5 min at 1.5 µL/min with 95% solution A (0.1% formic acid in
water) and 5% solution B (0.1% formic acid in acetonitrile) prior to sample injection.
A linear gradient was initiated 5 min after sample injection, ramping to 35% solution
A and 65% solution B after 50 min and 20% solution A and 80% solution B after 60
min. Mass spectra were acquired in the mass/charge (m/z) range of 400-1800.
Protein identification was carried out with the MS/MS search software Mascot 1.9
(Matrix Science, London, UK)(Perkins et al. 1999) and complementary analyses
with TurboSequest as implemented in the Bioworks Browser 3.2, build 41
(ThermoFinnigan).

29
5. Rat plasma collection and purification
Rat plasma was drawn from on Sprague-Dawley rat after anaesthetizing with
ether. Briefly, blood was withdrawn from the heart and placed in a plastic tube
containing 20 µL heparin (1000 U/mL) to prevent coagulation. Blood was
centrifuged at 4°C for 20 min at 230 x g and the supernatant (plasma) was collected
and stored at -20°C. Aliquots of rat plasma were purified using Sephacryl S-200 as
described for BALF purification (Section 4.4).

6. Western blot analysis
6.1. Detection of SP-A in BALF
Concentrated BALF (1 mL) was precipitated with trichloroacetic acid (TCA),
centrifuged at 4
o
C for 10 min whereby the supernatant was removed and the process
repeated with a mixture of ethanol/methanol instead of TCA. The pellet was left to
dry and stacking buffer (pH 8.5) was added to ensure that the final solution was
neutralized. The final concentration of treated BALF was 100x fold of the original
crude BALF. The samples were loaded onto a 12% SDS-PAGE and the gel was
electroblotted on nitrocellulose membranes overnight at 4
o
C and 40mV. Blocking
and all antibody incubations were done using Li-Cor buffer (Biosciences, Lincoln,
Nebraska). The blot was blocked in Li-cor blocking buffer for 1 h at room
temperature and then was imunoblotted with Whitsett’s anti-SP-A antibody (rat SP-
A raised in human 1:1000) for 2 h with constant shaking.
30
Blots were washed for 15 min with four changes of Tris-Tween buffer and then
incubated with secondary antibody (1:1000) for 1 h at room temperature with
constant shaking. Blots were washed as before and scanned at 700 nm wavelength
using the Li-Cor Scanner (Biosciences).

6.2. Detection of megalin in mouse kidney homogenate and RAECM-II
Kidneys from anaesthetized mouse were excised, dissected and homogenized
in RIPA buffer supplemented with proteinase inhibitor cocktail using PT-MR-2100
Polytron tissue homogenizer (Switzerland). Insoluble materials were pelleted by
centrifugation and the supernatant was collected and stored at -20ºC. On day 6 of
culture, twelve inserts of RAECM-II plated on 12-well Transwell plate were
detached and lysed using RIPA buffer supplemented with proteinase inhibitors. To
evaluate megalin expression, samples from mouse kidney and RAECM-II
homogenate containing 50 µg and 5 µg protein, respectively were subjected to non-
reducing SDS-PAGE followed by Western blotting with 1H2ab (anti-megalin
antibody kindly provided by Dr. Brown at Mass General Hospital) using the same
procedure described for SP-A.

7. Immunofluorescence and tannic acid staining of RAECM-II
Primary rat alveolar epithelial cells which were plated on tissue-cultured T75
plates, were examined for the presence of lamellar bodies that are indicative of type
II cell-like phenotype on days 3 and 5 using immunofluorescence and tannic acid
31
staining. Cells (0.58 x 10
6
cell/cm
2
) were plated on tissue-cultured chamber slides
and incubated with MDS. Immunostaining of RAECM-II on day 3 or 6 followed the
protocol described by Borok et al. Briefly, cells were fixed in cold methanol for 10
min at 4
o
C, rinsed with PBS, pH 7.2 and blocked with 5% BSA for at least 1 h at
room temperature. Slides were then rinsed with PBS and incubated with primary
antibody against lamellar body protein (1:1000) for 1 h at room temperature. The
monolayer was rinsed with PBS three times for 5 min and post-fixed in 3.7%
formalin/PBS for 5 min at room temperature and rinsed as before. Mounting fluid
was added and slide was coverslipped and sealed. Slides were analyzed for the
presence of lamellar bodies using Zeiss LSM 510 Meta NLO Confocal Imaging
System. Tannic acid staining followed the method suggested by Mason et al.
whereby cells were plated on 6-well cluster plate (Mason et al. 1985), washed with
PBS and fixed with 1.5% glutaraldehyde for 15 min. Cell monolayers were washed
with PBS and further fixed with 1.0% osmium tetroxide for 1.5 h. Further rinsing
was followed by overnight treatment of the monolayer with freshly prepared 1.0%
tannic acid in PBS, 6.8. Next day, the monolayer was washed with PBS and distilled
water and stored in 0.1% sodium azide. Lamellar bodies were analyzed by
differential interference contrast (DIC) light microscopy using Zeiss LSM 510 meta
NLO Confocal Imaging System.

32
8. Mannosylation of albumin
Mannosylation of albumin was preformed according to Kataoka et al.,
(Kataoka and Tavassoli 1984). Briefly, thiophosgene (80ul) was added to p-
aminophenyl mannose (0.185mmol) in 80% ethanol (10ml) and the reaction mixture
was stirred for 1.5 h at room temperature. Nitrogen was bubbled to remove
remaining odor and the solution was adjusted to pH 6.0 and evaporated in vacuo.
This aqueous solution made with 5ml water was added slowly to a solution of
albumin (0.44 µmol in 0.01 M of 15 mL borate buffer, pH 9.0) while stirring and the
mixture was maintained at pH 9.0 overnight at room temperature. The solution was
dialyzed against 0.15 M sodium chloride and stored at -20
o
C.
The number of mannose residues attached to albumin was determined according to
Dubois et al., who used phenol in the presence of sulfuric acid for quantitative
colorimetric micro-determination of sugars (Dubois 1956).

8. Statistical analyses
Data are presented as mean ± SD. For comparisons of multiple ( ≥ 3) group
means, one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc tests.
Student’s t-tests were used to compare means of two groups. P < 0.05 was
considered to be statistically significant.
33
Chapter Four: Results
I. Conditioned medium: Impact on insulin transport
1. Insulin transport in CMII
A four-fold increase in the apical-to-basolateral transport of the intact
125
I-
insulin across RAECM-II was observed in the presence of the conditioned medium
which was collected from the apical compartment of RAECM-II (CMII), as
compared to that in fresh serum-supplemented medium (SM) (Fig. 3). This increase
of
125
I-insulin transport was not observed when CMII was replaced by the
conditioned medium collected from the basolateral compartment of RAECM-II
monolayers. As shown in Fig. 3, CMII-enhanced apical-to-basolateral transport of
125
I-insulin was significant only across RAECM-II, but not RAECM-I (p>0.05). On
the other hand, CMII did not affect the basolateral-to-apical transport of
125
I-insulin
across either RAECM-I or RAECM-II (Fig. 4). Furthermore, conditioned medium
from apical compartment of RAECM-I (CMI) did not have any effect on apical-to-
basolateral transport of
125
I-insulin in either RAECM-I or RAECM-II (Fig. 5).
Taking into consideration that after 2 h incubation RAECM-I exhibited a lower
TEER value than that for RAECM-II (1,530 vs 2,850 ohm·cm
2
), transport of
125
I-
insulin in CMI or CMII was normalized to the amount of its transport in SM across
either RAECM-I or RAECM-II, respectively. Based on these observations,
subsequent studies were focused only on CMII for the enhancement of apical-to-
basolateral transport of insulin across RAECM-II.

34

0
0.1
0.2
0.3
0.4
0.5
SM aCM II filtCM II bCM II
transport (ng/well)

Figure 3. Effects of various media on apical-to-basolateral transport of insulin across
RAECM-II. Apical-to-basolateral transport of intact insulin, when measured at
apical donor concentration of 4 µg/ml across RAECM-II in culture medium
containing 10% NBS (SM), conditioned apical medium from RAECM-II (aCMII),
filtrate from type II apical conditioned medium (filtCMII) centrifuged with
Centricon-50 at 2000 g, at 4
o
C for 30 minutes, and conditioned basal medium from
RAECM-II (bCMII). Data represent mean ± SD (n=3). * = significantly different
(p<0.01) from all others.
*
35

Figure 4. Comparison of insulin transport across RAECM-I and RAECM-II.
Apical-to-basolateral (a-to-b) and basolateral-to-apical (b-to-a) transport of intact
insulin, when measured at donor concentration of 4 µg/ml across A) RAECM-I and
B) RAECM-II in culture medium containing 10% NBS (SM) and conditioned apical
medium from RAECM-II (CMII). Data represent mean ± SD (n=3). * =
significantly different (p<0.05) from all others.
0
0.2
0.4
0.6
0.8
1
1.2
SM CMII
transport (ng/well)
a-to-b
b-to-a
0
0.2
0.4
0.6
0.8
1
1.2
SM CMII
transport (ng/well)
a-to-b
b-to-a
*
A.
B.
36

0
50
100
150
200
250
300
350
SM CM I CM II
% transport (normalized against the
data for SM)
RAECM-I
RAECM-II

Figure 5. Comparison of insulin transport across RAECM-I and RAECM-II.
Apical-to-basolateral transport of intact insulin, when measured at apical donor
concentration of 4 µg/ml across RAECM-I and RAECM-II in culture medium
containing 10% NBS (SM), conditioned apical medium from RAECM-I (CMI) and
conditioned apical medium from RAECM-II (CMII). Data represent mean ± SD
(n=3). * = significantly different (p<0.05) from all others.

*
37
2. Biochemical and physical characterization of CMII

To determine a molecular weight range for the factor(s) causing enhanced
insulin transport across RAECM-II, CMII was centrifuged in a Centricon device (50
kD molecular weight cutoff) and
125
I-insulin was dosed in the resultant filtrate
(filtCMII). Figure 3 showed that enhanced insulin transport was not seen when
filtCMII was used for dosing, consistent with the observation that the enhancing
effect of CMII resides with soluble factor(s) whose molecular weight equals to or is
greater than 50 kDa. Next, when CMII was heated at 80
o
C for 15 min, a complete
loss of effect on insulin transport across RAECM-II was observed (Table 3),
indicating that the enhancing factor(s) may be protein(s). Moreover, when
ammonium sulfate was added to CMII, about 95% of enhancement activity of CMII
was retained in the precipitated fraction, suggesting that the enhancing effect for
insulin transport is afforded by protein(s) in CMII.

Table 3. Characterization of CMII.

Treatment % Activity Remaining
Heat (80
o
C, 15 min) 0% ± 0.04% (n=3)
Protein precipitation (4M (NH
4
)
2
SO
4
) 95% ± 2% (n=3)

38
We next investigated the effects of supplementing serum on alveolar
epithelial elaboration of the factor(s) for enhancing insulin transport. After six days
of culture with 1% newborn bovine serum (NBS) in lieu of 10% serum, culture
medium from apical compartment of RAECM-II was collected and tested for the
enhancing effects on insulin transport across RAECM-II. Figure 6 shows that when
conditioned medium collected from monolayers cultured in 1% NBS
supplementation instead of 10% NBS (denoted as 1% CMII and 10% CMII,
respectively), 1% CMI had a significantly lower enhancement of insulin transport
across RAECM-II compared to that of 10% CMII. This latter data set indicates that
the production of enhancing factor(s) by RAECM-II may require component(s)
present in NBS.
0
0.2
0.4
0.6
0.8
1
1.2
1%CMII 10%CMII 1%SM 10%SM
transport (ng/well)

Figure 6. Effect of culture medium containing 1% or 10% newborn bovine serum
(SM) on insulin transport across RAECM-II. Apical-to-basolateral transport of
intact insulin, when measured at apical donor concentration of 4 µg/ml across
RAECM-II in either 1% or 10% SM, or 1% or 10% CMII. Data represent mean ±
SD (n=3). * = significantly different (p<0.05) from all others.
*
39
3. Effects of putative factor(s) in CMII for enhancing insulin transport on
paracellular versus transcellular pathways
The effect of CMII on transport of a paracellular marker (mannitol) and a
fluid-phase endocytosis marker (horseradish peroxidase), were also investigated.
Mannitol dosed in SM, CMI and CMII all showed similar magnitude of transport
across RAECM-II (Fig. 7A), with no significant changes in TEER (2,500 ohm·cm
2
).
Horseradish peroxidase transport (presumably occurring via fluid-phase transcytosis)
across RAECM-II was similar in magnitude, irrespective of medium used for dosing
(i.e., SM, CMI and CMII) (Fig. 7B). Further implications of CMII factor(s) on
endocytosis in RAECM-II were assessed by incubating radiolabeled insulin in either
CMII or SM at 37, 16 or 4
o
C. Figure 8 shows that insulin transport (dosed in CMII)
at 16 and 4
o
C was decreased 90% from that observed at 37
o
C. In contrast, the
amount of insulin transport when dosed with SM, was decreased by approximately
40% and 70% at 16 and 4
o
C, respectively.
40

0
0.04
0.08
0.12
0.16
0.2
SM CM I CM II
transport (ng/well)

0
10
20
30
40
50
60
70
80
SM CM I CM II
transport (ng/well)

Figure 7. Effect of CMII on paracellular transport and fluid-phase transcytosis
across RAECM-II. Apical-to-basolateral transport of A)
14
C-mannitol measured at
apical donor concentration of 0.02 µg/ml and B) horseradish peroxidase measured at
apical donor concentration of 1 mg/ml across RAECM-II in culture medium
containing 10% NBS (SM), conditioned apical medium from RAECM-I (CMI) and
conditioned apical medium from RAECM-II (CMII).
A.
B.
41

0
0.1
0.2
0.3
0.4
37 16 4
temperature
o
C
transport (ng/well)
SM
CMII

Figure 8. Temperature dependence of CMII-enhanced insulin transport across
RAECM-II. Apical-to-basolateral transport of intact insulin, when measured at
apical donor concentration of 4 µg/ml across RAECM-II in culture medium
containing 10% NBS (SM) and conditioned apical medium from RAECM-II (CMII)
at 37, 16 and 4
o
C. Data represent mean ± SD (n=3). * = significantly different
(p<0.05) compared to the data observed at 37°C with SM; ** = significantly
different (p<0.01) compared to the data observed at 37°C with CMII; † =
significantly different (p<0.01) compared to the data obtained for SM at 37ºC.

*
*
** **
†
42
4. Effect of CMII on
125
I-insulin degradation
Throughout the transport experiments (apical-to-basolateral) across RAECM-
II, it was observed that a smaller fraction of intact radiolabeled insulin (measured by
trichloroacetic acid precipitation) was present in the basolateral compartment when
dosed in CMII compared to that of SM (5% vs 10%, respectively). Incubation of
125
I-insulin in CMII or SM for 2 h at 37
o
C did not yield significant degradation in
either medium (data not shown), indicating that the factor(s) in CMII was not a
contributing source of
125
I-insulin degradation. However as shown in Fig. 9, the
addition of a proteinase inhibitor cocktail, which included mostly cytoplasmic and
lysosomal proteinase inhibitors, to CMII resulted in a significantly higher percent of
intact
125
I-insulin in the basolateral compartment of RAECM-II compared to that of
CMII devoid of inhibitors (20% vs 5%). On the other hand, there was no significant
difference in the amount of intact
125
I-insulin transported in SM in the presence or
absence of proteinase inhibitors.
43

0
1
2
3
4
5
SM CM II SM +PI CM +PI
traansport (ng/well)

0
0.1
0.2
0.3
0.4
0.5
0.6
SM CM II SM +PI CM II+PI
transport (ng/well)

Figure 9. Effect of proteinase inhibitors (PI) on insulin transport across RAECM-II.
Transport of A) total and B) intact insulin measured at apical donor concentration of
4 µg/ml across RAECM-II in SM and CMII without or with proteinase cocktail
inhibitor (SM+PI, CMII+PI, respectively). Data represent mean ± SD (n=3). * =
significantly different (p< 0.05) compared to SM; ** = significantly different (p<
0.01) from all others.

A.
**
B.
**
*
44
5. Investigation of the involvement of surfactant protein in the transport
enhancement of labeled protein

5.1. Primary alveolar epithelial cells cultured on T-75 as source of SP-A

Since only a limited amount of conditioned medium was obtained from cells
cultured on Transwell plates, primary rat alveolar epithelial cells were grown on
tissue-cultured T-75 plates to allow for volume scale-up of conditioned medium.
However, it was first necessary to determine whether lamellar bodies, which are
characteristic of type II cells, were conserved throughout the culture period (6 days)
in T-75 flasks. Immunofluorescence or tannic acid staining of lamellar bodies in
RAECM-II (Figs. 10 and 11) were detected on day three of culture. However, by
day six these lamellar bodies were only slightly detectable (even with the addition of
KGF to culture medium) eliminating the possibility of using T-75 plates for the
purpose of obtaining larger amounts of medium conditioned by RAECM-II.

45

A. B.

Figure 10. Tannic acid staining of RAECM-II. Primary rat alveolar cells (2x10
5
cells/cm
2
) were plated on tissue culture-treated 6-well cluster plate and treated with
tannic acid staining on day A) three and B) six. Cells were visualized using DIC
light microscopy.

A. B.

Figure 11. Immunofluorescence staining of RAECM-II. Primary rat alveolar cells
(5x10
4
cells/cm
2
) were plated on tissue culture-treated chamber slides and assayed
with immunofluorescence staining using primary antibody (1:1000) against lamellar
bodies on day A) three and B) six. Cells were visualized using confocal microscopy.
RAECM-II day 3 RAECM-II day 6
RAECM-II day 3 RAECM-II day 6
46
5.2. Mannosylated albumin as a marker for mechanistic investigation
With the assumption that the factor(s) present in CMII may be a surfactant
protein (SP-A), albumin was chemically modified by addition of mannose residues
to be used as a tracer molecule since mannose is known to bind to the lectin domains
of SP-A. The conjugate ratio of albumin:mannose was 1: 26 as calculated from BCA
protein assay of albumin and sulfuric/phenol assay (Dubois 1956) of mannose
residues. Incubation of
125
I-Man-Alb (5 µg/mL) with either CMII or SM and
subsequent fractionation on Sephacryl S-200 did not yield any difference in the
binding pattern of radiolabeled Man-albumin to medium factor(s) as monitored by
radioactivity and absorption. Absorption of fractions collected from
125
I-Man-Alb
incubated with CMII showed two large peaks (#26 and #34) containing the majority
of proteins found in CMII, but only #26 corresponded to the peak containing
125
I-
Man-Alb (Fig. 12A). On the other hand, absorption values of fractions collected
from
125
I-Man-Alb incubated with SM showed two shoulders (#26 and #30) and a
major peak (#34), but
125
I-Man-Alb-containing fraction coincided again with only
fraction #26 (Fig. 12B). The absorption/(radioactivity) of fraction #26 was
0.7/(204,000) and 1.6/(470,00) in SM and CMII, respectively, which seems to
indicate a proportionality between the amount of protein present in fraction #26 and
the amount of
125
I-Man-Alb binding to that fraction. Standard Man-Alb applied to
the same column type and volume (S-200/40ml/PBS eluent) showed a peak at
fraction #34 (data not shown). The shift in elution peak of
125
I-Man-Alb from
fraction #34 to fractions #26 suggests that common binding factors are present in
47
both SM and CMII. Such factors likely correspond to proteins with lectin binding
properties (i.e., collectin family) and may include surfactant protein A in addition to
complement factors known to be present in serum.
48

0
50000
100000
150000
200000
250000
300000
350000
400000
450000
500000
0 1020 304050
fraction
cpm
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Abs at 280nm
cpm
Abs

0
25000
50000
75000
100000
125000
150000
175000
200000
225000
0 1020 3040 50
fraction
cpm
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Abs at 280 nm
cpm
Abs

Figure 12. Size exclusion gel chromatography of Man-Alb.
125
I-Man-Alb (5 µg/mL)
was incubated for 30 min at 37ºC in either A) CMII or B) SM and applied to a
Sephacryl S-200 column (1 cm x 40cm), and eluted with PBS. Fractions (1 mL)
were collected and measured for both radioactivity (cpm ♦) and absorbance at 280
nm (Abs ▲).
A.
B.
49
II. Bronchoalveolar lavage fluid: Impact on insulin transport
1. Transport of insulin across RAECM-II and -I, MDCK and Caco-2
The apical-to-basolateral transport of acid-precipitable
125
I-insulin (as
determined by TCA precipitation) across RAECM-II when dosed in the retentate
obtained from 7.5x BALF using a Centricon device YM-10 was increased by 4-fold
as compared to that dosed in KRP (Fig. 13). Our data indicate that ~5% and ~10%
of total transported
125
I-insulin dosed in 7.5x BALF and KRP, respectively, was acid-
precipitable. Moreover, incubating
125
I-insulin in 7.5x BALF or KRP at 37
o
C for 1 h
in a water bath resulted in ~20% and 1% dissociation of total insulin, respectively
(data not shown). This suggests that the lower percent of acid-precipitable
125
I-
insulin observed when dosed in 7.5x BALF and transported across RAECM-II was
partly a result of protease activity present in BALF but which was absent in KRP.
All data regarding insulin transport hereafter represent the acid-precipitable
125
I-
insulin which is most relevant to our study.
When transport of insulin across RAECM-I and -II was compared, 7.5x
BALF enhanced transport of insulin across both monolayer types (3- and 4- fold,
respectively), indicating that the putative factor(s) may be involved in enhancing
insulin transport across both type I and type II cell monolayers (Fig. 13). To
determine whether the transport-enhancing activity of BALF factor(s) was also
efficacious in other cell monolayers, both MDCK and Caco-2 cell monolayers were
used. As shown in Figure 14, the insulin transport-enhancing effect of 7.5x BALF
was absent when tested across both MDCK and Caco-2 monolayers. Furthermore,
50
transport of insulin across RAECM-II when dosed in filtrate (in place of retentate)
was not significantly different from transport observed with KRP. Since
enhancement of insulin transport was most prominent across RAECM-II, we chose
to focus on this cell monolayer for characterization and identification of the factor(s)
using 7.5x BALF retentate.
51

Figure 13. Effect of BALF on insulin transport across RAECM-I and -II. Apical-to-
basolateral transport of A) total and B) intact insulin, when measured at apical donor
concentration of 5 µg/ml across RAECM-I and -II in either KRP or 7.5x BALF.
Data represent mean ± SD (n=3). * = significantly different (p<0.05) from all others.

0
1
2
3
4
5
KRP B ALF
transport (ng/well)
RAECM-II
RAECM-I
*
0
20
40
60
80
100
120
KRP B ALF
transport (ng/well)
RAECM-II
RAECM-I
A.
*
B.
52

0
5
10
15
20
25
KRP BALF retentate BALF filtrate
transport (ng/well)
RAECM-II
MDCK
Caco-2

Figure 14. Transport of insulin across RAECM-II, MDCK and Caco-2. Apical-to-
basolateral transport of intact insulin, when dosed apically at 5 µg/mL in either KRP
or 7.5x BALF retentate or 7.5x BALF filtrate across RAECM-II, MDCK and Caco-2
cell monolayers. Data represent mean ± SD (n=3). * = significantly different
(p<0.05) from all others.
*
53
2. Effects of putative factor(s) in BALF on transcytosis pathways
2.1. Paracellular versus transcellular markers
The effect of BALF on transport of paracellular markers (mannitol and
sodium fluorescein) and fluid-phase endocytosis marker (FITC-dextran, 70 kDa)
across RAECM-II, was also investigated (Table 4). The permeability of
3
H-mannitol
and sodium fluorescein dosed in KRP and 7.5x BALF showed similar fluxes across
RAECM-II, with no significant changes in TEER. The permeability of FITC-
dextran across RAECM-II was similar in magnitude, irrespective of dosing vehicle
used (i.e., KRP or 7.5x BALF).

Table 4. Effect of the 7.5x BALF on the paracellular and transcellular pathways in
RAECM-II.

Papp (cm/sec) Solute
KRP BALF
Sodium fluorescein (1 mg/mL) 5.8 ± 0.9 (x 10
-7
) 5.0 ± 0.89 (x 10
-7
)
3
H-Mannitol (1 µCi/mL) 7.35 ± 1.2 (x 10
-7
) 7.93 ± 1.34 (x 10
-7
)
FITC-dextran 70 kDa (0.5 mg/mL) 3.2 ± 0.16 (x 10
-9
) 2.6 ± 0.65 (x 10
-9
)
TEER ohm·cm
2
2,000 ± 304 (n=24) 2,150 ± 450 (n=24)
Values are mean ± SD, (n=4).
TEER measured in ohm·cm
2
of RAECM-II did not change when either KRP or
BALF was used as dosing solution (2,000 ± 304 (n=24) and 2,150 ± 450 (n=24),
respectively).

54
2.2. Effects of temperature and endocytosis inhibitor
Effect of BALF factor(s) on transport across RAECM-II was assessed by
dosing insulin apically in either KRP or 7.5x BALF at 37º or 4ºC. Transport of
insulin dosed in KRP decreased by ~40% at 4ºC compared to that at 37
o
C (Fig.
15A). The 3.5-fold of enhanced insulin transport observed with BALF at 37ºC was
decreased to the same level of transport observed with KRP when temperature was
lowered to 4ºC. Although 40% of this decrease was due to reduced paracellular
transport of insulin across RAECM-II at 4ºC, the additional 42% reduction suggests
an involvement of an alternative pathway such as transcytosis, likely afforded by the
transport-enhancing effect of BALF factor(s). To further corroborate this
observation, we used monensin, which is an endocytosis inhibitor. Monensin did not
affect insulin transport observed with KRP (Fig. 15B). However, in presence of
monensin, the enhanced insulin transport observed with 7.5x BALF decreased by
~90% to a level similar to that of baseline (i.e., insulin transport observed with KRP
in the absence of monensin). Given that monensin did not change TEER (2,000 ±
230 (n=10) and 2,250 ± 150 (n=10) ohm·cm
2
in the absence and presence of
monensin, respectively), the inhibition is likely due to interference with an
endocytosis-related pathway by the inhibitor. These data indicate that BALF
factor(s) enhance transcellular transport of insulin, but do not affect paracellular or
fluid phase endocytosis transport of solutes.

55

0
1
2
3
4
37 4
temperature
o
C
transport (ng/well)
KRP
BALF

0
1
2
3
4
(-)monensin (+)monensin
transport (ng/well)
KRP
BALF

Figure 15. Effects of temperature and monensin on enhanced insulin transport across
RAECM-II. Apical-to-basolateral transport of intact insulin, when dosed apically at
5 µg/mL in KRP or 7.5x BALF across RAECM-II A) at 37 and 4ºC after 1 h, and
B) with or without 60 µM monensin at 37ºC after 1 h. Data represent ± SD (n=4). *
= significantly different (p<0.05) compared to data observed with 7.5x BALF at 4°C;
and † = significantly different (p<0.05) compared to data observed with 7.5x BALF
with monensin.
†
*
A.
B.
56
3. Apical-to-basolateral versus basolateral-to-apical transport of insulin across
RAECM-II
Transport of intact insulin across RAECM-II, when dosed in BALF, was
enhanced by ~4-fold in the apical-to-basolateral (a-to-b) direction compared to that
in the basolateral-to-apical (b-to-a) direction (Table 5, p<0.05). In the presence of
BALF, acid-precipitable fractions of
125
I-insulin (i.e., % intact) in the basolateral and
apical receiver compartments were 7% and 20%, respectively. On the other hand,
transport of intact insulin across RAECM-II in the a-to-b direction, when dosed in
KRP, was less than 1-fold higher than that observed in the b-to-a direction (Table 5,
p<0.05). Moreover, the acid-precipitable fraction of
125
I-insulin (dosed in KRP) in
the basolateral and apical receiver compartments were 14% and 60%. In other
words, although some of the insulin was being degraded by proteolytic enzymes
present in BALF (as noted earlier), further degradation was occurring inherently by
peptidases, which are likely present on the apical membrane of RAECM-II (as
observed with KRP) and that additional degradation might be taking place within
cytosolic or vesicular domains of the alveolar epithelial cell monolayers as a result of
a transcellular pathway for insulin transport afforded by BALF factor(s).
Furthermore, the unidirectional enhancement of intact insulin transport from a-to-b
and not b-to-a suggests that the effect of BALF factor(s) specifically affects the
apical membranes of RAECM-II.
57
Table 5. Apical-to-basolateral (a-to-b) and basolateral-to-apical (b-to-a) transport of
insulin (10 µg/mL) dosed in Kreb’s Ringer’s phosphate (KRP) solution and 7.5x
BALF across RAECM-II.

KRP (n=3) BALF (n=3)
Total (ng/well) Intact (ng/well) Total (ng/well) Intact (ng/well)
a-to-b 17.43 ±5.94 2.36 ±0.17 171.07 ±9.87 12.54 ±1.23
b-to-a 2.48 ±0.29

1.42 ± 0.18

12.54 ±1.10

2.62 ±0.63
Values are mean ± SD, (n=3).

4. Characterization of insulin transport-enhancing factor(s) in BALF
4.1. Molecular sizing
The retentates of the 7.5x BALF using Centricon devices YM-10, -30, or -50
were used as dosing vehicles to examine the effect of size-fractionated BALF on
insulin transport across RAECM-II. As shown in Table 6, insulin transport-
enhancing effect was highest in the retentate obtained from the Centricon device
YM-10 and the transport was decreased to 56% when dosed in retentates obtained
from Centricon devices YM-30 and -50. A 12% SDS-PAGE revealed that similar
bands present (Fig. 16) in all retentates, but that the intensity/level of protein
expression was greater in the experiments using retentate from the 10 kDa
ultrafiltration compared to that of the 30 and 50 kDa cutoff.

58
4.2. Temperature and trypsin treatment
In order to examine the nature of the factor(s) contributing to the
enhancement of insulin transport across RAECM-II, 7.5x BALF concentrate
obtained from the Centricon device YM-10 was heated at 80ºC for 15 min or
digested with trypsin and quenched with soybean trypsin inhibitor (Table 6).
Dosing
125
I-insulin in the heat-denatured or the trypsin-digested BALF concentrates,
decreased its transport to 4% and 20%, respectively, of that observed with untreated
the BALF concentrate (p<0.05).

Table 6. Effect of different treatments of 7.5x BALF retentate on enhancing insulin
transport across RAECM-II.

Dosing vehicle %
125
I-insulin transport as compared to
BALF retentate using 10 kDa cutoff
a

BALF retentate from 10 kDa 100
BALF
b
retentate from 30 kDa 56 ± 8.32
BALF
c
retentate from 50 kDa 56 ± 6.31

KRP

20 ± 0.63
BALF
d
retentate treated at 80ºC for 15 min 4 ± 0.35
BALF
e
retentate treated with 10 µM trypsin
and 20 µM soybean trypsin inhibitor
20 ± 2.08
Values are mean ± SD, (n=3).
a
BALF was concentrated 7.5x in a Centricon device YM-10 and retentate was used as dosing vehicle
for
125
I-insulin transport across RAECM-II.
b
BALF was concentrated 7.5x in a Centricon device YM-30 and used as dosing vehicle for
125
I-
insulin transport across RAECM-II.
c
BALF was concentrated 7.5-fold in a Centricon device YM-50 and used as dosing vehicle for
125
I-
insulin transport across RAECM-II.

d
7.5x BALF retentate from a Centricon device YM-10 was denatured at 80ºC for 15 min.
e
7.5x BALF retentate from a Centricon device YM-10 was digested with 10 µM trypsin and
quenched with 20 µM soybean trypsin inhibitor
59

Figure 16. Polyacrylamide gel electrophoresis of BALF retentate. BALF was
concentrated 7.5 fold using Centricon devices with different molecular weight cut
offs and aliquots from the retentates were examined by 12% SDS-PAGE. Lane 1,
BALF retentate from YM-50. Lane 2, BALF retentate from YM-10. Lane 3, BALF
retentate from YM-30. Lane 4, rainbow marker. Staining was with Coomassie Blue.
1 2 3 4
220 K
97 K
66 K

45 K

30 K

20 K

14.3 K
60
5. Surfactant protein A (SP-A) as a “putative enhancing factor in both CMII and
BALF
5.1. Western blot of SP-A in BALF
One possible explanation for enhancement of insulin transport observed
specifically across RAECM-II when using CMII or BALF as dosing vehicle, maybe
due to the inherent secretion of surfactant proteins by RAECM-II in cell culture or
by type-II pneumocytes in the alveolar epithelium. The complexity of CMII
prevented us from purifying and detecting SP-A directly; however, the presence of
SP-A in CMII was an assumption based on the maintenance of lamellar bodies in
RAECM-II and the increased expression of SP-A mRNA when KGF-supplemented
culture medium was used (Borok et al. 1998). Analysis of SP-A in BALF by
Western blotting, was possible after concentrating crude rat lavage fluid by ~100-
fold (Fig. 17). It was difficult to estimate the amount of SP-A from our experiment
but it is known that normal rat lung lavage may contain approximately 8-10µg of SP-
A/lung (Haagsman et al. 1987).

36 kDa
28 kDa
Figure 17. Western blot of SP-A present in rat BALF. BALF was concentrated 10 fold
using a Centricon device YM-10. 1 mL aliquot from 10x BALF was precipitated using
trichloroacetic acid. The supernatant was removed and the precipitate was dried with
ethanol and resuspended with 10 µL of loading buffer. Lane 1, rat SP-A; Lane 2, SP-A
from rat BALF. The proteins were separated by 12% sodium dodecyl sulfate-
polyacrylamide gel electrophoresis under reducing conditions and immunoblotted with
Whitsett’s anti-SP-A antibody. Molecular masses were determined by comparison with
known molecular markers.
61
5.2. Effect of SP-A on insulin transport across RAECM-II
On day five of cell culture, different concentrations of rat SP-A (1, 3, and 10
µg/mL, a generous gift from Dr. McCormack) and
125
I-insulin dosed in KRP
supplemented with 0.1% albumin serum showed a slight but insignificant increase of
insulin transport when dosed in presence of SP-A compared to that dosed in control
buffer. The enhancement of insulin transport in presence of SP-A was low in
comparison to that seen with BALF (Figs. 18A and 18B). There are several
explanations to this observation. First, receptors for SP-A in our RAECM-II model
have never been investigated and the only supporting data are Northern blots
obtained of SP-A mRNA in rat alveolar cells cultured on Transwells and
supplemented with MDSF (Borok et al. 1998). Studies that have identified high-
affinity receptors for SP-A, have been restricted to either freshly isolated rat type II
pneumocytes or those cultured for not more than 24 h (Kuroki et al. 1988), (Stevens
et al. 2001). Second, both CMII and BALF are complex mixtures of proteins, lipids,
and other factors which may act synergistically resulting in an enhancement of
insulin transport across RAECM-II. Addition of SP-A antibody decreased
significantly the enhanced insulin transport by ~33% (Fig. 18B), but without a
control human antibody it was difficult to determine whether the decrease was due to
blocking of SP-A or non-specific binding.

62

0
1
2
3
4
SP-A 1µg/ml SP-A 3 µg/ml SP-A 10 µg/ml Alb 0.1%
transport (ng/well)

0
2
4
6
8
BALF BALF+ anti SP-A KRP
transport (ng/well)

Figure 18. Effect of human SP-A and anti-SP-A antibody on insulin transport across
RAECM-II. Apical-to-basolateral transport of intact insulin, when measured at
apical donor concentration of 4 µg/ml across RAECM-II in A) increasing
concentration of rat SP-A, and B) BALF with or without anti SP-A antibody. Data
represent mean ± SD (n=3). * = significantly different (p<0.05) from all others.
*
A.
B.
63
6. Purification of BALF proteins and their effect on insulin transport

6.1. Using ultracentrifugation and lyophilization
When rat lungs were instilled with large amounts of KRP (30 mL/rat), the
resultant BALF contained proteins in very dilute concentrations (0.2 mg/mL).
Concentrating BALF by ~7.5-fold in a Centricon YM-10 device increased the
protein concentration to ~0.5-1 mg/mL. For partial purification purposes, further
concentration and decrease of BALF volume was necessary. Lyophilization is
considered an efficient method for concentration as long as the activity of the protein
of interest is not compromised. BALF retentate, obtained by centrifugation, was
lyophilized and reconstituted in 2 mL water and desalted using a Sephadex G-50
column (Fig. 19). Insulin transport coincided with peak fractions of the desalted,
lyophilized BALF indicating that that lyophilization did not abolish BALF factor(s)
activity. For further purification, peak fractions (15-18), which were pooled and
concentrated to 2 mL (Fig. 20A), were applied to a Sephacryl S-200 column. The
elution profile (Fig. 20B) showed a single peak (c) with a small shoulder (b).
Fractions: b (17-20), c (21-24), and d (25-28) as indicated in (Fig. 20B), were
pooled, concentrated, and labeled as fractions: B, C, and D and tested for their effect
on insulin transport across RAECM-II (Fig. 21A). It was observed that the amount
of apical-to-basolateral transport of insulin dosed in fraction B (A
280
0.3) was
approximately 2-fold higher compared to that dosed in fraction C (A
280
0.7) (Fig.
21A) and 4-fold higher compared to that dosed in PBS. Fraction C also yielded a
slight but not significant enhancement of insulin transport as compared to transport
64
observed with PBS (3.4 ng/well and 1.43 ng/well), respectively). Analysis of
fractions B, C and D using a 7.5% SDS-PAGE and Coomassie staining showed that
several bands resembling high molecular weight proteins were expressed
predominantly in fraction B (Fig. 21B) suggesting the involvement of such moieties
in increasing the amount of insulin transport across RAECM-II.

Figure 19. Transport of insulin across RAECM-II dosed in lyophilized BALF. Rat
lavage was concentrated 7.5-fold in a Centricon YM-10 device and lyophilized. The
lyophilized BALF powder was reconstituted with 2 mL water and desalted using
Sephadex G-50 column (1 cm x 40 cm), PBS (pH 7.4). Fractions (1 mL) were
measured at absorbance 280 nm and used for dosing solution to study apical-to-
basolateral transport of insulin (5µg/mL) across RAECM-II.

0
1
2
3
4
5
6
7
0 1020 304050
fracti on
transport (ng/well)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Abs at 280nm
transport
absorbance
65

Figure 20. Salting out and purification of BALF. Rat lavage was concentrated 7.5-
fold in a Centricon YM-10 device. A) 7.5x BALF was lyophilized, reconstituted
and desalted using Sephadex G-50 column (1 cm x 40 cm). Fractions (1 mL) were
measured at absorbance 280 nm. B) Fractions (15-18) from G-50 gel
chromatography were pooled and concentrated to ~2 mL and purified using
Sephacryl S-200 column (1 cm x 40 cm). Fractions (1 mL) were measured at 280
nm. PBS (pH 7.4) was the eluting buffer used for G-50 and S-200 gel
chromatography.

B
0
0.05
0.1
0.15
0.2
0.25
0 1020 3040
fraction
Abs at 280 nm
b
c
d
B.
0
0.2
0.4
0.6
0.8
1
0 10203040
fraction
Abs at 280 nm
A.
66

0
100
200
300
400
500
600
700
BC D PBS
Pooled fractions
transport % of control
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Abs at 280 nm
transport
Abs

1 2 3 4
200
116

97

66

45

31
Figure 21. Relative transport activity of S-200 purified BALF fractions. After
purifying lyophilized 7.5x BALF using Sephacryl S-200 column, fractions
denoted as b, c, and d from Fig. 20, were pooled, concentrated and labeled as
(B, C, and D). A) Absorbance profile ( ♦) of fractions (B, C and D) at 280 nm
and transport (solid bars) of intact insulin, when measured at apical donor
concentration of 5 µg/ml dosed in fractions B, C, D, and control buffer (PBS)
across RAECM-II. Data represent mean ± SD (n=4). * = significantly
different (p<0.05) from all others. B) 7.5% SDS-PAGE representing: Lane 1,
fraction B; Lane 2, fraction C; Lane 3, fraction D; Lane 4, broad range
molecular weight marker. Staining was with Coomassie Brilliant Blue.
A.
B.
*
67
6.2. Using tangential fluid filtration
The use of ultracentrifugation as means of concentrating and purifying BALF
was inconvenient for protein identification using LC-MS/MS. We therefore used a
tangential fluid filtration (TFF) system as an alternative method to concentrate
approximately 500 mL of BALF to a final volume 25 mL (i.e., 20x BALF), which
contained ~2.5 mg/mL of protein. This 20x BALF retained insulin transport-
enhancing activity (data not shown). Lyophilization, which could help in storing this
concentrated form of BALF, led to an abolishment of BALF factor(s) enhancing
activity of insulin transport across RAEMC-II (data not shown). Purification of 20x
BALF resulted in two major peaks (Fig. 22A) fraction #22 (high molecular weight
fraction) and fraction #29 (albumin-rich fraction). Figure 22A shows that the
amount of insulin transported when dosed in fraction #22 or #29 (~ 200 ug/mL of
protein) was 6.7 ng/well or 3.5 ng/well, respectively, indicating that the fraction
consisting of the high molecular weight proteins, as observed previously, maintained
the transport-enhancing activity even with the use of TFF as a method for
concentrating large volumes of BALF. Fraction #22 which was loaded onto a 7.5%
SDS-PAGE, and analyzed with Coomassie staining, revealed three prominent high
molecular weight protein bands at molecular weight of

~200 kDa (Fig. 22B). In-gel
digestion and LC-MS/MS of the three bands suggested the proteins may be: alpha-1-
inhibitor III, murinoglobulin gamma 2, and pregnancy-zone protein (Table 7), all of
which belong to the macroglobulin family.
68

0
1
2
3
4
5
6
7
8
18
20
22
24
26
28
30
32
34
36
fraction #
transport (ng/well)
0
0.1
0.2
0.3
0.4
0.5
0.6
Abs at 280 nm
transport
Abs

Figure 22. Elution profile and SDS-PAGE of BALF proteins. BALF was
concentrated 20-fold using TFF system, and purified by applying to a Sephacryl S-
200 column (1 cm x 40 cm), PBS (pH 7.4)). A) Absorbance profile ( ♦) at 280 nm of
fractions (1 mL) from purified 20x BALF and the corresponding apical-to-
basolateral transport (closed bar) of
125
I-insulin dosed apically at 10 µg/mL in
fractions #22 and #29 across RAECM-II. Data represent mean ± SD (n=3). * =
significantly different (p<0.05) from the enhanced insulin transport observed with
fraction #29. B) BALF protein detected by 7.5% SDS-PAGE. Lane 1: fraction #22
of purified 20x BALF. Lane 2: rainbow molecular weight markers (myosin,
220,000; phosphorylase b, 97,000; bovine serum albumin 66,000).
*
A.
B.
69
Table 7. Analysis and identification of proteins, obtained from SDS-PAGE gel
bands of rat BALF (fraction #22), using LC-MS/MS and Mascot Search Engine.

a
Score is the summation of all observed unique peptides, which is based on individual peptide scores
(ion scores) >36 and that are indicative of identity and extensive homology.
b
The % sequence coverage is determined by the percentage of amino acids matched to the identified
protein based on observation of tryptic peptides with high confidence match (p<0.05) as per Mascot
scoring.

The following data are detailed scoring obtained from the Mascot Search
Engine for the analyzed proteins: a) alpha-1-inhibitor III and fingerprint of peptide
#12, b) murinoglobulin 2, and c) pregnancy-zone protein.

a) gi|83816939 Mass: 165038 Total score: 462 Peptides matched: 12
alpha-1-inhibitor III [Rattus norvegicus]

Query Observed Mr(expt) Mr(calc) Delta Miss Score Rank Peptide
1 473.49 944.96 943.53 1.43 0 45 1 HGIPFFVK
2 549.22 1096.42 1095.65 0.77 0 41 1 VLIVEPEGIK
4 662.53 1323.04 1321.72 1.32 0 40 1 GDPIPNEQVLIK
5 724.33 1446.65 1445.83 0.82 0 54 1 MLSGFIPLKPTVK
6 770.50 1538.98 1538.76 0.22 0 (38) 1 TPSVTVQSSGSFSQK
7 770.90 1539.78 1538.76 1.02 0 54 1 TPSVTVQSSGSFSQK
9 796.78 1591.54 1590.81 0.73 0 58 1 AHFSVMGDILSSAIK
10 807.93 1613.84 1612.79 1.05 0 44 1 ALMAYAFALAGNQEK
12 839.84 1677.66 1676.78 0.88 0 48 1 ISLCHGNPTFSSETK
13 560.54 1678.59 1676.78 1.81 0 (36) 1 ISLCHGNPTFSSETK
14 981.22 1960.43 1958.93 1.50 0 41 1 LPSSEEEESLDINIEGAK
15 710.00 2126.98 2125.04 1.94 0 47 1 VHLSFSPSQSLPASQTH

Protein name Database
accession #
Mass Score
a

(p<0.05)

% Sequence
coverage
b

alpha-1 inhibitor III
precursor (Rattus
norvegicus)
gi 34858400 164986 623 16
murinoglobulin 2
(Rattus norvegicus)
gi 62647940 162752 677 14
pregnancy-zone protein
(Rattus norvegicus)
gi 21955142 168422 330 7
70

MS/MS Fragmentation of ISLCHGNPTFSSETK Found in gi|83816939, alpha-1-inhibitor III
[Rattus norvegicus]
Monoisotopic mass of neutral peptide (Mr): 1676.78
Ions Score: 48 Matches (Bold Red): 36/144 fragment ions using 76 most intense peaks

b) gi|50657404 Mass: 162855 Total score: 523 Peptides matched: 13
murinoglobulin 2 [Rattus norvegicus]

Query Observed Mr(expt) Mr(calc) Delta Miss Score Rank Peptide
1 473.33 944.65 943.53 1.12 0 (36) 1 HGIPFFVK
2 473.67 945.32 943.53 1.79 0 48 1 HGIPFFVK
3 645.10 1288.19 1287.66 0.53 0 42 1 VTASPQSLCGLR
4 686.27 1370.52 1369.74 0.78 1 38 1 VKTVPLTCNNPK
5 724.40 1446.79 1445.83 0.96 0 51 1 MLSGFIPLKPTVK
6 770.98 1539.94 1538.76 1.18 0 64 1 TPSVTVQSSGSFSQK
7 788.83 1575.65 1574.81 0.84 0 57 1 AHFSVMGDILSSAIK
9 796.87 1591.73 1590.81 0.92 0 (46) 1 AHFSVMGDILSSAIK
10 796.93 1591.85 1590.81 1.04 0 (56) 1 AHFSVMGDILSSAIK
11 866.77 1731.52 1730.87 0.65 0 77 1 FSIDTSSISGYSLNIK
12 957.86 1913.70 1912.95 0.75 0 36 1 YMVLVPSQLYTETPEK
13 980.81 1959.61 1958.93 0.68 0 84 1 LPSSEEEESLDINIEGAK
14 709.93 2126.77 2125.04 1.73 0 40 1 VHLSFSPSQSLPASQTHM

c) gi|21955142 Mass: 168422 Total score: 253 Peptides matched: 5
pregnancy-zone protein [Rattus norvegicus]

Query Observed Mr(expt) Mr(calc) Delta Miss Score Rank Peptide
1 545.74 1089.46 1087.55 1.91 0 54 1 VPDTITEWK
2 676.74 1351.47 1349.75 1.73 0 56 1 ALLAYAFALAGNR
3 750.76 1499.51 1497.75 1.76 0 80 1 VSGSGCVYLQTSLK
6 790.08 1578.15 1577.77 0.39 0 34 1 THITNAFNWLSMK
9 896.88 1791.74 1790.91 0.83 0 36 1 YNILPEAEGEAPFTLK

71
7. Effect of protein concentration of BALF fraction #22 on insulin transport
across RAECM-II
Table 8 shows a concentration dependent increase of enhanced insulin
transport with increasing amounts of protein in BALF fraction #22. The absence of
transport saturation, as would have been expected due to a receptor-mediated
enhancement effect of BALF factor(s), is complicated by the presence of more than
one protein in the sample. All three macroglobulins (detected by LC-MS/MS) are
taken up into the cell through a membrane bound receptor with varying kinetics and
processing and the impact of each protein must be studied separately to establish a
valid conclusion. However, it is evident, from the amount of insulin transported in
20x BALF versus that in BALF fraction #22 (26.16 ng/well vs 15.5 ng/well,
respectively) that a combination of factor(s) is involved in enhancing insulin
transport across RAECM-II, however, our method of purification underscored the
involvement of macroglobulins as major contributors of this enhancement.

Table 8. Effect of protein concentration in BALF fraction #22 on enhancing
transport of insulin (dosed apically at 20 µg/ml) across RAECM-II.

% of
fraction # 22
Abs at 280
nm
Protein
(ug/mL)
Intact
(ng/well)
% intact
100% 0.447 293 15.6 ±2.48 5%
80% 0.332 213 12.4 ±0.22 6%
50% 0.204 120 8.05 ±2.98 8%
15% 0.06 10 2.59 ±2.59 5.6%
20x BALF 1.484 26.16 ±3.51 4%
PBS 2.02 ±0.06 11%

72
8. Effect of BALF fraction #22 on insulin transport across RAECM-I and -II
We had previously shown (Fig. 13) that 7.5x BALF enhanced transport of
insulin across RAECM-I and RAECM-II by 3- and 4-fold, respectively as compared
to KRP. When using fraction #22 from purified BALF (Refer to Materials and
Methods, Section 4.4) the insulin transport-enhancing activity observed previously
with BALF across RAECM-I was abolished (Fig. 23B). For RAECM-II, however,
fraction #22 enhanced transport of intact insulin ~2-fold from that observed with
PBS in contrast to ~4-fold, which was observed previously when 7.5x BALF was
used as dosing solution (Fig. 13). Similar to the data observed with 7.5x BALF,
around 5% of total transported insulin across RAECM-II was acid-precipitable (i.e.,
intact) when dosed in fraction #22 from purified BALF. Taken together, these data
provide supporting evidence that a number of factor(s) in BALF are likely involved
in enhancing insulin transport across RAECM with each depending on its
concentration, cell association and overall interaction(s).
73

0
10
20
30
40
50
PBS BALF (#22)
transport (ng/well)
RAECM-II
RAECM-I

0.00
0.50
1.00
1.50
2.00
2.50
3.00
PBS BALF (#21)
transport (ng/well)
RAECM-II
RAECM-I

Figure 23. Effect of purified BALF on insulin transport across RAECM-I and -II.
Apical-to-basolateral transport of A) total and B) intact insulin, when measured at
apical donor concentration of 10 µg/ml across RAECM-I and -II in either fraction
#22 of the BALF or saline phosphate buffer (PBS). Data represent mean ± SD
(n=3). * = significantly different (p<0.05) from all others.
*
*
A.
B.
74
9. Comparison of rat plasma to BALF
9.1. Purification of rat plasma
Based on the LC-MS/MS analysis of the purified BALF fraction #21 (i.e.
with high molecular weight proteins), the results indicated that alpha-1-inhibitor III
precursor ( α
1
I
3
) was the most abundant protein. Since α
1
I
3
is not available
commercially, we decided to use rat plasma as an alternative source of α
1
I
3
. The
concentration of alpha-1-inhibitor III in normal rat plasma is about 30 µM (Gliemann
and Sottrup-Jensen 1987). Rat plasma was obtained as mentioned in Materials and
Methods Section 5. Figure 24 shows that partial purification of rat plasma using S-
200 yielded an absorption profile pattern comparable to that of BALF. However, the
concentration of proteins in rat plasma was several folds higher than BALF as
depicted by the absorbance value and SDS-PAGE (Figs. 24 and 25). It was,
therefore, necessary to normalize protein concentration of the rat plasma fractions to
that of BALF. A 1:100 dilution of the collected fractions from the rat plasma sample
gave absorption values that were similar to the fractions obtained from BALF after
S-200 partial purification. When insulin was dosed in fractions obtained from BALF
purification or their counterpart from rat plasma purification (after 1:100 dilution),
the latter showed no enhancement of insulin transport in any fraction across
RAECM-II (data not shown). This observation could have been either due to
masking of α
1
I
3
by other proteins in the rat plasma or simply due to extensive
dilution of α
1
I
3
.

75

0
0.5
1
1.5
2
10 15 20 25 30 35 40
fracti on
Abs at 280 nm

Figure 24. Purification of rat plasma. Rat plasma was applied to S-200 column (1
cm x 40 cm) and eluted with PBS (pH 7.4). Fractions (1 mL) were collected and
diluted 1:5 with PBS and absorbance was measured at 280 nm.

Figure 25. SDS-PAGE of purified rat plasma and BALF fractions. Lane 1, albumin
(3 µg); Lane 2, broad range marker; Lane 3, desalted BALF (3 µg); Lane 4, rat
plasma fraction #22 (1:5); Lane 5, rat plasma fraction #24 (1:5); and Lane 6, rat
plasma fraction #29 (1:5).
1 2 3 4 5 6 1 2 3 4 5 6
76
9.2. Effect of rat plasma on transport of insulin
We sought to use various dilutions of rat plasma fractions and compare their
effect with that of fractions from BALF. Figure 26 shows that rat plasma (1:10)
fraction 24 (representing high molecular weight proteins) resulted in higher transport
of intact insulin across RAECM-II as compared to rat plasma (1:10) fraction #29.
This is in accord to what has been observed with insulin transport when dosed in 20x
BALF fractions #22 and #29 (Fig. 22A). Furthermore, the transport of insulin was
higher in fraction #24 from rat plasma diluted at 1:10 as compared to that diluted at
1:25. That is also in concurrence to what we had observed regarding a dose-
dependent effect of BALF (Table 8). However, what remains ambiguous is that 7.5x
BALF fractions 22-24, which contained a lower amount of protein (Abs 0.045)
compared to the amount in rat plasma fraction #24 (1:25, Abs 0.16), showed higher
transport of insulin across RAECM-II. This could be due to oversaturation of the
receptor involved in endocytosis of α
1
I
3
or otherwise the presence of high and low
affinity receptors that depend on ligand concentration or the presence of several
factors such α
1
I
3
and pregnancy zone protein that may share common receptors for
endocytosis. Another explanation could be because the α
1
I
3
present in BALF is the
activated form of the macroglobulin in contrast to the native form usually found in
the plasma. We therefore decided to use a homologue of α
1
I
3
, human alpha-2
macroglobulin, which is commercially available and which has been studied more
extensively in literature.
77

0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
plasma 1:10 plasma 1:25 BALF buffer
transport (ng/well)
fraction (24)
fraction (29+30)
fraction (22-24)
fraction (29+31)
buffer

Figure 26. Comparison of the effect of rat plasma and BALF on insulin transport
across RAECM-II. Rat plasma and 7.5x BALF were both partially purified using
Sephacryl S-200 (1 cm x 40 cm) using PBS (pH 7.4) as eluting buffer. Chosen
fractions from rat plasma purification were diluted and used as dosing vehicle.
Apical-to-basolateral transport of intact insulin, when measured at apical donor
concentration of 10µg/mL across RAECM-II. Data represent mean ± SD (n=3). * =
significantly different (p<0.05) from all others. ‡ = significantly different (p<0.05)
from all other rat plasma fractions and control buffer. Values in bold indicate
absorbance of each fraction measured at 280 nm.

‡
0.55
0.16
0.54 0.215
0.13
0.045
*
78
III. Macroglobulin family: Impact on insulin transport

1. Inhibition of BALF-mediated enhancement using anti-megalin antibody
1.1. Western blot of megalin
Type II pneumocytes are known to express megalin or gp330 receptor
(Moestrup 1994). A western blot of lysed RAECM-II grown on Transwell plates for
six days revealed a band which was reactive with mouse gp330 monoclonal antibody
1H2 (Fig. 27). This 1H2 antibody was reactive with the positive control gp330
obtained from mouse kidney homogenate, an organ known for its high gp330
expression levels.

Figure 27. Western blot of megalin from RAECM-II and mouse kidney
homogenate. RAECM-II from 12 inserts of 12-well Transwell plate were lysed with
RIPA buffer supplemented with proteinase inhibitor cocktail. Mouse kidneys were
dissected and homogenized using RIPA buffer supplemented with proteinase
inhibitor cocktail. Lane 1, rainbow marker; Lane 2, RAECM-II megalin; Lane 3,
mouse kidney megalin. The proteins were separated by 7.5% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis under non-reducing conditions and
immunoblotted with 1H2 anti-megalin antibody.
1 2 3
79
1.2. Competition with anti-megalin antibody
Subsequently, we wanted to determine whether gp330 was involved in the
enhancement of insulin transport mediated by factor(s) present in S-200 purified 7.5x
BALF fractions (17-20) also designated as BALF fraction B (refer to Sec. 11.1)
hereafter. As depicted in Fig. 28, incubation of 1H2 with the BALF fraction B
blocked approximately 40% of the apical-to-basolateral enhanced insulin transport
across RAECM-II as compared to BALF fraction B alone (5.9 ng/well vs 9.8
ng/well, respectively). There was no significant decrease of insulin transport when
co-incubated with mouse IgG, as a control. The inhibition of BALF-mediated
enhancement of insulin transport across RAECM-II was incomplete as evident by the
significantly lower amount of insulin transport when dosed in PBS (1.78 ng/well).
This suggests that other receptors for macroglobulins, such as LRP, may be a more
prominent pathway of BALF-mediated insulin transcytosis across RAECM-II.
80

0
2
4
6
8
10
12
BALF (fraction B)
BALF (fraction B)+IgG
BALF (fraction B)+1H2
PBS
transport (ng/well)

Figure 28. Inhibitory effects of anti-megalin antibody (1H2) on the enhanced insulin
transport RAECM-II.
125
I-insulin (10 µg/mL) was dosed apically in S-200 purified
7.5x BALF fraction B alone or in the presence of either mouse IgG (50 µg/mL) or
1H2 (50 µg/mL) or PBS. Apical-to-basolateral transport of intact insulin was
measured across RAECM-II. Data represent mean ± SD (n=3). * = significantly
different (p<0.05) from transport with BALF (fraction B) and PBS.
*
81
2. Comparative studies of
125
I-insulin transport in BALF versus alpha-2
macroglobulin
2.1. Effect of alpha-2 macroglobulin on insulin transport
Given that 1H2 antibody was unable to completely inhibit the enhanced
transport of insulin offered by BALF fraction B, we investigated the role of LDL
receptor-related protein (LRP) as an alternative receptor that could be involved in the
endocytosis of the insulin-protein complex in RAECM-II. Alpha-2 macroglobulin
( α
2
M) is homologous to α
1
I
3
and is available commercially. Both proteins belong to
the macroglobulin family and function as universal proteinase inhibitors via a similar
mechanism (Gliemann and Sottrup-Jensen 1987). They are also taken up by the LRP
receptor. However, only the activated form of α
2
M (i.e., one which has trapped a
proteinase and undergone a conformational change) binds to its cognate receptor and
is taken up by the cell. Figure 29 shows that when insulin is incubated with α
2
M in
the presence of trypsin, an enhancement of insulin transport similar to that with
BALF is observed across RAECM-II. This increase, however, does not occur when
insulin is incubated with α
2
M alone or with preformed α
2
M-trypsin complex.

2.2. Mechanism of alpha-2 macroglobulin and insulin interaction
To explain the mechanism of interaction between insulin and α
2
M in the
absence or presence of trypsin, the two mixtures were incubated at 37°C for 15 min
and applied to Sephacryl S-300 for purification. The absorption profile of the
mixture plotted against radioactivity, revealed two overlapping peaks for absorbance
82
and radioactivity due to the α
2
M·
125
I-insulin complex. Binding of insulin to α
2
M was
approximately 40-fold higher in the presence of trypsin as compared to that in the
absence of trypsin (Fig.30). The percentage of insulin bound to α
2
M in the presence
of trypsin was about 10% of total insulin used.

0
2
4
6
8
10
12
Mac
Mac/try
Mac/ins/try
BALF
BALF+ Mac
buffer
transport (ng/well)

Figure 29. Comparison of insulin transport in BALF versus alpha 2-macroglobulin.
125
I-insulin was incubated with α
2
M (2.8 nM), preformed α
2
M-trypsin complex, α
2
M
in the presence of trypsin (10 µM), BALF or BALF in combination with α
2
M.
Apical-to-basolateral transport of intact insulin, when measured at apical donor
concentration of 10 µg/mL across RAECM-II. Data represent mean ± SD (n=3). * =
significantly different (p<0.05) from all others. ** = significantly different (p<0.01)
compared to treatments with α
2
M and α
2
M-trypsin complex.

*
**
83

0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25
fraction
Abs at 280
0
200
400
600
800
1000
1200
1400
cpm
Abs
cpm

0
0.01
0.02
0.03
0.04
0.05
0 5 10 15 20 25 30
fracti on
Abs at 280 nm
0
5000
10000
15000
20000
25000
30000
35000
40000
cpm
Abs
cpm

Figure 30. Elution profile of α
2
M and
125
I-insulin using S-300 gel chromatography.
α
2
M (300 nM) and
125
I-insulin (20 µg/mL) were incubated at 37°C for 15 min. in A)
absence and B) presence of 10 µM trypsin. The mixture was applied to Sephacryl S-
300 column (1 cm x 40 cm) and eluted with PBS, pH 7.4. Fractions (1mL) were
analyzed for absorbance at 280nm ( ♦) and for radioactivity ( ■).
A.
B.
84
3. Inhibition of BALF-mediated enhancement of insulin using activated alpha-2
macroglobulin
3.1. Receptor binding of alpha-2 macroglobulin with RAECM-II
α
2
M has been shown to bind to LRP, which is present on various interstitial
cells (macrophages and fibroblasts), epithelial and specialized cells such as
hepatocytes, cells of the choroid plexus, ciliary body, neurons, and alveolar epithelial
cells (Zheng et al. 1994) (Moestrup et al. 1992). To determine whether
internalization of activated α
2
M (2.8 nM) bound to radiolabeled insulin was receptor
mediated, the complex was incubated with increasing concentrations of activated
α
2
M bound to unlabeled insulin. As depicted in Fig. 31, transport of intact insulin
decreased by about 60% in the presence of 50 nM activate α
2
M·insulin complex and
to baseline transport in the presence of 400 nM activated α
2
M·insulin complex. No
significant decrease was observed in the presence of equimolar concentrations of
albumin denoting that the inhibition was not due to non-specific binding.

85
0
20
40
60
80
100
120
140
1 10 100 1000
α2M•insulin complex (nM)
% of control
Ins-Mac-tryp
Alb

Figure 31. Transport of α
2
M·
125
I-insulin complex across RAECM-II. α
2
M (2.8nM)
was activated in presence of
125
I-insulin (10 µg/mL) and trypsin at 37 ºC for 20 min.
In parallel increasing concentrations of α
2
M (8.4-400 nM) were activated in presence
of unlabeled insulin and trypsin in the same manner described above. Increasing
equimolar concentrations of albumin were used as control. Apical-to-basolateral
transport of α
2
M·
125
I-insulin complex was measured as % control across RAECM-II
in presence of increasing concentrations of α
2
M·insulin complex ( ■), and albumin
( ▲). Data represent mean ± SD (n=3).

86
3.2. Competition with activated alpha-2 macroglobulin complex
The transport of insulin across RAECM-II when dosed in BALF fraction B,
was decreased approximately 60% in the presence of excess activated α
2
M·insulin
complex (200 nM) as compared to that in absence of the activated α
2
M·insulin
complex (Fig. 32). Based on the presumption that the major factor in BALF fraction
B is α
1
I
3
and the knowledge that α
2
M is a ligand of LRP, may suggest that α
1
I
3
and
α
2
M share a common receptor for endocytosis in RAECM-II.

0
2
4
6
8
10
12
14
BALF (fraction B)
PBS
transport (ng/well)

Figure 32. Inhibitory effect of activated α
2
M on the enhanced insulin transport in
BALF across RAECM-II. Apical-to-basolateral transport of intact insulin, when
measured at apical donor concentration of 10 µg/mL and dosed in BALF fraction B
alone or in the presence of excess activated α
2
M·insulin complex (200nM), or in PBS
across RAECM-II. Data represent mean ± SD (n=3). * = significantly different
(p<0.05) from all others.

*
87
Chapter Five: Discussion
1. Effect of CMII factor(s) on insulin transport
Our findings demonstrate that conditioned medium (CMII), obtained from
cultured type II cell-like alveolar epithelial monolayers (RAECM-II), contains (a)
protein factor(s) which can increase transport of insulin across the alveolar epithelial
barrier. This conclusion is based on the following observations: 1) neither newborn
bovine serum components nor keratinocyte growth factor (KGF) per se, both used in
culture medium, produced increases in trans-alveolar epithelial transport of insulin;
2) when centrifuged using a 50 kDa molecular weight cut-off Centricon filter,
enhancement of transport was only in the retentate of the CMII; 3) the factor(s) in
CMII was heat labile and precipitated in 50% ammonium sulphate; 4) conditioned
medium from type I cell-like alveolar epithelial cell monolayers (RAECM-I), CMI,
did not increase peptide transport, suggesting the absence in CMI of (a) factor(s)
secreted predominantly by RAECM-II; and 5) the increase of insulin transport
enhanced by CMII was observed only across RAECM-II, but not RAECM-I.
Absorption enhancers in protein/peptide delivery have been studied
extensively (Patton et al. 2004) (Hussain et al. 2004). Some of the enhancers act by
loosening tight junctions which in turn allows increased paracellular transport
(ethanol, chitosan, bile salts) (Ma et al. 1999) (Machida et al. 2000). However, CMII
did not change permeability of mannitol (a paracellular marker) when compared to
SM and, in addition, did not cause any significant decrease in TEER, indicating that
CMII most likely acts on transcellular routes for insulin absorption. The use of
88
proteinase inhibitors has been successful in enhancing bioavailability of various
peptides (Yamamoto et al. 1994) (Morita et al. 1994) (Machida et al. 2000). Insulin,
which transports paracellularly at baseline (Patton 1996), when administered through
the lungs, is likely to be most susceptible to membrane-associated aminopeptidase
(Forbes et al. 1999). Although peptide degradation is a major concern, incubating
insulin in SM and CMII for 2 h at 37
o
C showed no significant degradation upon
TCA precipitation. Therefore, possible degradation of insulin takes place either on
the surface cell membrane or within the cell after insulin uptake. In this study, using
proteinase inhibitor cocktail protected insulin from degradation in CMII rather than
SM. This might seem to contradict what has been published regarding insulin
absorption enhancement in presence of proteinase inhibitors (Yamamoto et al. 1994)
(Forbes et al. 1999); however, the inhibitor cocktail composition in our study, which
mainly protected against cytosolic and lysosomal proteinases, was different from
reagents used in other studies (Yamamoto et al. 1994) (Fukuda et al. 1995) (Forbes
et al. 1999) and might not have had the same protective activity against insulin
degradation by aminopeptidase present on cell membrane surface. These results may
indicate a partial rerouting of insulin transport through the transcellular pathway,
exposing it to degradation by cytosolic proteinases, which include insulin degrading
enzyme, known to be present in cultured type II alveolar cells (Hsu and Bai 1998).
Regardless of the high degradation, this transcellular pathway can maintain the
amount of insulin transport at a higher level than that in cells incubated with SM.
89
Bile salts and lipid surfactants have also been studied as enhancers in protein
formulations for pulmonary delivery (Patton 1996) (Kobayashi et al. 1994) (Bur et
al. 2006). The mechanism of action has been attributed to opening of tight junctions,
production of insulin monomers, and regional phase changes and/or pore formation
due to reverse micelle formation in cell membranes (Johansson et al. 2002) (Hirai
1981) (Gordon et al. 1985). When 1, 2-dipalmitoyl phosphatidylcholine (DPPC), a
component of pulmonary surfactants, was incorporated with a mixture of liposome
and insulin and delivered via pulmonary routes, a significant hypoglycemic effect
was observed, suggesting a possible effect of the lipid surfactant on membrane
perturbation (Mitra et al. 2001). The results of the current study, however, indicate
that the factor is a protein. Whether this factor can cause de-aggregation of insulin
hexamers or otherwise cause phase changes in biomembranes remains to be
elucidated. Moreover, membrane re-arrangement or fluidization seems unlikely
since enhanced insulin transport was observed only in the apical-to-basal direction
and not the other way around. Furthermore, CMII from the apical but not basolateral
compartment of RAECM-II showed enhancement of insulin transport, indicating the
possibility that the factor(s) were solely secreted and may also be recycled at the
apical membrane.
The involvement of endocytosis as an enhancement mechanism for
transalveolar epithelial insulin transport in the presence of conditioned medium from
RAECM-II was also investigated. Although CMII did not have marked effects on
triggering fluid-phase endocytosis as determined by the marker horseradish
90
peroxidase, effect of CMII on insulin transport was abolished when incubated at 16
and 4
o
C. By contrast, insulin dosed in SM followed the conventional rule of passive
diffusion with a decrease of ~50% when incubated at 16
o
C (Hastings et al. 2004).
Taken together, these results, along with the effect of the proteinase inhibitor
cocktail on insulin transport, may suggest that the factor(s) present in CMII could
either be interacting with insulin or influencing the cellular membrane thereby
ultimately leading to a rerouting of insulin transport from a paracellular pathway to a
transcellular pathway other than nonspecific pinocytosis (i.e., possibly through
macro-pinocytosis or receptor-mediated endocytosis). To date, studies involving
receptor-mediated endocytosis as a target mechanism for enhancing peptide delivery
have used ligand-based conjugates such as transferrin conjugated to insulin, GCSF or
HRP (Widera et al. 2003) (Shah and Shen 1995) (Deshpande et al. 1994). However,
whether or not insulin is transported via receptor-mediated transcytosis when
complexed with factor(s) in CMII, remains to be determined.
It is known that type II pulmonary epithelial cells secrete lipids and protein
surfactants, a characteristic which is maintained for the length of cell culture (seven
days) upon supplementation of culture medium with KGF (Borok et al. 1998). To
determine whether these secreted factor(s) are implicated in enhancing peptide
transport, it was necessary to at least partially purify conditioned medium and narrow
down the range of proteins involved. However, the major limitation was the
presence of newborn bovine serum proteins in CMII that made it difficult to
accomplish such a step.
91
Culturing alveolar cells in medium supplemented with only 1% instead of 10%
serum abolished the effect of CMII which required a different approach for this
study.

2. Mechanistic investigation of the effect of SP-A on insulin transport
The assumption that surfactant protein A (SP-A) may be involved in
enhancement of insulin transport was based on the observation that the activity
afforded by conditioned medium collected from RAECM-II was explicitly seen
across RAECM-II. SP-A, which binds to surfactant lipids in the lungs (Wissel et al.
2001), also belongs to the collectin family that bind to mannose residues present on
cell or particle surfaces (Sidobre et al. 2000).
Using
125
I-mannosylated-albumin as a marker that may bind and track the
effect of SP-A in CMII, failed to yield any difference in its binding profile as
compared to SM. Collectins are also present in serum but SM had no effect on
enhancing insulin transport across RAECM-II. If binding was indeed through the
carbohydrate binding domain of SP-A, then species difference between the rat SP-A
and bovine collectin prevented a similar transport pattern in SM versus CMII across
RAECM-II. Given that the complexity of serum in the medium was a limiting
component to further assess the effects of CMII factor(s) on insulin transport, we
decided to use lavage fluid obtained from rat lungs as an appropriate substitute to
CMII and as a better representation of the physiological components of the alveolar
lining fluid. Similar to the effect of CMII, bronchoalveolar lavage fluid (BALF) also
92
enhanced transport of insulin across RAECM-II. A Western blot of a concentrated
aliquot of BALF showed that it contained SP-A, but incubation of rat SP-A (a gift
from Dr. McCormick) with insulin did not significantly enhance transport of the
insulin across RAECM-II. There are several explanations to this: 1) the effect
elicited by SP-A may be multi-factorial and may require the presence of other
cofactors that may influence its aggregation or binding to insulin; 2) the assumption
that SP-A receptor exists on RAECM-II is based on mRNA detection (Borok et al.
1998), but neither the expression or the function of SP-A receptor has been reported
in this model; and 3) the factor(s) expressed specifically by type II pneumocytes
could be a protein other than SP-A which is involved in enhancement of insulin
transport across RAECM-II seen in both CMII and BALF. Mason et al., emphasized
the importance of the cell culture system in influencing the secretion, processing and
transport of SP-A (Mason et al. 2002). Similarly, optimization of conditions such as
cell culture, incubating medium (pH, salts, ionic strength) and activity of protein are
necessary to further investigate the effect of SP-A on transport of insulin or any other
protein candidate across RAECM-II.

3. Effect of factor(s) in BALF on insulin transport
As mentioned earlier, it was observed that rat lung lavage fluid contains a
factor(s) which enhances transport of insulin across RAECM. It was necessary,
first, to concentrate BALF, which represents a diluted sample of rat epithelial lining
fluid, by ~7.5-fold to obtain significant transport-enhancing activity. The lack of
93
such an enhancement across Caco-2 or MDCK cell monolayers suggests tissue
specificity of the factor(s) involved. Molecular sizing using a Centricon device
suggested that the factor(s) is greater than 50 kDa. The insulin transport-enhancing
effect of 7.5x BALF was decreased upon heat deactivation and tryptic-digestion
suggesting that the factor involved is a protein.
Transport across the alveolar epithelial barrier may take place via paracellular
or transcellular routes, depending on the size, shape, or charge of the transported
molecule. For insulin (5.8 kDa), transport is thought to be predominantly
paracellular as indicated by its low permeability value across RAECM (Yamahara et
al. 1994a) as well as its small diameter (~2.2nm) relative to the estimated pore
diameter of the alveolar epithelium (Patton 1996). Our results showed that transport
of insulin when dosed in 7.5x BALF across RAECM-II was not due to enhanced
passive diffusion via paracellular routes is evidenced by the unchanged permeability
of two paracellular markers,
3
H-mannitol and sodium fluorescein, when dosed in
either KRP or 7.5x BALF. Moreover, 7.5x BALF did not affect TEER (>2,000
ohm·cm
2
). Fluid phase endocytosis did not seem to be the main route for insulin
transport, as no significant difference in transcytosis of the marker, FITC-dextran (70
kDa), was noted with 7.5x BALF or KRP as the dosing solution. These findings
provide evidence that a non-diffusional mechanism is responsible for enhancement
of insulin transport across RAECM in the presence of BALF factor(s).

94
To investigate whether enhancing effects of BALF factor(s) involved cell
energy-requiring process(es), insulin transport was studied at 37ºC and 4°C.
Lowering the temperature from 37ºC to 4°C reduced insulin transport by ~37% when
dosed in KRP. This is comparable to the decrease (~40%) in flux of small molecular
weight dextran (4 kDa) as a result of lowering temperature from 37ºC to 4ºC and is
consistent with a predominant diffusional pathway (Matsukawa et al. 1997). The
enhanced transport of insulin observed with 7.5x BALF at 37ºC was reduced to
baseline transport similar to that observed for insulin with KRP at 4ºC, which implies
that an additional pathway besides paracellular transport and one that involves an
energy-dependent mechanism such as transcytosis may be underway (Kim and Malik
2003). Monensin, an ionophore reported to interfere with Na
+
/H
+
exchange in
endosomes, has been used in various studies to determine the involvement of
clathrin-mediated receptor endocytosis of certain ligands.(Hastings et al.
1994)
,
(Holladay et al. 1999) By altering the acidic pH of the endosome
environment, monensin disrupts dissociation of ligand-receptor complex and
prevents recycling of the receptor to the plasma membrane, thereby inhibiting
endocytosis.(Hastings et al. 2004) When monensin was added to 7.5x BALF, there
was a reduction of insulin transcytosis by ~50% across RAECM-II, whereas
monensin had no effect on insulin transport when dosed in KRP. Taken together,
these observations suggest the involvement of a receptor on RAECM-II, with a
cognate factor(s) present in BALF, which seems to endocytose both factor(s) and
insulin simultaneously. The identity of such receptor(s) remains to be determined.
95
Our results show that the transport of insulin dosed in 7.5x BALF across
RAECM-II was increased despite of a higher percent of degradation resulting from a
“transcellular” pathway as compared to the paracellular pathway observed with
KRP-dosed insulin. There are several issues that must be considered when trying to
explain the degradation profile of insulin: 1) degradation of insulin occurs in BALF
due to proteolytic enzymes present in the ELF; 2) transport of KRP-dosed insulin
results in higher percentage of degradation across RAECM-II compared to RAECM-
I (our observations) suggesting that more proteolytic activity occurs in type II
pneumocytes due to differences in peptidase and/or proteinase expression levels as a
result of cell transdifferentiation (Forbes et al. 1999); and 3) transport of BALF-
dosed insulin causes further degradation of insulin across RAECM-II possibly due to
an encounter with lysosomal proteases as transport of insulin shifts from diffusional
to non-diffusional pathway in presence of BALF factor(s), a phenomenon also
observed with CMII (Bahhady et al. 2007). Regardless of the high degradation
profile resulting from apical-to-basolateral transport of BALF-dosed insulin, the final
amount of intact insulin remains highest in this condition. These results as well as
our findings, that at least two of the factors belonged to the macroglobulin family
(these findings are discussed in detail below), may indicate that the association
between BALF factor(s) and insulin protected insulin from degradation
predominantly by peptidases expressed on apical membrane of pneumocytes (Forbes
et al. 1999) (Yamahara et al. 1994a). Once bypassing these membrane-bound
peptidases, as a result of insulin interaction with macroglobulin, the intracellular fate
96
of insulin becomes that of its associated ligand. This hyopothesis is discussed
further in the next section.
Our observation of a unidirectional enhancement of insulin flux strictly from
apical-to-basolateral direction of RAECM-II, implies that if indeed a transcellular
pathway is involved, then the effect of BALF factor(s) is limited to the apical side
and not the basal side of RAECM-II (e.x., apical expression of receptor). Although
insulin was inherently degraded more from a-to-b rather than b-to-a when dosed in
KRP (86.5% vs 43%, respectively), the percent of insulin degraded in BALF
surpassed that seen in KRP in both directions probably due to proteolytic enzymes
present in BALF. Also, as mentioned earlier, the enhanced transport afforded by
BALF factor may be occurring through an endosomal-lysosomal pathway and may
expose insulin to an additional degradative environment.
It is evident from the SDS-PAGE analysis of crude, concentrated BALF that
a large number of proteins are represented in the epithelial lining fluid of the lungs.
In fact proteomic data from human or rat BALF depict an array of more than fifty
proteins ranging from plasma proteins to proteins produced by cells endogenous to
the lung (Wattiez and Falmagne 2005) (Signor et al. 2004). The BALF factor(s)
contributing to the insulin transport-enhancing activity was detected in the higher
molecular weight fraction obtained from the S-200 partial purification gel
chromatography. Enhanced transport of insulin observed with BALF fractions was
~4-fold or ~1-fold in those fractions containing proteins >100 kDa or <100 kDa,
respectively. Furthermore, 7.5x BALF enhanced insulin transport 2-fold more than
97
purified 7.5x BALF (fraction #22). These findings suggest that enhancement of
insulin transport across RAECM-II requires a combination of factors, but that it is an
effect elicited predominantly by large molecular weight proteins. LC-MS/MS
analysis, after in-gel digestion of the high molecular weight protein bands, identified
the proteins as: alpha-1-inhibitor III, murinoglobulin gamma 2, and pregnancy-zone
protein, all which belong to the macroglobulin family.

4. Mechanistic investigation of the effect of alpha-1-inhibitor III ( α
1
I
3
) and
alpha-2 macroglobulin ( α
2
M) on insulin transport
Given that alpha-1-inhibitor III ( α
1
I
3
) was the most abundant protein among
the high molecular weight protein fraction, we decided to further investigate its
involvement in enhancing insulin transport across RAECM-II. In fact, α
1
I
3
and
murinoglobulin 2 may resemble identical proteins as suggested by Saito and
Sinohara (Saito and Sinohara 1985). α
1
I
3
(mwt 200 kDa) is a homologue of alpha-2
macroglobulin ( α
2
M; mwt 700 kDa), which is found in a number of species including
human (Gliemann and Sottrup-Jensen 1987). As proteinase inhibitors, both share a
common mechanism whereby a proteinase-susceptible region (known as the “bait”
region) lures the endopeptidase and leads to cleavage of the macroglobulin (Salvesen
and Barrett 1980). Cleaving of the bait region exposes β-cysteinyl- γ-glutamyl thiol
esters which in turn attack lysine residues in the proteinase resulting in a
conformational change that in turn immobilizes the proteinase and lead to its
inactivation (Sottrup-Jensen et al. 1981). This conformational change allows the
98
exposure of the C-terminal residues, which share a similar homology in both α
1
I
3
and
α
2
M (Enghild et al. 1989), and which are necessary for ligand/receptor binding.
Moreover, it is only this form of proteinase-activated α
1
I
3
or α
2
M that is able to bind
with high affinity to its cognate receptor: the low density lipoprotein receptor-related
protein/ α
2
M receptor (LRP/ α
2
MR) found on various cells.
Based on these similarities, we decided to use human α
2
M (which is available
commercially) as the prototype model for investigating its effect on insulin transport
across RAECM-II. Various studies have shown that in addition to its proteinase
inhibitory activity, α
2
M exhibits nonproteolytic ligand binding capacity (Gron and
Pizzo 1998). The ligands include growth factors (Mathew et al. 2003) (Liu et al.
2001), antigens (Chu and Pizzo 1993), and insulin (Chu et al. 1991), among others.
The interaction between insulin and α
2
M is mainly covalent and occurs only in the
presence of a proteinase, e.g., trypsin. In our experiments, binding of radiolabeled
insulin to α
2
M occurred exclusively in the presence of trypsin, supporting the
mechanism put forth by Chu et al. (Chu et al. 1991). Whether a similar mechanism
occurs between α
1
I
3
present in BALF and insulin could only be speculated. On the
one hand, it is well established that both α
1
I
3
and α
2
M share a common proteinase
inhibitory mechanism; but on the other hand, the simultaneous presence of insulin is
a prerequisite for its incorporation during the trypsin-induced conformational change
of α
2
M (Chu et al. 1991). Although it is probable that α
1
I
3
-

proteinase has already
been formed in BALF, this does not completely exclude the presence of free α
1
I
3

partly because it exists in a state of equilibrium and partly due to the reversibility of
99
the thiol ester bond under certain conditions (Gron et al. 1996). Furthermore,
association of preformed α
1
I
3
-

proteinase to insulin could be a result of non-covalent
binding due to possible electrostatic interactions though such binding was shown not
to exceed 10-15% in case of α
2
M with insulin or lysozyme (Chu and Pizzo 1993).
Our results have shown that the enhancement of insulin transport in the
presence of BALF was most pronounced across RAECM-II. The lack of such
observation in Caco-2 or MDCK cell monolayers may be due to the absence of
receptor expression. Similarly as RAECM-II transdifferentiate to RAECM-I, a
change in expression level of various proteins occurs possibly explaining why the
transport-enhancing activity was abolished with the use of fraction #22 from S-200
purified BALF. As mentioned earlier, activated α
1
I
3
has been shown to bind to
LRP/ α
2
MR (Gliemann and Sottrup-Jensen 1987). This receptor, as defined by
immunohistochemical labeling, exists on bronchial and alveolar epithelium of human
tissue (Moestrup et al. 1992). However, there is conflicting data regarding its
presence on rat alveolar epithelium (Zheng et al. 1994) (Stefansson et al. 1995). In
fact a closely related receptor, gp330 (known as megalin), which is also a member of
the low density lipoprotein receptor family (Fig. 33), shows a high distribution in rat
type II pneumocytes (Zheng et al. 1994). Gp330 is expressed on the apical surface
of type II pneumocytes and is involved in clathrin-mediated endocytosis of various
ligands (Moestrup 1994). We have also shown through Western blotting that
megalin is expressed by RAECM-II culture model used in this study. Although
gp330 and LRP share several common ligands, α
2
M binds exclusively to LRP
100
(Moestrup 1994). This, however, dos not necessarily eliminate the possibility that
α
1
I
3
may be endocytosed by gp330 since differences in structure and affinity kinetics
do exist in both macroglobulins (Moestrup and Gliemann 1991).

Figure 33. Macroglobulin family and low density lipoprotein receptor family
(LDLR). LDL-related receptor protein-1 (LRP-1) and LRP-2 also known as gp330
or megalin, very low density lipoprotein receptor (VLDLR) and LDLR are members
of the LDLR family and bind a broad and partially overlapping spectrum of
proteins/ligands exposing regions rich in positively charged amino acid residues
(Moestrup 1994).

Macroglobulin family
Low density lipoprotein receptor family
(LDLR)
LRP-1
LRP-2
gp330/megalin
VLDLR LDLR
α1-inhibitorIII
pregnancy zone protein
α2-macroglobulin
X
?
?
101
Our findings indicate that the enhanced transport of insulin afforded by fraction B
from purified BALF (fraction #22 and fraction B are identical but are obtained using
either TFF or ultracentrifugation, respectively) was partially inhibited in the presence
of anti-megalin inhibitor (1H2) or excess activated α
2
M·insulin complex (~40% or
~60%, respectively). The ineffectiveness of either ligands to completely abolish the
amount of enhanced insulin transport across RAECM-II likely suggests that an
involvement of more than two receptors and possibly a number of different
pathways, other than paracellular or fluid-phase endocytosis. Furthermore, knowing
that fraction B contains a mixture of macroglobulins ( α
1
I
3
and PZP) each defined by
different receptor affinities and intracellular processes limits our interpretation and
requires a more detailed mechanistic investigation of each ligand and its cognate
receptor in RAECM-II. For example, Moestrup et al. showed that receptor-binding
of activated α
2
M complex to LRP followed two different dissociation constants
depending on α
2
M concentration (Fig. 34.). On the other hand, α
1
I
3
did not fit this
model since it is a monomer of α
2
M and thereby bound to LRP through its single
domain with a higher dissociation constant (Moestrup and Gliemann 1991).

Figure 34. Schematic of a postulated interaction between activated α
2
M or α
1
I
3

complex with LRP. The figure is from ref. (Moestrup SK, 1991).
102
Another important aspect of the α
1
I
3
/ α
2
M-insulin internalization is the
intracellular trafficking. Both receptors, LRP and megalin, are internalized via the
clathrin-mediated pathway. Whether the final destination of the intracellular
trafficking of the ligand once taken up by LRP or megalin is the lysosome, depends
on both the ligand and the cell type involved. Our results show that transport of
either total or intact insulin dosed in BALF fraction #22 was enhanced from a-to-b
across RAECM-II. Addition of megalin or LRP competitors reduced both total and
intact amounts of the enhanced insulin transport in BALF fraction B. These findings
suggest that the mechanism of enhanced insulin transport-activity afforded by BALF
factor(s) likely involves trafficking through an endosomal/lysosomal pathway.
It is well documented that most α
2
M complexes are internalized by certain
cells and accumulated in lysosomes within 10-20 min (Morita et al. 1994).
Previously, Shimizu et al., had shown that when α
2
M was chemically conjugated to
insulin through a disulfide bond, internalization of the complex by the α
2
M receptor
in hepatoma cells did not stimulate tyrosine aminotransferase activity (Ito et al.
1984), but stimulation did occur upon internalization of the conjugate via the insulin
receptor. On the other hand, some ligands such as Pseudomonas exotoxin
complexed to α
2
M, bind with high affinity to LRP in fibroblasts and manage to exert
toxicity signifying its capability of evading lysosomal degradation (Kounnas et al.
1993). In addition to its proteinase inhibitory action, α
2
M serves as a binding carrier
that modulates the biological activity of certain proteins such as cytokines, growth
factors, leptin, among others. The binding may include hydrophobic interactions or
103
otherwise covalent bonding through amide, ester, or disulfide linkage. The
dissociation of bound ligand is usually a result of membrane-bound reductases or
endosomal/lysosomal reduction and/or acidification. In summary, several studies
indicate that the intracellular processing of α
2
M complexes is determined by cell
type, ligand, ligand/ α
2
M association, and that the outcome could include intracellular
sequestration, endosomal translocation to cytosol, or transcytosis in the case of
polarized cells (Borth 1992).
Megalin, which belongs to the same family as α
2
MR/LRP, is exclusively
found on certain epithelial cells. Similar to α
2
MR/LRP, megalin-mediated uptake of
ligand usually results in lysosomal accumulation. However, a recent study shows
that transcytosis of retinol-binding protein across renal proximal tubule cells via the
megalin receptor did not result in appreciable lysosomal degradation (Marino et al.
2001). In contrast, as shown by the same study, lysozyme, a substrate of megalin,
was transported and degraded in the lysosomes suggesting a ligand-dependence
intracellular trafficking. In addition, megalin expressed on different cells has lead to
different intracellular processing of the same ligand as in the case of apolipoprotein J
(Zlokovic et al. 1996). Although megalin is present on type II pneumocytes, it has
not been studied extensively and its physiological role is focused on the uptake of
lipoprotein lipase for the synthesis of lipid surfactants. Whether megalin in type II
pneumocytes may be involved in transcytosis of certain ligands is an area worth
exploring.

104
Chapter Six: Conclusion
Relevant to protein and peptide drug delivery, the lungs have proven to be
useful for systemic absorption. Established reasons are inherent to the lung’s large
surface area, thin epithelium, rich blood supply, and extracellular proteinase
inhibitory activity (Patton et al. 2004). However, studies regarding mechanistic
information of interaction between the lung epithelium and its microenvironment in
drug delivery are still lacking. In particular, the discrepancy between the low in vitro
peptide transport across cultured RAECMs and the high in vivo bioavailability of
several peptide and protein drugs is still largely unsolved.
Our results showed that CMII enhanced insulin transport across RAECM-II
but not RAECM-I. Factors in BALF, on the other hand, seemed to exert an
enhancing effect on both alveolar epithelial cell types. It is difficult to make a
comparison of the components present in both biological fluids. In the first scenario
(involving CMII) we hypothesized that a factor released explicitly by RAECM-II
could be SP-A which also proved to be a constituent of BALF. However, purified
SP-A (gift from McCormik) failed to substantiate this hypothesis. This does not
necessarily eliminate involvement of SP-A, but may suggest that a combination of
factors may be required to elicit the transport-enhancing effect observed with CMII
and BALF. In the second scenario (involving purified BALF), we determined
through LC-MS/MS that at least one of the factors contributing to enhanced insulin
transport across RAECM-II is α
1
I
3
, a member of the macroglobulin family. Could a
common thread exist that may link observations from both CMII and BALF? The
105
secretion of surfactant proteins by type II pneumocytes also involves a concomitant
processing by cathepsins. Therefore cathepsins are secreted by cultured rat type II
alveolar epithelium and are additionally present in BALF (Ishii et al. 1991). Given
that α
2
M is a component of bovine serum, which is used to supplement cell culture
medium, and that cathepsins represent one of the substrates for macroglobulins,
provides a milieu that facilitates the incorporation of insulin through the mechanism
described previously. The subsequent transcytosis of a larger amount of insulin in
RAECM-II as compared to RAECM-I in the presence of purified BALF may be due
to the absence of a receptor (on RAECM-I) or the presence of a receptor with either
low abundance and/or affinity, since several phenotypic changes are known to occur
when RAECM-II transdifferentiate into RAECM-I in vitro (Borok et al. 1998)
(Dobbs et al. 1985). An hypothesis worth exploring, is whether factors in BALF
may have induced the expression of particular receptor proteins on type II
pneumocytes and led to the observed enhancement of insulin transport. Although
induction of protein expression requires time, post-translational modification could
be stimulated within a short period. Furthermore, the expression of the factor(s)
throughout RAECM-II culture period may in turn affect the presence of receptors as
in the case of cytokine-induced expression of alpha-2 macroglobulin receptor on
alveolar macrophages ((Hussaini et al. 1990). A recent study showed that lung fluid
from patients with acute lung injury decreased expression of sodium and chloride
transport proteins on type II pneumocytes (Lee et al. 2007). On the other hand,
Gandhi et al., demonstrated that a globulin fraction of purified pulmonary edema
106
fluid increased expression of sodium channels in type II pneumocytes (Gandhi et al.
2007). These findings underscore the importance of proteins within the lung fluid in
regulating the cellular machinery of the alveolar epithelium.
Given that alveolar type II cells cover about 5% of the total alveolar surface,
the additional effect afforded by an endogenous factor may contribute far less to total
net absorption of insulin in vivo. This scenario may also be true if peptides are
transported across predominately type I cells of the in vivo lung. It seems, however,
that the alveolar microenvironment provides a plethora of factors which include
proteins, lipids and mucopolysaccharide that may have synergistic effects for
facilitating the transport of some proteins across an otherwise tight epithelial barrier.
Nevertheless, interpretation of our data must be done with caution given the
limitations in translating in vitro data on drug transport to in vivo situation as evident
in the studies utilizing IgG or albumin in the context of their cognate receptors, FcRn
and gp60, respectively (Sakagami 2006). It is true that rat primary alveolar cell
culture provides a simplistic approach in unraveling some of the mechanistic
pathways afforded by the alveolus, yet, the in vivo implications remain to be
determined.
In summary, we show that conditioned medium from rat alveolar type II cell-
like monolayers and bronchoalveolar lavage fluid contain (a) protein factor(s) which
facilitate(s) transcellular transport of insulin across primary RAECM-II.
Observations from both studies highlight the importance of the alveolar epithelium in
maintaining lung homeostasis through various mechanisms, including endocytosis
107
(Kim and Malik 2003) and may further explain the high bioavailability of several
peptide and protein drugs in vivo. The present study underscores the relevance of the
alveolar microenvironment to insulin delivery and sheds insights on the possible
mechanisms involved.

108
Chapter Seven: Future Perspectives
It would be of interest to investigate whether the transport-enhancing effect
of BALF is applicable to protein drugs other than insulin, which are currently being
used via pulmonary administration such as: growth hormone, interferon, etc. (Patton
and Byron 2007). The next question would be whether alpha-2 macroglobulin may
be used as a drug delivery agent? For example, in ARDS, it has been found that IL-8
complexes to α
2
M resulting in several outcomes: 1) protection of IL-8 against
proteolytic degradation; 2) acting as a reservoir; and 3) prevention sticking of IL-8 to
interstitial matrix (Kurdowska et al. 2002).
Increasing drug retention for peptide or protein drugs is important to
overcome the short systemic duration which usually necessitates frequent
administration (Patton and Byron 2007). One example of inherent tissue retention is
seen with some steroids that are retained in the tissue through esterification of their
long-chain fatty acids thereby prolonging their effect (Esmailpour et al. 1997).
Using α
2
M to increase the retention of certain short acting peptides/small proteins
that are substrates for exopeptidases is a valid proposal; however, incorporation of
peptides is limited to those with large molecular weight, to avoid degradation by the
simultaneous presence of the proteinase. However, to overcome such a problem, the
peptide may be incorporated via activation of the α
2
M with a small nucleophile that
results in a similar conformational change afforded by the proteinase (Gron and
Pizzo 1998). Since the complex is naturally occurring, it would not elicit an
antigenic immune response observed with chemically conjugated protein-ligand
109
complexes. The remaining question would be whether this is a suitable approach for
macromolecule drugs such as proteins whereby extended retention in the lungs may
cause an increased susceptibility to macrophage engulfment. It is true that α
2
M may
protect the protein from degradation, but on the other hand the presence of α
2
M
receptors on alveolar macrophages may limit its use as a delivery agent unless the
purpose is to target macrophages, for example in the case of tuberculosis. This
approach has been successful in delivering α
2
M conjugated α-galactosidase to
fibroblasts from patients with Fabry’s disease (Osada et al. 1987), or α
2
M conjugated
viral proteins to macrophages for increased anti-viral antibody secretion (Osada et al.
1988). On the other hand, the unique presence of megalin on type II pneumocytes
could also be utilized as a pathway for transcytosis of ligands given that the
intracellular trafficking does not involve lysosomal degradation as in the case of
thyroglobulin transcytosis across thyrocyte cell line (Marino et al. 2000). This could
be accomplished by using anti-megalin antibody as a carrier attached to the potential
cargo that would be targeted specifically for transcytosis across type II pneumocytes.
In fact, a study by Pan et al showed that a high-capacity transport system mediates
the transcytosis of receptor-associated protein (RAP), which is a substrate for both
LRP-1 and -2, across the blood brain barrier through LRP-2 receptor expressed on
the apical membrane of the cerebral vasculature (Pan et al. 2004).
Ultimately, the implementation of alpha-2 macroglobulin as an adjunct
delivery agent for protein drug inhalation will involve certain restrictions that depend
on the lung condition and the presence of concomitant disease that may increase or
110
decrease expression of proteins in the lavage fluid and interfere with the absorption
of inhaled drugs. For example, cathepsin L is increased in the lavage fluid of
smokers (Takahashi et al. 1993) and since cathepsin is one of the substrates of alpha-
2 macroglobulin, increased insulin absorption may occur. Needless to say that these
pathological situations will restrict any sort of systemic inhalation therapy, including
insulin.
Although the proteins that we have identified as being involved in enhancing
transport of insulin across RAECM-II belong to the macroglobulin family, the
presence of a plethora of factors in the BALF provides a limitless pool of candidates
to be tested (Fig. 35). Moreover, the intrinsic mechanisms provided by the human
physiology such as the numerous cellular processing pathways may assist in the
advancement of developing more specific and targeted drug delivery agents
necessary for the success of treating resilient human ailments.

111

Figure 35. Summary of findings. 1) The drug, which is intended for pulmonary
delivery (in this case insulin) presumably, associates with factor(s) in epithelial
lining fluid, such as the protein shown in this study which belongs to the
macroglobulin family. 2) This complex (macroglobulin/insulin) is internalized
through clathrin-mediated endocytosis via a receptor which belongs to the low-
density lipoprotein receptor related protein. The fate of the endocytosed complex is
unknown but may include: 3) translocation to the cytosol and either degradation by
cytosolic degrading enzymes; 4) lysosomal uptake and degradation by lysosomal
proteases; and 5) transcytosis and permeation of insulin in amounts surpassing that
observed for insulin in control buffer or medium.

epithelial
lining fluid
drug
lysosomal degradation
transcytosis
translocation
endocytosis
3
4
5
1
2
112
References
Bahhady R, Kim KJ, Borok Z, Crandall ED, Shen WC (2007) Enhancement of
Insulin Transport across primary rat alveolar epithelial cell monolayers by
endogenous cellular factor(s). Pharm Res. In press
Benson BJ, Hawgood S, Williams MC (1984) Role of apoprotein and calcium ions in
surfactant function. Exp Lung Res 6:223-236
Birn H, Zhai X, Holm J, Hansen SI, Jacobsen C, Christensen EI, Moestrup SK
(2005) Megalin binds and mediates cellular internalization of folate binding protein.
Febs J 272:4423-4430
Bitonti AJ, Dumont JA (2006) Pulmonary administration of therapeutic proteins
using an immunoglobulin transport pathway. Adv Drug Deliv Rev 58:1106-1118
Bitonti AJ, Dumont JA, Low SC, Peters RT, Kropp KE, Palombella VJ, Stattel JM,
Lu Y, Tan CA, Song JJ, Garcia AM, Simister NE, Spiekermann GM, Lencer WI,
Blumberg RS (2004) Pulmonary delivery of an erythropoietin Fc fusion protein in
non-human primates through an immunoglobulin transport pathway. Proc Natl Acad
Sci U S A 101:9763-9768
Borok Z, Lubman RL, Danto SI, Zhang XL, Zabski SM, King LS, Lee DM, Agre P,
Crandall ED (1998) Keratinocyte growth factor modulates alveolar epithelial cell
phenotype in vitro: expression of aquaporin 5. Am J Respir Cell Mol Biol 18:554-
561
Borth W (1992) Alpha 2-macroglobulin, a multifunctional binding protein with
targeting characteristics. Faseb J 6:3345-3353
Brandes ME, Finkelstein JN (1989) Stimulated rabbit alveolar macrophages secrete a
growth factor for type II pneumocytes. Am J Respir Cell Mol Biol 1:101-109
Bur M, Huwer H, Lehr CM, Hagen N, Guldbrandt M, Kim KJ, Ehrhardt C (2006)
Assessment of transport rates of proteins and peptides across primary human alveolar
epithelial cell monolayers. Eur J Pharm Sci 28:196-203
113
Castranova V, Rabovsky J, Tucker JH, Miles PR (1988) The alveolar type II
epithelial cell: a multifunctional pneumocyte. Toxicol Appl Pharmacol 93:472-483
Christensen EI, Moskaug JO, Vorum H, Jacobsen C, Gundersen TE, Nykjaer A,
Blomhoff R, Willnow TE, Moestrup SK (1999) Evidence for an essential role of
megalin in transepithelial transport of retinol. J Am Soc Nephrol 10:685-695
Chu CT, Pizzo SV (1993) Receptor-mediated antigen delivery into macrophages.
Complexing antigen to alpha 2-macroglobulin enhances presentation to T cells. J
Immunol 150:48-58
Chu CT, Rubenstein DS, Enghild JJ, Pizzo SV (1991) Mechanism of insulin
incorporation into alpha 2-macroglobulin: implications for the study of peptide and
growth factor binding. Biochemistry 30:1551-1560
Deshpande D, Toledo-Velasquez D, Wang LY, Malanga CJ, Ma JK, Rojanasakul Y
(1994) Receptor-mediated peptide delivery in pulmonary epithelial monolayers.
Pharm Res 11:1121-1126
Dobbs LG, Williams MC, Brandt AE (1985) Changes in biochemical characteristics
and pattern of lectin binding of alveolar type II cells with time in culture. Biochim
Biophys Acta 846:155-166
Dubois M, Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F. (1956) Colorimetric
method for determination of sugars and related substances. Analytical Chemistry
28:350-356
Enghild JJ, Thogersen IB, Roche PA, Pizzo SV (1989) A conserved region in alpha-
macroglobulins participates in binding to the mammalian alpha-macroglobulin
receptor. Biochemistry 28:1406-1412
Esmailpour N, Hogger P, Rabe KF, Heitmann U, Nakashima M, Rohdewald P
(1997) Distribution of inhaled fluticasone propionate between human lung tissue and
serum in vivo. Eur Respir J 10:1496-1499

114
Ferkol T, Cohn LA, Phillips TE, Smith A, Davis PB (2003) Targeted delivery of
antiprotease to the epithelial surface of human tracheal xenografts. Am J Respir Crit
Care Med 167:1374-1379
Forbes B, Wilson CG, Gumbleton M (1999) Temporal dependence of ectopeptidase
expression in alveolar epithelial cell culture: implications for study of peptide
absorption. Int J Pharm 180:225-234
Fukuda Y, Tsuji T, Fujita T, Yamamoto A, Muranishi S (1995) Susceptibility of
insulin to proteolysis in rat lung homogenate and its protection from proteolysis by
various protease inhibitors. Biol Pharm Bull 18:891-894
Gallaher TK, Wu S, Webster P, Aguilera R (2006) Identification of biofilm proteins
in non-typeable Haemophilus Influenzae. BMC Microbiol 6:65
Gandhi SG, Rafii B, Harris MS, Garces A, Mahuran D, Chen XJ, Bao HF, Jain L,
Eaton DC, Otulakowski G, O'Brodovich H M (2007) Effects of cardiogenic edema
fluid on ion and fluid transport in the adult lung. Am J Physiol Lung Cell Mol
Physiol. In progress
Gliemann J, Sottrup-Jensen L (1987) Rat plasma alpha 1-inhibitor3 binds to
receptors for alpha 2-macroglobulin. FEBS Lett 221:55-60
Gonzalez R, Yang YH, Griffin C, Allen L, Tigue Z, Dobbs L (2005) Freshly isolated
rat alveolar type I cells, type II cells, and cultured type II cells have distinct
molecular phenotypes. Am J Physiol Lung Cell Mol Physiol 288:L179-189
Gordon GS, Moses AC, Silver RD, Flier JS, Carey MC (1985) Nasal absorption of
insulin: enhancement by hydrophobic bile salts. Proc Natl Acad Sci U S A 82:7419-
7423
Gron H, Pizzo SV (1998) Nonproteolytic incorporation of protein ligands into
human alpha 2-macroglobulin: implications for the binding mechanism of alpha 2-
macroglobulin. Biochemistry 37:6009-6014

115
Gron H, Thogersen IB, Enghild JJ, Pizzo SV (1996) Structural and functional
analysis of the spontaneous re-formation of the thiol ester bond in human alpha 2-
macroglobulin, rat alpha 1-inhibitor-3 and chemically modified derivatives. Biochem
J 318 ( Pt 2):539-545
Gumbleton M (2001) Caveolae as potential macromolecule trafficking compartments
within alveolar epithelium. Adv Drug Deliv Rev 49:281-300
Gumbleton M, Hollins AJ, Omidi Y, Campbell L, Taylor G (2003) Targeting
caveolae for vesicular drug transport. J Control Release 87:139-151
Haagsman HP (1998) Interactions of surfactant protein A with pathogens. Biochim
Biophys Acta 1408:264-277
Haagsman HP, Diemel RV (2001) Surfactant-associated proteins: functions and
structural variation. Comp Biochem Physiol A Mol Integr Physiol 129:91-108
Haagsman HP, Hawgood S, Sargeant T, Buckley D, White RT, Drickamer K,
Benson BJ (1987) The major lung surfactant protein, SP 28-36, is a calcium-
dependent, carbohydrate-binding protein. J Biol Chem 262:13877-13880
Hastings RH, Folkesson HG, Matthay MA (2004) Mechanisms of alveolar protein
clearance in the intact lung. Am J Physiol Lung Cell Mol Physiol 286:679-689
Hastings RH, Summers-Torres D, Yaszay B, LeSueur J, Burton DW, Deftos LJ
(1997) Parathyroid hormone-related protein, an autocrine growth inhibitor of
alveolar type II cells. Am J Physiol 272:L394-399
Hastings RH, Wright JR, Albertine KH, Ciriales R, Matthay MA (1994) Effect of
endocytosis inhibitors on alveolar clearance of albumin, immunoglobulin G, and SP-
A in rabbits. Am J Physiol 266:544-552
Heinemann L, Pfutzner A, Heise T (2001) Alternative routes of administration as an
approach to improve insulin therapy: update on dermal, oral, nasal and pulmonary
insulin delivery. Curr Pharm Des 7:1327-1351
116
Herzog V, Fahimi HD (1973) A new sensitive colorimetric assay for peroxidase
using 3,3'-diaminobenzidine as hydrogen donor. Anal Biochem 55:554-562
Hirai SY, T.; Mima, H. (1981) Mechanisms for the enhancement of the nasal
absorption of insulin by surfactants. Int J Pharm 9:173-184
Holladay SR, Yang Z, Kennedy MD, Leamon CP, Lee RJ, Jayamani M, Mason T,
Low PS (1999) Riboflavin-mediated delivery of a macromolecule into cultured
human cells. Biochim Biophys Acta 1426:195-204
Hsu MC, Bai JP (1998) Investigation into the presence of insulin-degrading enzyme
in cultured type II alveolar cells and the effects of enzyme inhibitors on pulmonary
bioavailability of insulin in rats. J Pharm Pharmacol 50:507-514
Hussain A, Arnold JJ, Khan MA, Ahsan F (2004) Absorption enhancers in
pulmonary protein delivery. J Control Release 94:15-24
Hussaini IM, Srikumar K, Quesenberry PJ, Gonias SL (1990) Colony-stimulating
factor-1 modulates alpha 2-macroglobulin receptor expression in murine bone
marrow macrophages. J Biol Chem 265:19441-19446
Ishii Y, Hashizume Y, Watanabe T, Waguri S, Sato N, Yamamoto M, Hasegawa S,
Kominami E, Uchiyama Y (1991) Cysteine proteinases in bronchoalveolar epithelial
cells and lavage fluid of rat lung. J Histochem Cytochem 39:461-468
Ito F, Ito S, Shimizu N (1984) Transmembrane delivery of polypeptide hormones
bypassing the intrinsic cell surface receptors: a conjugate of insulin with alpha 2-
macroglobulin (alpha 2M) recognizing both insulin and alpha 2M receptors and its
biological activity in relation to endocytic pathways. Mol Cell Endocrinol 36:165-
173
Johansson F, Hjertberg E, Eirefelt S, Tronde A, Hultkvist Bengtsson U (2002)
Mechanisms for absorption enhancement of inhaled insulin by sodium taurocholate.
Eur J Pharm Sci 17:63-71

117
John TA, Vogel SM, Minshall RD, Ridge K, Tiruppathi C, Malik AB (2001)
Evidence for the role of alveolar epithelial gp60 in active transalveolar albumin
transport in the rat lung. J Physiol 533:547-559
Kataoka M, Tavassoli M (1984) Synthetic neoglycoproteins: a class of regents for
detection of sugar-recognizing substances. J Histochem Cytochem 32:1091-1098
Khoo JC, Vance JE, Mahoney EM, Jensen D, Wancewicz E, Steinberg D (1984)
Neutral triglyceride lipase in macrophages. Arteriosclerosis 4:34-40
Khubchandani KR, Snyder JM (2001) Surfactant protein A (SP-A): the alveolus and
beyond. FASB J 15:59-69
Kim KJ, Fandy TE, Lee VH, Ann DK, Borok Z, Crandall ED (2004) Net absorption
of IgG via FcRn-mediated transcytosis across rat alveolar epithelial cell monolayers.
Am J Physiol Lung Cell Mol Physiol 287:616-622
Kim KJ, Malik AB (2003) Protein transport across the lung epithelial barrier. Am J
Physiol Lung Cell Mol Physiol 284:247-259
Kim KJ, Matsukawa Y, Yamahara H, Kalra VK, Lee VH, Crandall ED (2003)
Absorption of intact albumin across rat alveolar epithelial cell monolayers. Am J
Physiol Lung Cell Mol Physiol 284:L458-465
Kobayashi S, Kondo S, Juni K (1994) Study on pulmonary delivery of salmon
calcitonin in rats: effects of protease inhibitors and absorption enhancers. Pharm Res
11:1239-1243
Komada F, Iwakawa S, Yamamoto N, Sakakibara H, Okumura K (1994)
Intratracheal delivery of peptide and protein agents: absorption from solution and dry
powder by rat lung. J Pharm Sci 83:863-867
Kounnas MZ, Henkin J, Argraves WS, Strickland DK (1993) Low density
lipoprotein receptor-related protein/alpha 2-macroglobulin receptor mediates cellular
uptake of pro-urokinase. J Biol Chem 268:21862-21867
118
Kurdowska AK, Geiser TK, Alden SM, Dziadek BR, Noble JM, Nuckton TJ,
Matthay MA (2002) Activity of pulmonary edema fluid interleukin-8 bound to
alpha(2)-macroglobulin in patients with acute lung injury. Am J Physiol Lung Cell
Mol Physiol 282:L1092-1098
Kuroki Y, Mason RJ, Voelker DR (1988) Alveolar type II cells express a high-
affinity receptor for pulmonary surfactant protein A. Proc Natl Acad Sci U S A
85:5566-5570
Lee JW, Fang X, Dolganov G, Fremont RD, Bastarache JA, Ware LB, Matthay MA
(2007) Acute lung injury edema fluid decreases net fluid transport across human
alveolar epithelial type II cells. J Biol Chem. In press
Leslie CC, McCormick-Shannon K, Cook JL, Mason RJ (1985) Macrophages
stimulate DNA synthesis in rat alveolar type II cells. Am Rev Respir Dis 132:1246-
1252
Li L, Mathias NR, Heran CL, Moench P, Wall DA, Smith RL (2006) Carbopol-
mediated paracellular transport enhancement in Calu-3 cell layers. J Pharm Sci
95:326-335
Lin PM, Wright JR (2006) Surfactant protein A binds to IgG and enhances
phagocytosis of IgG-opsonized erythrocytes. Am J Physiol Lung Cell Mol Physiol
291:L1199-1206
Liu Q, Ling TY, Shieh HS, Johnson FE, Huang JS, Huang SS (2001) Identification
of the high affinity binding site in transforming growth factor-beta involved in
complex formation with alpha 2-macroglobulin. Implications regarding the
molecular mechanisms of complex formation between alpha 2-macroglobulin and
growth factors, cytokines, and hormones. J Biol Chem 276:46212-46218
Luton F, Mostov KE (1999) Transduction of basolateral-to-apical signals across
epithelial cells: ligand-stimulated transcytosis of the polymeric immunoglobulin
receptor requires two signals. Mol Biol Cell 10:1409-1427
Ma TY, Nguyen D, Bui V, Nguyen H, Hoa N (1999) Ethanol modulation of
intestinal epithelial tight junction barrier. Am J Physiol 276:G965-974
119
Machida M, Hayashi M, Awazu S (2000) The effects of absorption enhancers on the
pulmonary absorption of recombinant human granulocyte colony-stimulating factor
(rhG-CSF) in rats. Biol Pharm Bull 23:84-86
Marino M, Andrews D, Brown D, McCluskey RT (2001) Transcytosis of retinol-
binding protein across renal proximal tubule cells after megalin (gp 330)-mediated
endocytosis. J Am Soc Nephrol 12:637-648
Marino M, Zheng G, Chiovato L, Pinchera A, Brown D, Andrews D, McCluskey RT
(2000) Role of megalin (gp330) in transcytosis of thyroglobulin by thyroid cells. A
novel function in the control of thyroid hormone release. J Biol Chem 275:7125-
7137
Mason RJ, Leslie CC, McCormick-Shannon K, Deterding RR, Nakamura T, Rubin
JS, Shannon JM (1994) Hepatocyte growth factor is a growth factor for rat alveolar
type II cells. Am J Respir Cell Mol Biol 11:561-567
Mason RJ, Lewis MC, Edeen KE, McCormick-Shannon K, Nielsen LD, Shannon JM
(2002) Maintenance of surfactant protein A and D secretion by rat alveolar type II
cells in vitro. Am J Physiol Lung Cell Mol Physiol 282:L249-258
Mason RJ, Walker SR, Shields BA, Henson JE, Williams MC (1985) Identification
of rat alveolar type II epithelial cells with a tannic acid and polychrome stain. Am
Rev Respir Dis 131:786-788
Mathew S, Arandjelovic S, Beyer WF, Gonias SL, Pizzo SV (2003) Characterization
of the interaction between alpha2-macroglobulin and fibroblast growth factor-2: the
role of hydrophobic interactions. Biochem J 374:123-129
Matsukawa Y, Lee VH, Crandall ED, Kim KJ (1997) Size-dependent dextran
transport across rat alveolar epithelial cell monolayers. J Pharm Sci 86:305-309
Matsukawa Y, Yamahara H, Yamashita F, Lee VH, Crandall ED, Kim KJ (2000)
Rates of protein transport across rat alveolar epithelial cell monolayers. J Drug
Target 7:335-342

120
McConahey PJ, Dixon FJ (1980) Radioiodination of proteins by the use of the
chloramine-T method. Methods Enzymol 70:210-213
Mitra R, Pezron I, Li Y, Mitra AK (2001) Enhanced pulmonary delivery of insulin
by lung lavage fluid and phospholipids. Int J Pharm 217:25-31
Moestrup SK (1994) The alpha 2-macroglobulin receptor and epithelial
glycoprotein-330: two giant receptors mediating endocytosis of multiple ligands.
Biochim Biophys Acta 1197:197-213
Moestrup SK, Gliemann J (1991) Analysis of ligand recognition by the purified
alpha 2-macroglobulin receptor (low density lipoprotein receptor-related protein).
Evidence that high affinity of alpha 2-macroglobulin-proteinase complex is achieved
by binding to adjacent receptors. J Biol Chem 266:14011-14017
Moestrup SK, Gliemann J, Pallesen G (1992) Distribution of the alpha 2-
macroglobulin receptor/low density lipoprotein receptor-related protein in human
tissues. Cell Tissue Res 269:375-382
Morimoto K, Yamahara H, Lee VH, Kim KJ (1993) Dipeptide transport across rat
alveolar epithelial cell monolayers. Pharm Res 10:1668-1674
Morita T, Yamamoto A, Takakura Y, Hashida M, Sezaki H (1994) Improvement of
the pulmonary absorption of (Asu1,7)-eel calcitonin by various protease inhibitors in
rats. Pharm Res 11:909-913
Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F,
Christensen EI, Willnow TE (1999) An endocytic pathway essential for renal uptake
and activation of the steroid 25-(OH) vitamin D3. Cell 96:507-515
Osada T, Kuroda Y, Ikai A (1987) The fate of internalized alpha-2-macroglobulin:
alpha-galactosidase conjugate in fibroblasts from Fabry's hemizygote. Biochem
Biophys Res Commun 143:954-958
121
Osada T, Noro N, Kuroda Y, Ikai A (1988) Antibodies against viral proteins can be
produced effectively in response to the increased uptake of alpha 2-
macroglobulin:viral protein conjugate by macrophages. Biochem Biophys Res
Commun 150:883-889
Pan W, Kastin AJ, Zankel TC, van Kerkhof P, Terasaki T, Bu G (2004) Efficient
transfer of receptor-associated protein (RAP) across the blood-brain barrier. J Cell
Sci 117:5071-5078
Patel L (2007) Cationic cell penetrating peptides: Characterization of transport
properties in epithelial cell and their utilization as delivery systems for protein and
petide drugs. In: Pharmaceutical Sciences. Los Angeles, University of Southern
California, (Thesis).
Patton JS (1996) Mechanisms of macromolecule absorption by the lungs. Adv Drug
Del Rev 19:3-36
Patton JS, Byron PR (2007) Inhaling medicines: delivering drugs to the body through
the lungs. Nat Rev Drug Discov 6:67-74
Patton JS, Fishburn CS, Weers JG (2004) The lungs as a portal of entry for systemic
drug delivery. Proc Am Thorac Soc 1:338-344
Perkins DN, Pappin DJ, Creasy DM, Cottrell JS (1999) Probability-based protein
identification by searching sequence databases using mass spectrometry data.
Electrophoresis 20:3551-3567
Pizzo SV, Wu, S.M. (2001) Alpha-macroglobulins and kunins, Lippincott Williams
& Wilkins, Philadelphia
Reynolds HY (2000) Use of bronchoalveolar lavage in humans--past necessity and
future imperative. Lung 178:271-293
Rooney SA, Gobran LI, Umstead TM, Phelps DS (1993) Secretion of surfactant
protein A from rat type II pneumocytes. Am J Physiol 265:L586-590
122
Ross GF, Morris RE, Ciraolo G, Huelsman K, Bruno M, Whitsett JA, Baatz JE,
Korfhagen TR (1995) Surfactant protein A-polylysine conjugates for delivery of
DNA to airway cells in culture. Hum Gene Ther 6:31-40
Ross GF, Notter RH, Meuth J, Whitsett JA (1986) Phospholipid binding and
biophysical activity of pulmonary surfactant-associated protein (SAP)-35 and its
non-collagenous COOH-terminal domains. J Biol Chem 261:14283-14291
Saito A, Sinohara H (1985) Rat plasma murinoglobulin: isolation, characterization,
and comparison with rat alpha-1- and alpha-2-macroglobulins. J Biochem (Tokyo)
98:501-516
Sakagami M (2006) In vivo, in vitro and ex vivo models to assess pulmonary
absorption and disposition of inhaled therapeutics for systemic delivery. Adv Drug
Deliv Rev 58:1030-1060
Sakamoto N, Mukae H, Fujii T, Ishii H, Yoshioka S, Kakugawa T, Sugiyama K,
Mizuta Y, Kadota J, Nakazato M, Kohno S (2005) Differential effects of alpha- and
beta-defensin on cytokine production by cultured human bronchial epithelial cells.
Am J Physiol Lung Cell Mol Physiol 288:L508-513
Salvesen GS, Barrett AJ (1980) Covalent binding of proteinases in their reaction
with alpha 2-macroglobulin. Biochem J 187:695-701
Shah D, Shen WC (1995) The paradox of transferrin receptor-mediated drug
delivery--intracellular targeting or transcellular transport? J Drug Target 3:243-245
Shen Z, Zhang Q, Wei S, Nagai T (1999) Proteolytic enzymes as a limitation for
pulmonary absorption of insulin: in vitro and in vivo investigations. Int J Pharm
192:115-121
Sidobre S, Nigou J, Puzo G, Riviere M (2000) Lipoglycans are putative ligands for
the human pulmonary surfactant protein A attachment to mycobacteria. Critical role
of the lipids for lectin-carbohydrate recognition. J Biol Chem 275:2415-2422
123
Signor L, Tigani B, Beckmann N, Falchetto R, Stoeckli M (2004) Two-dimensional
electrophoresis protein profiling and identification in rat bronchoalveolar lavage
fluid following allergen and endotoxin challenge. Proteomics 4:2101-2110
Simionescu D, Simionescu M (1983) Differentiated distribution of the cell surface
charge on the alveolar-capillary unit. Characteristic paucity of anionic sites on the
air-blood barrier. Microvasc Res 25:85-100
Sottrup-Jensen L, Petersen TE, Magnusson S (1981) Mechanism of proteinase
complex formation with alpha 2-macroglobulin. Three modes of trypsin binding.
FEBS Lett 128:127-132
Sottrup-Jensen L, Sand O, Kristensen L, Fey GH (1989) The alpha-macroglobulin
bait region. Sequence diversity and localization of cleavage sites for proteinases in
five mammalian alpha-macroglobulins. J Biol Chem 264:15781-15789
Spiekermann GM, Finn PW, Ward ES, Dumont J, Dickinson BL, Blumberg RS,
Lencer WI (2002) Receptor-mediated immunoglobulin G transport across mucosal
barriers in adult life: functional expression of FcRn in the mammalian lung. J Exp
Med 196:303-310
Stefansson S, Kounnas MZ, Henkin J, Mallampalli RK, Chappell DA, Strickland
DK, Argraves WS (1995) gp330 on type II pneumocytes mediates endocytosis
leading to degradation of pro-urokinase, plasminogen activator inhibitor-1 and
urokinase-plasminogen activator inhibitor-1 complex. J Cell Sci 108 ( Pt 6):2361-
2368
Stevens PA, Wissel H, Zastrow S, Sieger D, Zimmer KP (2001) Surfactant protein A
and lipid are internalized via the coated-pit pathway by type II pneumocytes. Am J
Physiol Lung Cell Mol Physiol 280:L141-151
Takahashi H, Ishidoh K, Muno D, Ohwada A, Nukiwa T, Kominami E, Kira S
(1993) Cathepsin L activity is increased in alveolar macrophages and
bronchoalveolar lavage fluid of smokers. Am Rev Respir Dis 147:1562-1568
Thorley AJ, Goldstraw P, Young A, Tetley TD (2005) Primary human alveolar type
II epithelial cell CCL20 (macrophage inflammatory protein-3alpha)-induced
dendritic cell migration. Am J Respir Cell Mol Biol 32:262-267
124
Trapnell BC, Whitsett JA (2002) Gm-CSF regulates pulmonary surfactant
homeostasis and alveolar macrophage-mediated innate host defense. Annu Rev
Physiol 64:775-802
Tronde A, Norden B, Marchner H, Wendel AK, Lennernas H, Bengtsson UH (2003)
Pulmonary absorption rate and bioavailability of drugs in vivo in rats: structure-
absorption relationships and physicochemical profiling of inhaled drugs. J Pharm Sci
92:1216-1233
Watkins CA, Rannels DE (1979) In situ perfusion of rat lungs: stability and effects
of oxygen tension. J Appl Physiol 47:325-329
Wattiez R, Falmagne P (2005) Proteomics of bronchoalveolar lavage fluid. J
Chromatogr B Analyt Technol Biomed Life Sci 815:169-178
Wattiez R, Hermans C, Bernard A, Lesur O, Falmagne P (1999) Human
bronchoalveolar lavage fluid: two-dimensional gel electrophoresis, amino acid
microsequencing and identification of major proteins. Electrophoresis 20:1634-1645
Widera A, Kim KJ, Crandall ED, Shen WC (2003) Transcytosis of GCSF-transferrin
across rat alveolar epithelial cell monolayers. Pharm Res 20:1231-1238
Williams MC (2003) Alveolar type I cells: molecular phenotype and development.
Annu Rev Physiol 65:669-695
Williams MC (1984) Endocytosis in alveolar type II cells: effect of charge and size
of tracers. Proc Natl Acad Sci U S A 81:6054-6058
Wissel H, Lehfeldt A, Klein P, Muller T, Stevens PA (2001) Endocytosed SP-A and
surfactant lipids are sorted to different organelles in rat type II pneumocytes. Am J
Physiol Lung Cell Mol Physiol 281:L345-360
Wright JR, Wager RE, Hawgood S, Dobbs L, Clements JA (1987) Surfactant
apoprotein Mr = 26,000-36,000 enhances uptake of liposomes by type II cells. J Biol
Chem 262:2888-2894
125
Yamahara H, Lehr CM, Lee VHL, Kim KJ (1994a) Fate of insulin during transit
across rat alveolar epithelial cell monolayers. Eur J Pharm Biopharm 40:294-298
Yamahara H, Morimoto K, Lee VH, Kim KJ (1994b) Effects of protease inhibitors
on vasopressin transport across rat alveolar epithelial cell monolayers. Pharm Res
11:1617-1622
Yamamoto A, Umemori S, Muranishi S (1994) Absorption enhancement of
intrapulmonary administered insulin by various absorption enhancers and protease
inhibitors in rats. J Pharm Pharmacol 46:14-18
Zheng G, Bachinsky DR, Stamenkovic I, Strickland DK, Brown D, Andres G,
McCluskey RT (1994) Organ distribution in rats of two members of the low-density
lipoprotein receptor gene family, gp330 and LRP/alpha 2MR, and the receptor-
associated protein (RAP). J Histochem Cytochem 42:531-542
Zlokovic BV, Martel CL, Matsubara E, McComb JG, Zheng G, McCluskey RT,
Frangione B, Ghiso J (1996) Glycoprotein 330/megalin: probable role in receptor-
mediated transport of apolipoprotein J alone and in a complex with Alzheimer
disease amyloid beta at the blood-brain and blood-cerebrospinal fluid barriers. Proc
Natl Acad Sci U S A 93:4229-4234

Abstract (if available)

Abstract Pulmonary delivery of peptides and proteins has been studied for decades and the recent approval of Exubera® by the US and European markets, has made non-invasive delivery of these drug macromolecules through the lungs attainable. Despite the success of several proteins and peptides in achieving appreciable systemic bioavailability relative to other non-invasive routes of administration such as nasal and oral, the underlying mechanisms of absorption remain ambiguous. Certainly the large surface area, rich blood supply, and minimal proteolytic activity compared to the gastrointestinal tract, may explain the lung's efficiency in providing a suitable environment for drug absorption. However, it is by understanding the intricate cellular machinery provided by the alveolar epithelium (proposed site of systemic absorption), that further mechanisms of drug absorption may be unraveled. In this dissertation, we investigated the effects of conditioned medium obtained from cultured primary rat alveolar epithelial cell monolayers (CMII), and rat bronchoalveolar lavage fluid (BALF) on the transport of insulin across the alveolar epithelium. Our findings indicated that protein factor(s) present in both CMII and BALF, significantly enhanced the transport of insulin across rat alveolar epithelial cell monolayers (RAECM). The mechanism of enhancement was not due to paracellular leakage or fluid-phase endocytosis, but involved transcellular pathway(s), possibly receptor-mediated endocytosis. Upon partial purification of BALF, the factor(s) with most prominent effect for enhancing insulin transport across RAECM resided in the fraction containing high molecular weight proteins (>200 kDa). The proteins, which were identified by LC-MS/MS, belonged to the macroglobulin family. The mechanism afforded by macroglobulin partially explained the insulin transport-enhancing effect observed with BALF and CMII across RAECM suggesting that other factors may be involved.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cationic cell penetrating peptides: characterization of transport properties in epithelial cells and their utilization as delivery systems for protein and peptide drugs

PDF

Characterization of rice-derived proinsulin-transferrin fusion protein and its oral bioactivity as a basal insulin analogue

PDF

Characterization of the transferrin oligomer and its potential application in drug delivery

Asset Metadata

Creator Bahhady, Rana (author)

Core Title Characterization and identification of endogenous factor(s) that enhance insulin transport across primary rat alveolar epithelial cell monolayers

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 09/17/2007

Defense Date 07/19/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag insulin,OAI-PMH Harvest,pulmonary drug delivery

Language English

Advisor Shen, Wei-Chiang (committee chair), Kim, Kwang-Jin (committee member), Mircheff, Austin (committee member), Okamoto, Curtis Toshio (committee member)

Creator Email bahhady@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m815

Unique identifier UC1204072

Identifier etd-Bahhady-20070917 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-553332 (legacy record id),usctheses-m815 (legacy record id)

Legacy Identifier etd-Bahhady-20070917.pdf

Dmrecord 553332

Document Type Thesis

Rights Bahhady, Rana

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu