![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 14, Issue 7, 2959-2971, July 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Medicine, VA Palo Alto Health Care System, Palo Alto, California
94304, and the Digestive Disease Center, Stanford University School of
Medicine, Palo Alto, CA 94305;
German Cancer Research Center, Heidelberg, Germany
Submitted August 8, 2002;
Accepted February 28, 2003
Monitoring Editor: Keith Mostov
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
His (R80H) genomic (18 kb) and cDNA
K20 in cells and mice. Arg80 of K20 is conserved in most keratins, and its
mutation in epidermal keratins causes several skin diseases. R80H but not
wild-type K20 generates disrupted keratin filaments in transfected cells.
Transgenic mice that overexpress K20 R80H have collapsed filaments in small
intestinal villus regions, when expressed at moderate levels, whereas
wild-type K20-overexpressing mice have normal keratin networks. Overexpressed
K20 maintains its normal distribution in several tissues, but not in the
pancreas and stomach, without causing any tissue abnormalities. Hence, K20
pancreatic and gastric expression is regulated outside the 18-kb region.
Cross-breeding of wild-type or R80H K20 mice with mice that overexpress
wild-type K18 or K18 that is mutated at the conserved K20 Arg80-equivalent
residue show that K20 plays an additive and compensatory role with K18 in
maintaining keratin filament organization in the intestine. Our data suggest
the presence of unique regulatory domains for pancreatic and gastric K20
expression and support a significant role for K20 in maintaining keratin
filaments in intestinal epithelia. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
;
Ku et al., 1999). The
so-called "soft" keratins (i.e., those excluding the "hard'
keratins found in epidermal appendages such as hair) make up the IF proteins
of epithelial cells. Soft keratins (K) consist of >20 members (K1–20)
that include types I (K9-K20) and II (K1-K8) IF proteins. All epithelial cells
express one or more type I and type II keratins as noncovalent heteropolymers.
Regardless of the number of keratins found in a given epithelial cell, the
ratio of type I to type II keratins is 1:1 (reviewed by
Moll et al., 1982;
Fuchs and Weber, 1994;
Herrmann and Aebi, 2000;
Coulombe and Omary, 2002). For
example, hepatocytes express K8/18 exclusively, whereas intestinal epithelial
cells express K7/8 (type II) and K18/19/20 (type I) at variable levels.
Because cell culture models have not been revealing in understanding
keratin function, transgenic animal models were generated for several keratins
as a handle toward understanding their function. These models provided clear
evidence for keratins as serving an essential role in protecting cells from a
variety of mechanical and nonmechanical forms of stress, particularly in the
skin and liver (Fuchs and Weber,
1994; Magin et al.,
2000; Coulombe and Omary,
2002; Omary et al.,
2002; Oshima,
2002). The disease phenotypes of these mouse models also led to
the identification of several human diseases as keratin-associated diseases
(Fuchs and Cleveland, 1998;
Omary et al., 2002).
For example, transgenic mice (R89C mice) that overexpress human (h) K18 that
is mutated at a highly conserved Arg (Arg89Cys; R89C) develop mild
chronic hepatitis and a marked predisposition to drug-induced liver injury, in
association with hepatocyte and pancreatic acinar cell keratin filament
disruption (Ku et al.,
1995,
1996; Toivola et al.,
1998,
2000a;
Omary et al., 2002).
Arg89 of K18 was chosen because it is mutated in K14 and K5 (the equivalent
residue is Lys in some type II keratins) in some patients with the severe form
of the blistering skin disorder epidermolysis bullosa simplex
(Fuchs and Coulombe, 1992).
The phenotype of the R89C mice led to the search and subsequent identification
of K8/18 mutations in patients with cryptogenic forms of liver disease (Ku
et al., 1997,
2001). Dramatic keratin
filament collapse in R89C mice was noted in the liver and pancreas but not in
the small or large intestine, likely due to the compensatory presence of other
type I keratins (K19 and K20) that are either absent in the liver
(Ku et al., 1995) or
present in small amounts in the pancreas
(Toivola et al.,
2000b).
We focused on studying K20, which is the most recently identified keratin
protein (Moll et al.,
1990,
1993;
Quaroni et al., 1991;
Calnek and Quaroni, 1993).
Human K20 has several unique properties, including its relatively greater
sequence divergence compared with other type I keratins, with only 58%
identical amino acids in the conserved 
-helical "rod"
domain with the closest keratin, K14; unique distribution that is nearly
confined to the gastric and intestinal epithelium, urothelium, bladder, and
Merkel cells; expression in the more differentiated enterocytes along the
crypt-villus differentiation axis; and lack of immunological cross-reaction of
most anti-keratin antibodies (Moll et
al., 1993). Human K20 has 79% amino acid identity with rat
K20 (Chandler et al.,
1993). Although nothing is known regarding K20 function, it has
been used as a marker for the histological classification of tumors
(Moll et al., 1992;
Harnden et al., 1999;
Wildi et al., 1999;
Leech et al., 2001).
The observation that K20 is coexpressed in the same intestinal cell type as
other type I keratins, including K18 and K19 suggest that these three type I
keratins may have redundant or complementary functions.
In this study, we generated transgenic mice that overexpress wild-type (WT)
and ArgHis (R80H) K20 (corresponding to the conserved Arg89 in hK18) to
begin addressing K20 function in digestive epithelia. We focused on the highly
conserved Arg because it has provided a useful target in studying K18 (Ku
et al., 1995,
1996).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
), which recognizes human K18 without cross-reacting with
mouse keratins; rat mAbs Troma-I (Developmental Studies Hybridoma Bank,
University of Iowa, Iowa City, IA), Troma-II and -III [kindly provided by Drs
Helene Baribault at Tularik (South San Francisco, CA) and Thomas Magin at the
University of Bonn (Bonn, Germany), respectively], which recognize mouse K8,
K18, at Tularik (South San Fracnisco, CA) and K19, respectively; rabbit
antibody 4668, which is an anti-K18 peptide antibody (raised against the
peptide RPVSSAASVYAGA; G; Ku et
al., 1998) that recognizes human and mouse K18; mouse mAb
LF6, which recognizes human K8 and is generated by immunizing mice with high
salt extracts of human K8/18 isolated from HT29 cells (our unpublished
observations); mouse anti-K20 mAb (clone Ks20.8), which recognizes mouse and
human K20 and anti-chromogranin A mAb (NeoMarkers, Fremont, CA); mouse
anti-K20 mAb (clone ITKs20.10), which recognizes human but not mouse K20 and
mouse anti-K7 mAb (RCK105), which recognizes mouse and human K7 (Progen;
Heidelberg, Germany); and rabbit antibody 8592
(Ku et al., 1995),
which recognizes human and mouse K8/18.
Transgene Construct and Generation of Transgenic Lines
The human K20 genomic DNA (18 kb) and cDNA (1.35 kb) were mutated at codon
CGTCAT to convert Arg80His. The mutation was first introduced into
K20 cDNA by using a QuikChange site-directed mutagenesis kit (Stratagene, La
Jolla, CA) and was confirmed by DNA sequencing. For the 18-kb genomic DNA, an
XhoI fragment (15 kb) was first subcloned in pBluescript SK+ and then
an internal BstI fragment (8 kb) was cut out to generate a workable
DNA fragment (10 kb, including the vector) to introduce the mutation. After
mutation, the DNA was religated to regenerate the 18-kb gene. Mutant and WT
K20 genes were injected into pronuclei of fertilized FVB/n eggs. Four
heterozygous K20 transgenic lines that manifest germline expression of WT (WT1
and WT2) and K20 R80H (M1 and M2) K20 were selected. The transgene copy
numbers were estimated using tail vein DNA of test animals and a densitometric
standard curve generated using a plasmid that expresses genomic human K20, as
described previously (Ku et al.,
1995). Presence of the transgene was verified by polymerase chain
reaction screening with hK20-specific primers that generated a 550-base pair
fragment corresponding to the first intron of the K20 gene: (+)
5'CCGGAATAAAGTTCATCCTCTG-3' and (-)
5'-GACATTTCTTTTAGCCCCTGTG-3'. The PCR conditions were 94°C for
2 min and then 35 cycles at 94°C for 1 min, 55°C for 1 min, and
72°C for 1 min. The WT1, WT2, M1, and M2 heterozygous K20-overexpressing
lines were also cross-bred with homozygous transgenic mice that overexpress WT
hK18 (TG2 mice) or hK18 R89C mice (F22 and F30 lines) that were previously
described in detail (Ku et al.,
1995,
1996; Toivola et al.,
1998,
2000a).
Sequencing of Mouse K20 and Generation of Antibodies
A database search
(www.ncbi.nlm.nih.gov),
by using the human K20 cDNA sequence, indicated the presence of an expressed
sequence tag clone that likely represents mouse K20. This clone was purchased
(American Type Culture Collection), sequenced, and identified as a full-length
mouse K20 cDNA. The sequence was deposited in GenBank as "MuK20"
no. AF473907
[GenBank]
. The sequence of mouse K20, in alignment with rat and human K20,
is shown in Table 1. MuK20 was
subcloned into a PET23A(+) bacterial expression system, followed by partial
purification of the mouse K20 protein by using HiTrap Q and SP columns
(Amersham Biosciences AB, Uppsala, Sweden). The partially purified protein was
used to immunize BALB/c mice followed by generation of mAbs by using standard
procedures. One clone that recognized mouse but not human K20, termed XQ1, was
further subcloned and used for immunoblotting.
|
Transfection and Immunofluorescence Staining
The genomic and cDNA constructs of hK20 were cotransfected with hK8 cDNA,
by using LipofectAMINE (Invitrogen, Carlsbad, CA), into BHK-21 cells (to allow
relatively high yield biochemical extraction of keratins) or NIH-3T3 cells
(for cell staining). NIH-3T3 cells were cultured on chamber slides for
2–3 d after transfection and then fixed in methanol (3 min, -20°C).
Fixed cells were "blocked" for 15 min with phosphate-buffered
saline (PBS) containing 2% bovine serum albumin (buffer B) and then
coincubated with anti-K20 mAb in buffer B (30 min). After washing three times
with PBS, cells were blocked with buffer B containing 2% normal goat serum (15
min) followed by incubating with Texas Red-conjugated goat anti-mouse antibody
for 20 min and then washing with PBS.
Mouse tissues were frozen in optimum cutting temperature compound, sectioned and fixed in acetone (-20°C, 10 min) and then stained with mouse or rabbit anti-keratin antibodies as described above. For double staining, mouse and rabbit primary antibodies were used, followed by Texas Red and fluorescein isothiocyanate-conjugated secondary antibodies. Images of single confocal sections of stained cells and tissues were obtained with a Nikon TE300 microscope coupled to a Bio-Rad MRC1024ES confocal microscope.
Keratin Isolation, Western and Northern Blotting
High salt extraction (HSE) was used to isolate keratins from cultured
transfected cells or from tissues of transgenic and nontransgenic mice
(Ku et al., 1995).
SDS-PAGE was done using 8 or 10% acrylamide gels. Immunoblotting was carried
out after transferring HSE to polyvinylidene difluoride membranes followed by
blocking with 5% nonfat dry milk and then incubating with anti-keratin
antibodies, washing, incubating with peroxidase-conjugated goat anti-mouse,
anti-rabbit, or anti-rat immunoglobulins. Immune-reactive bands were
visualized using enhanced chemiluminescence (PerkinElmer Life Sciences,
Boston, MA).
Northern blot analysis was done with total tissue RNA isolated from the small intestine and colon of various mouse lines by using an RNeasy Midi kit (QIAGEN, Chatsworth, CA). Total RNA (20 µg/lane) was loaded then separated using 1% denaturing formaldehyde gels and transferred to nylon membranes. Blots were UV cross-linked, prehybridized for 2 h, and then hybridized with 32P-labeled cDNA probes for hK20, mouse K20, and mouse glyceraldehyde-3-phosphate dehydrogenase (overnight, 68°C). The blots were then washed and visualized by autoradiography.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
|
|
|
Effect of the K20 Arg80His (R80H) Mutation on Keratin Filament
Assembly in Transfected Cells
We targeted Arg80 of K20 for mutation, given the conserved nature of this
arginine among IF proteins and the significant impact of its mutation in
causing several skin diseases involving epidermal keratins
(Fuchs and Coulombe, 1992).
Furthermore, mutation of the equivalent residue in K18 (Arg89Cys) and
expression in transfected cells, or in mice as a transgene, results in keratin
filament collapse (Ku et al.,
1995). We generated a K20 Arg80His mutation, and chose His
because an Arg-to-His mutation entailed introducing a single nucleotide
substitution (CGTCAT) and because the equivalent K14 residue has a
similar mutation (Arg125His) in patients with epidermolysis bullosa
simplex (Coulombe et al.,
1991).
We used cDNA (c) and genomic (g) human K20 constructs to generate the K20 R80H mutation (Figure 4). Transfection of various combinations of cK8/cK18, cK8/cK20, or cK8/gK20 (WT or R80H K20) into baby hamster kidney (BHK) cells showed that the transfected proteins can be detected by immunoblotting of HSE of the transfected cell and confirmed the predicted reactivity of the keratin-specific antibodies (Figure 4A). Immunofluorescence staining of NIH-3T3 cells cotransfected with WT K8 and either WT or R80H K20 (genomic and cDNA) showed that WT K20 generates filaments, whereas R80H K20 (both genomic and cDNA) result in collapse of the keratin filaments into distinct dots (Figure 4B). Thus, the R80H residue of K20 is critical for keratin filament organization in cultured cells.
|
Characterization of Transgenic Mice that Overexpress Human WT and
R80H K20
We generated four K20 transgenic mouse lines, two that overexpress WT hK20
(WT1 and WT2) and two that overexpress hK20 R80H (M1 and M2). The transgene
copy number was 30 and 35 for WT1 and WT2, and 2 and 20 for M1 and M2,
respectively. hK20 mRNA (Figure
5A) and protein (Figure
5B) expression in the small intestine and colon of the transgenic
lines were confirmed. As expected, hK20 was not detected in control
nontransgenic mice (Figure 5A, lanes 5 and 10; B, lanes 1 and 6), whereas endogenous mouse keratins were
detected in all the mouse lines (Figure 5,
A and B). The hK20 protein is present at low levels in the small
intestine and colon of M1 mice compared with the other transgenic mouse lines
(Figure 5B), in a manner that
parallels the transgene copy number.
We used high salt extraction, which provides a near quantitative recovery
of the total keratin pool, to assess the relative content of individual
keratins in the small intestine of the various mouse lines. As shown in
Figure 2A, the K20 transgene
protein expression levels were WT1 
WT2 > M2 >> M1, which
parallel the K20 transgene copy number (hK20 expression in the M1 line is
difficult to discern by Coomassie staining). Notably, K20 overexpression
results in down-regulation of K19 steady-state protein levels
(Figure 2A, compare lane 3 with
lane 4 or 5). The specific bands (for one-dimensional gels,
Figure 2A) and spots (for
two-dimensional gels, Figure
2B) were assigned to individual keratins by using two-dimensional
gel analysis (Figure 2B) and
immune blotting of the one- and two-dimensional gels with IF-specific
antibodies (our unpublished data).
We then compared the tissue distribution of hK20 with endogenous mK20 by blotting with antibody's specific to mouse and human K20. Due to unavailability of antibodies that adequately recognize mouse K20 by immunoblotting, we generated a mAb (termed XQ1) that recognizes mouse but not human K20 (see MATERIALS AND METHODS). For this, we identified and sequenced a cDNA clone that corresponds to mK20 (Table 1). As shown in Figure 2C, hK20 maintained its expression in various segments of the small intestine (ileum, jejunum, and duodenum) and colon [right (ascending) and left (descending) colon]. The K20 transgene product also maintained its tissue/cell expression in the bladder (Figure 6, g and l) and in the scarcely found Merkel cells of the skin (Figure 6, j and o). The K20 transgene product was absent from the liver, as expected (Figure 2C, lane 7; confirmed by immune staining; our unpublished data). However, hK20 was also absent from mouse stomach and pancreas in contrast to its endogenous mouse K20 counterpart (Figures 2C for stomach and 6 for stomach and pancreas; pancreatic mK20 levels are too low to detect by Western blotting). This finding suggests that the regulatory elements that direct gastric and pancreatic expression of mK20 are likely to be outside of the 18-kb hK20 genomic construct that we used and/or that mK20 transcriptional machinery does not recognize the corresponding human sequences.
|
Effect of the R80H K20 Mutation on Keratin Filament Organization in
Tissues
We examined keratin filament organization in tissues that overexpressed the
R80H K20 mutant compared with similar tissues that overexpressed WT hK20.
Mutant and WT hK20 integrate into the existing endogenous mouse keratin
filament networks, which is detectable in all the tissues examined
(Figure 6). There was some
mosaicism in hK20 expression, particularly in the colon, in that not all
epithelial tip cells uniformly expressed K20 (our unpublished data). This
mosaic expression phenomenon was observed in other keratin-overexpressing
transgenic mouse models such as those that overexpress mK19
(Bader and Franke, 1990).
Tissues expressing WT K20 in the WT1 and WT2 lines had normal-looking
filaments in all cases tested (Figure
7, a, b, e, f, i, and j; our unpublished data). However, in M2
mice, the keratin filaments collapsed into cytoplasmic dots and thick cables
as noted in the ileum (Figure
7, g and h) and bladder (our unpublished data). Filament
disruption was not seen in the ileum
(Figure 7, k and l) or bladder
(our unpublished data) of M1 mice likely due to lower hK20 expression level in
this line, which suggests that the K20 R80H mutation behaves in a
dominant-negative manner when overexpressed at moderate levels. Combined with
the cell transfection experiments of Figure
4, our results show that the R80H mutation causes keratin filament
disruption in vitro and in vivo. There was no obvious effect on tissue
histology after testing all organs that expressed WT or R80H hK20 in WT1/2 and
M1/2 transgenic mice (our unpublished data).
|
Transgenic Mouse Cross-Breeding Demonstrates Reversible Rescue or
Collapse of Keratin Filaments in the Small Intestine, Depending on the Dosage
of WT versus Mutant K18 and K20
We tested the functional redundancy of K18 and K20 at the level of filament
organization by breeding mice that overexpress WT hK18 (TG2 line) or that
overexpress hK18 R89C (F22 and F30 lines), with the K20 mouse lines WT2 (which
is similar to WT1) or M1 (which overexpresses low levels of K20). We chose for
the initial breeding the M1 rather than the M2 line because M1 mice have
normal-looking keratin filaments in the small intestine. As reported
previously (Ku et al.,
1995), WT K18-overexpressing mice (TG2) have normal-looking
keratin filaments throughout their intestine (our unpublished data), whereas
F22 (Figure 8A, e) and F30 (our
unpublished data) have normal keratin filaments in the majority of epithelial
cells but are found collapsed into cytoplasmic dots and thick cables in goblet
cell of the small intestine (Figure
8A,e), likely related to the strong K18 staining in these cells
(Figure 1e). Interestingly,
keratin filament collapse occurs only in the F30 x M1 but not in the F22
x M1 ileums, nor in the ileums (or colons; our unpublished data) of the
TG2 x WT2 or F30 x WT2 cross-breeds
(Figure 9A). Analysis of the
individual keratins in the small intestine of these cross-bred mice was also
carried out. In TG2 and F30 ileums, the relative K18 levels were h > m and
h m, respectively (Figure
9B, lanes 1 and 2). These levels were not significantly affected
after cross-breeding with WT1 or M1 mice
(Figure 9B, lanes 3–6).
In the F30 x M1 mouse intestine, keratin filament disruption
(Figure 9A) is likely caused by
the addition of a very small amount of K20 that is barely perceptible by
Coomassie staining.
|
|
The above-mentioned analysis was extended by mating M2 K20 mice with F22 or F30 mice to test for an additive K18 + K20 mutational effect and by TG2 x M2 and F22 x WT1 matings to test for rescue of the mutant K18 R89C or K20 R80H by wild-type K20 or K18 overexpression, respectively. When hK18 WT (TG2) and R89C (F22 or F30) mice are mated with M2 mice, the filament network in small intestinal epithelia of TG2 x M2 mice nearly normalize compared with M2 mice (compare Figure 8A, a and b). This indicates that wild-type hK18 compensates for mutant hK20 in filament organization and thus rescues the M2 phenotype. Similarly, the goblet cell phenotype of F22 x WT1 mice (Figure 8A, e) is also rescued when wild-type hK20 is overexpressed (Figure 8A, f). The filament disruption noted in F22 x M2 is similar to that seen in F30 x M2 (Figure 8A, c and d). Confirmation of the expression of hK18 and hK20 proteins in the various intercrosses was demonstrated by keratin-specific immune blotting as shown in Figure 8B. A schematic summary of the filament organization phenotype of these interbreedings is shown in Figure 10. Together, our results indicate that K20 and K18 are functionally redundant at the level of keratin filament assembly in the intestine.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
;
O'Guin et al., 1987;
Takemura et al.,
1988), which is schematically summarized in
Figure 3. The major points that
are likely to be biologically relevant are: 1) K18 and K20 are
differentiation-specific keratins in the small intestine. For example, K20 is
primarily expressed in the more differentiated villus cells, whereas K18 is
nearly absent in many of these cells (except for goblet cells) and is
preferentially found in basal and midzonal regions of the epithelium. 2) K18
and K20 show significant differences in their distribution patterns in the
small and large intestine. In that regard, K20 is restricted to a more
pronounced epithelial tip distribution in the colon, whereas K18 is more
abundant and less restricted in its colonic distribution, compared with the
small intestine. In contrast, K8 and K19 are found nearly uniformly throughout
the epithelium of the small intestine and colon. K20 is more abundant in the
small versus the large intestine, which provides a likely explanation for the
dominant negative effect of the R80H mutation in the ileum but not in the
colon. 3) There seems to be mouse and human species differences in K20
expression. For example, K20 expression in mouse small intestine is most
prominent in the ileum and, even so, is still relatively minor compared with
K19. In contrast, K20 is a major keratin in human small intestine,
particularly in the duodenum (Moll et
al., 1990).
Targeting the Conserved Arginine of IF Proteins to Study Keratin and
Other IF Protein Biology
In this study, Arg80 of K20 was targeted for mutation in a similar manner
to the previous targeting of the same conserved K18 arginine (Arg89), which is
located within the N-terminal aspect of helix 1A of the rod domain. This
arginine is conserved in most IF proteins, including lamins, and across
species, including snail IF proteins
(Fuchs and Coulombe, 1992).
Mutation of this conserved arginine occurs in several epidermal keratins and
results in a severe phenotype (Coulombe
et al., 1991; Fuchs
and Weber, 1994). Its targeting to generate an animal model to
study keratin function and potential disease association was first described
in transgenic mice that overexpressed K18 R89C. Mutation of the conserved
arginine causes keratin filament disassembly in cultured cells that express
K18 (Ku et al., 1995)
or K20 (Figure 4) and in
tissues of transgenic mice that express K18
(Ku et al., 1995) or
K20 (Figure 7). Collectively,
the findings in this study that defined an in vivo role for K20 in keratin
filament organization and the findings of the multiple studies that examined
overexpression of K18 R89C in transgenic mice (reviewed by
Omary et al., 2002)
predict that the approach of overexpressing IF proteins with mutations that
target this conserved arginine are likely to be fruitful.
Phenotype of Transgenic Mice That Overexpress Human K20 R80H
The major phenotype of the R80H K20 transgenic mice is a filament
organization defect in enterocytes of the small intestine, which parallels the
expression level of the mutant K20 protein. This defect occurs in the ileum of
M2 but not M1 mice that, in turn, express lower K20 levels, and does not occur
in the colon of M2 mice likely due to the higher relative level of K18 and the
lower relative level of K20 in colonocytes. Disruption of keratin filaments
did not result in a histological or disease-like phenotype based on several
tests that we carried out. For example, the stool weight in WT1 and M2 mice
was similar and both mouse strains were equally susceptible to
rotavirus-induced infection and had similar virus shedding profiles (our
unpublished data). The dichotomy between keratin filament disruption and the
presence or lack of a physiological effect on the involved cell/tissue under
basal or stress conditions seems to depend on the affected cell/tissue type.
For example, mice that express K18 R89C have disrupted keratin filaments in
hepatocytes and pancreatic acinar cells. Despite similar keratin filament
disruption in both cell types, the liver becomes highly susceptible to injury,
whereas the pancreas remains relatively resilient
(Toivola et al.,
2000a). Lack of an obvious physiological/histologic phenotype in
M2 (or F30 x M2) mice may relate to: 1) the observation that keratin
filament disruption did not involve the entire epithelium, 2) the presence of
gene modifiers, or 3) the possibility that a given keratin may serve different
functions depending on the cell/tissue location of its expression.
The 18-kb genomic construct we used to generate the WT1/2 and M1/2
transgenic lines faithfully maintained K20 expression in the small and large
intestine, bladder, and Merkel cells. Regional distribution of the K20
transgene was also maintained (i.e., surface versus base of the epithelium).
Two organs that did not maintain hK20 expression were the stomach and
pancreas. Expression of K20 in gastric epithelia is well established
(Moll et al., 1990),
whereas expression in the pancreas depends on the species and the experimental
conditions. For example, several studies report the presence of K20 in rat
ductal cells (Moll et al.,
1990; Bouwens,
1998), whereas ductal expression of mouse K20 is detectable at
very low levels (Figure 6) but
is induced in an autoimmune diabetes model
(Anastasi et al. 1999)
or after caerulein-induced pancreatitis (our unpublished observations). We did
not observe pancreatic hK20 transgene induction after caerulein-induced
pancreatitis (our unpublished data) or under basal conditions
(Figure 6). Therefore, it is
likely that the regulatory elements that control K20 expression in the stomach
and pancreas are outside the 18-kb genomic region that we used in this study.
This differs from K18 whereby a genomic sequence of 10-kb contained all the
necessary elements for normal tissue specific expression in simple epithelia
(Abe and Oshima, 1990).
However, we cannot exclude the possibility that gastric/pancreatic regulatory
factors may not recognize the human transgene.
Functional Redundancy of K18 and K20 at the Level of Keratin Filament
Organization
The dominant negative filament organization phenotype noted in small
intestinal enterocytes of the M2 transgenic line supports an in vivo role for
K20 in keratin filament organization. This role is further substantiated by
the intermixed cross-breeding of the transgenic mice that overexpress
wild-type K18 or K20 or mutant K18 or K20 (summarized in
Figure 10). Hence, wild-type
K18 rescues mutant K20 and vice versa, and the effects of the K20 and K18
mutations are additive in the same cell in terms of their filament disruptive
capacity. This provides in vivo evidence that K18 and K20 serve redundant
functions in terms of keratin filament organization in the intestine. In
addition, our findings provide an explanation for a K20-mediated sparing role
of keratin filament organization in the majority of transgenic mouse
enterocytes that overexpress K18 R89C (Ku
et al., 1995), although K19 is likely to play a bigger
"sparing" role given its abundance as the major type I keratin in
the intestine and its distribution throughout the intestinal epithelium
(Figures 1,
2,
3). Previous in vitro studies
showed that K18, compared with K20, is a preferred partner for binding with K8
but K20 and K8 do associate and form filaments
(Hofmann and Franke 1997) as
confirmed in this study in transfected cells in culture
(Figure 4) and by
colocalization of filaments containing K20 and K18 in tissues
(Figure 6). Availability of
transgenic mice that overexpress wild-type and mutant K20 should provide
useful in vivo models to study K20 function and regulation.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
Footnotes |
|---|
Corresponding author. E-mail address:
mbishr{at}stanford.edu.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Anastasi, E., Ponte, E., Gradini, R., Bulotta, A., Sale, P., Tiberti, C., Okamoto, H., Dotta, F., and Mario, U.D. (1999). Expression of Reg and cytokeratin 20 during ductal cell differentiation and proliferation in a mouse model of autoimmune diabetes. Eur. J. Endocrinol. 141, 644-652.[Abstract]
Bader, B.L., and Franke, W.W. (1990). Cell type-specific and efficient synthesis of human cytokeratin 19 in transgenic mice. Differentiation 45, 109-118.[CrossRef][Medline]
Bouwens, L. (1998). Cytokeratins and cell differentiation in the pancreas. J. Pathol. 184, 234-239.[CrossRef][Medline]
Calnek, D., and Quaroni, A. (1993). Differential localization by in situ hybridization of distinct keratin mRNA species during intestinal epithelial cell development and differentiation. Differentiation 53, 95-104.[CrossRef][Medline]
Chandler, J.S., Calnek, D., and Quaroni, A. (1993). Identification and characterization of rat intestinal keratins. Molecular cloning of cDNAs encoding cytokeratin 8, 19, and a new 49-KDa type I cytokeratin (cytokeratin 21) expressed by differentiated intestinal epithelial cells. J. Biol. Chem. 266, 11932-11938.
Chou, C.F., Riopel, C.L., Rott, L.S., and Omary, M.B. (1993). A significant soluble keratin fraction in "simple epithelial cells: lack of an apparent phosphorylation and glycosylation role in keratin solubility. J. Cell Sci. 105, 433-445.[Abstract]
Coulombe, P.A., Hutton, M.E., Letai, A., Hebert, A., Paller, A.S., and Fuchs, E. (1991). Point mutations in human14 genes of epidermolysis bullosa simplex patients: genetic and functional analyses. Cell 66, 1301-1311.[CrossRef][Medline]
Coulombe, P.A., and Omary, M.B. (2002). "Hard" and "soft" principles defining the structure, function and regulation of keratin intermediate filaments. Curr. Opin. Cell Biol. 14, 110-122.[CrossRef][Medline]
Fuchs, E., and Coulombe, P.A. (1992). Of mice and men: genetic skin diseases of keratin. Cell 69, 899-902.[CrossRef][Medline]
Fuchs, E., and Weber, K. (1994). Intermediate filaments: structure, dynamics, function and disease. Annu. Rev. Genet. 63, 345-382.
Fuchs, E., and Cleveland, D.W. (1998). A structural
scaffolding of intermediate filaments in health and disease.
Science 279,
514-519.
Franke, W.W., Mahmood, N., Appelhans, B., Schmid, E., Freudenstein, C., Osborn, M., and Weber, K. (1979). The organization of cytokeratin filaments in the intestinal epithelium. Eur. J. Cell Biol. 19, 255-268.[Medline]
Harnden, P., Mahmood, N., and Southgate, J. (1999). Expression of cytokeratin 20 redefines urothelial papillomas of the bladder. The Lancet 353, 974-977.[CrossRef][Medline]
Herrmann, H., and Aebi, U. (2000). Intermediate filaments and their associates: multi-talented structural elements specifying cytoarchitecture and cytodynamics. Curr. Opin. Cell Biol. 12, 79-90.[CrossRef][Medline]
Hofmann, I., and Franke, W.W. (1997). Heterotypic interactions and filament assembly of type I and type II cytokeratins in vitro: viscometry and determination of relative affinities. Eur. J. Cell Biol. 72, 122-132.[Medline]
Ku, N.-O., Gish, R., Wright, T.L., and Omary, M.B.
(2001). Keratin 8 mutations in patients with cryptogenic liver
disease. N. Engl. J. Med. 344,
1580-1587.
Ku, N.-O., Liao, J., and Omary, M.B. (1998). Phosphorylation of human keratin 18 serine-33 regulates binding to 14–3-3 proteins. EMBO J. 17, 1892-1906.[CrossRef][Medline]
Ku, N.-O., Michie, S.A., Oshima, R.G., and Omary, M.B.
(1995). Chronic hepatitis, hepatocyte fragility and increased
soluble phosphoglycokeratins in transgenic mice expressing a keratin 18
conserved arginine mutant. J. Cell Biol.
131,
1303-1314.
Ku, N.-O., Michie, S.A., Soetikno, R.M., Resurreccion, E.Z., Broome, R.L., Oshima, R.G., and Omary, M.B. (1996). Susceptibility to hepatotoxicity in transgenic mice that express a dominant-negative human keratin18 mutant. J. Clin. Investig. 98, 1034-1046.[Medline]
Ku, N.-O., Wright, T.L., Terrault, N.A., Gish, R., and Omary, M.B. (1997). Mutation of human keratin 18 in association with cryptogenic cirrhosis. J. Clin. Investig. 99, 19-23.[Medline]
Ku, N.-O., Zhou, X., Toivola, D.M., and Omary, M.B. (1999). The cytoskeleton of digestive epithelia in health and disease. Am. J. Physiol. 277, G1108-G1137.
Leech, S.N., Kolar, A.J., Barrett, P.D., Sinclair, S.A., and
Leonard, N. (2001). Merkel cell carcinoma can be distinguished
from metastatic small cell carcinoma using antibodies to cytokeratin 20 and
thyroid transcription factor 1. J. Clin. Pathol.
54,
727-729.
Magin, T.M., Hesse, M., and Schroder, R. (2000). Novel insights into intermediate-filament function from studies of transgenic and knockout mice. Protoplasma 211, 140-150.[CrossRef]
Moll, R., Franke, W.W., Schiller, D.L., Geiger, B., and Krepler, R. (1982). The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 31, 11-24.[CrossRef][Medline]
Moll, R., Lowe, A., Laufer, J., and Franke, W.W. (1992). Cytokeratin 20 in human carcinomas. A new histodiagnostic marker detected by monoclonal antibodies. Am. J. Pathol. 140, 427-447.[Abstract]
Moll, R., Schiller, D., and Franke, W.W. (1990).
Identification of protein IT of the intestinal cytoskeleton as a novel type I
cytokeratin with unusual properties and expression patterns. J. Cell
Biol. 111,
567-580.
Moll, R., Zimbelmann, R., Golsdchmidt, M., Keith, M., Laufer, J., Kasper, M., Koch, P., and Franke, W.W. (1993). The human gene encoding cytokeratin 20 and its expression during fetal development and in gastrointestinal carcinomas. Differentiation 53, 75-93.[Medline]
O'Guin, M., Galvin, S., Schermer, A., and Sun, T.-T. (1987). Pattern of keratin expression defines distinct pathways of epithelial development and differentiation. Curr. Top. Dev. Biol. 22, 97-125.[Medline]
Omary, M.B., Ku, N.-O., and Toivola, D.M. (2002). Keratins: guardians of the liver. Hepatology 35, 251-257.[CrossRef][Medline]
Oshima, R.G. (2002). Apoptosis and keratin intermediate filaments. Cell Death Differ. 9, 486-492.[CrossRef][Medline]
Quaroni, A., Calnek, D., Quaroni, E., and Chandler, J.S.
(1991). Keratin expression in rat intestinal crypt and villus
cells. J. Biol. Chem. 266,
11923-11931.
Takemura, R., Masaki, T., and Hirokawa, N. (1988). Developmental organization of the intestinal brush-border cytoskeleton. Cell Motil. Cytoskeketon 9, 299-311.[CrossRef][Medline]
Toivola, D.M., Omary, M.B., Ku, N.-O., Peltola, O., Baribault, H., and Eriksson, J.E. (1998). Protein phosphatase inhibition in normal and keratin 8/18 assembly-incompetent mouse strains supports a functional role of keratin intermediate filaments in preserving hepatocyte integrity. Hepatology 28, 116-128.[CrossRef][Medline]
Toivola, D.M., Ku, N.-O., Ghori, N., Lowe, A.W., Michie, S.A., and Omary, M.B. (2000a). Effects of keratin filament disruption on exocrine pancreas-stimulated secretion and susceptibility to injury. Exp. Cell Res. 255, 156-170.[CrossRef][Medline]
Toivola, D.M., Baribault, H., Magin, T.M., Michie, S.A., and Omary, M.B. (2000b). Simple epithelial keratins are dispensable for cytoprotection in two pancreatitis models. Am. J. Physiol. 279, G1343-1354.
Wildi, S., Kleeff, J., Maruyama, H., Maurer, C.A., Friess, H., Buchler, M.W., Lander, A.D., and Korc, M. (1999). Characterization of cytokeratin 20 expression in pancreatic and colorectal cancer. Clin. Cancer Res. 10, 2840-2847.
This article has been cited by other articles:
|
|
 |
|
  D. M. Toivola, I. Nakamichi, P. Strnad, S. A. Michie, N. Ghori, M. Harada, K. Zeh, R. G. Oshima, H. Baribault, and M. B. Omary Keratin Overexpression Levels Correlate with the Extent of Spontaneous Pancreatic Injury Am. J. Pathol., April 1, 2008; 172(4): 882 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Zhong, P. Strnad, D. M. Toivola, G.-Z. Tao, X. Ji, H. B. Greenberg, and M. B. Omary Reg-II Is an Exocrine Pancreas Injury-Response Product That Is Up-Regulated by Keratin Absence or Mutation Mol. Biol. Cell, December 1, 2007; 18(12): 4969 - 4978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Schweizer, P. E. Bowden, P. A. Coulombe, L. Langbein, E. B. Lane, T. M. Magin, L. Maltais, M. B. Omary, D. A.D. Parry, M. A. Rogers, et al. New consensus nomenclature for mammalian keratins J. Cell Biol., July 17, 2006; 174(2): 169 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Zhou, M. Cadrin, H. Herrmann, C.-H. Chen, R. J. Chalkley, A. L. Burlingame, and M. B. Omary Keratin 20 Serine 13 Phosphorylation Is a Stress and Intestinal Goblet Cell Marker J. Biol. Chem., June 16, 2006; 281(24): 16453 - 16461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-Z. Tao, D. M. Toivola, Q. Zhou, P. Strnad, B. Xu, S. A. Michie, and M. B. Omary Protein phosphatase-2A associates with and dephosphorylates keratin 8 after hyposmotic stress in a site- and cell-specific manner. J. Cell Sci., April 1, 2006; 119(Pt 7): 1425 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Wada-Hiraike, O. Imamov, H. Hiraike, K. Hultenby, T. Schwend, Y. Omoto, M. Warner, and J.-A. Gustafsson Role of estrogen receptor beta in colonic epithelium PNAS, February 21, 2006; 103(8): 2959 - 2964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Habtezion, D. M. Toivola, E. C. Butcher, and M. B. Omary Keratin-8-deficient mice develop chronic spontaneous Th2 colitis amenable to antibiotic treatment J. Cell Sci., May 1, 2005; 118(9): 1971 - 1980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Wong, R. Domergue, and P. A. Coulombe Overcoming Functional Redundancy To Elicit Pachyonychia Congenita-Like Nail Lesions in Transgenic Mice Mol. Cell. Biol., January 1, 2005; 25(1): 197 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Omary, P. A. Coulombe, and W.H. I. McLean Intermediate Filament Proteins and Their Associated Diseases N. Engl. J. Med., November 11, 2004; 351(20): 2087 - 2100. [Full Text] [PDF]
|
|
|
|
|
|
B. Zhong, Q. Zhou, D. M. Toivola, G.-Z. Tao, E. Z. Resurreccion, and M. B. Omary Organ-specific stress induces mouse pancreatic keratin overexpression in association with NF-{kappa}B activation J. Cell Sci., May 1, 2004; 117(9): 1709 - 1719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
