ABBS 2005,38(03): Stable Skin-specific Overexpression of Human CTLA4-Ig in Transgenic Mice through Seven Generations

Abstract        Skin graft rejection is
a typical cellular immune response, mainly mediated by T cells. Cytotoxic T
lymphocyte associated antigen 4-immunoglobin (CTLA4-Ig) extends graft survival
by blocking the T cell co-stimulation pathway and inhibiting T cell activation.
To investigate the efficacy of CTLA4-Ig in prolonging skin graft survival,
human CTLA4-Ig (hCTLA4-Ig) was engineered to overexpress in mouse skin by
transgenesis using the K14 promoter. Reverse transcription-polymerase chain
reaction (RT-PCR) and Western blot assay indicated that the expression of
CTLA4-Ig remained skin-specific and relatively constant compared to the
internal control protein, AKT, through seven generations. The presence and
concentration of the hCTLA4-Ig protein in transgenic mouse sera was determined
by enzyme-linked immunosorbent assay (ELISA), and the results indicated that
the serum CTLA4-Ig concentration also remained constant through generations.
Survival of transgenic mouse skins grafted onto rat wounds was remarkably
prolonged compared to that of wild-type skins from the same mouse strain, and
remained comparable among all seven generations. This suggested that the
bioactive hCTLA4-Ig protein was stably expressed in transgenical mice through
at least seven generations, which was consistent with the stable skin-specific CTLA4-Ig
expression. The results demonstrated that the transgenic expression of
hCTLA4-Ig in skin driven by the K14 promoter remained constant through
generations, and a transgenic line can be established to provide transgenic
skin with extended survival reproducibly.

Key words        cytotoxic T lymphocyte
antigen 4-Ig; transgenic mice; skin; K14 promoter

Cytotoxic T lymphocyte
associated antigen 4-immunoglobin (CTLA4-Ig) is a soluble fusion chimeric protein
consisting of the extracellular region of cytotoxic T lymphocyte associated
antigen 4 (CTLA4) on the N-terminus and the constant region of human IgG on the
C-terminus. By blocking the main co-stimulatory signal pathway of T cell,
B-7/CD28-CTLA4, CTLA4-Ig is capable of inhibiting T cell activation and
rendering T cells to an anergy state [1–3]. Using
CTLA4-Ig protein, many researchers have prolonged allo- or xeno-graft survival
in different experimental models [4–8], and some
even induced donor-specific immune tolerance [9]. Skin grafting is a regular
method to cover large burn wounds in burn patients, but immune rejection is the
main hurdle against the survival of heterologous (allo- or xeno-) skin grafts.
The immune rejection initiated by grafted skin is a typical acute cellular
immune response mainly mediated by T cells. Because of its effective inhibition
of T cell activation by blocking co-stimulation, CTLA4-Ig is a potential
therapeutic molecule that might extend skin graft survival.

Although CTLA4-Ig has
the capacity to prolong graft survival and even induce immune tolerance, it has
the potential to inhibit the immune system extensively because its mechanism is
non-antigen specific when it is administered continuously, systemically or
directly into the recipient. At present, in most experiments using CTLA4-Ig to
prolong graft survival, CTLA4-Ig protein or its recombinant adenovirus is
administered directly into the recipient. For patients with large burns, the
severe damage to the skin makes them extremely vulnerable to exogenous
infections. Therefore it is inconvenient and insecure to administer CTLA4-Ig
protein or its recombinant adenovirus directly into burn patients to prolong
skin graft survival.

A better choice may be
to express CTLA4-Ig in the skin graft. First, if CTLA4-Ig was expressed in the
skin graft, the activated and infiltrating T cells against skin antigen would
be more ready to be regulated by CTLA4-Ig molecules produced by the skin graft in
situ, and the remaining T cells of the recipient would be subjected to a
much lower concentration of CTLA4-Ig protein diffused from grafted skin. The
survival of the skin graft could be further prolonged and the probability of
extensive inhibition of recipient immunity by CTLA4-Ig would be further
reduced. Second, it is a time-consuming, difficult, costly and complex process
to obtain purified CTLA4-Ig protein or a high titer of recombinant adenovirus.
If CTLA4-Ig was transgenically expressed in the skin graft, there would be no
need to prepare purified CTLA4-Ig protein or recombinant adenovirus, and
therefore the cost of therapy would be remarkably reduced.

In clinical practice,
pig skin is frequently used to cover large burn wounds. We sought to produce a
transgenic pig line skin-specifically overexpressing human CTLA4-Ig (hCTLA4-Ig)
protein at a constant level through generations and over lifetimes, and used
this line as a reproducible source to obtain a transgenic skin graft capable of
prolonged survival for clinic purposes. To test this strategy in mice we have
previously produced transgenic mouse line skin-specifically expressing
hCTLA4-Ig [10].

In this paper, we have
demonstrated that CTLA4-Ig was skin-specifically overexpressed in the
transgenic line at a constant level through seven generations, further
indicating the feasibility and practicability of the strategy mentioned above.

Materials and Methods

Generation of hCTLA4-Ig
transgenic mice

hCTLA4-Ig transgenic
mice were generated, as described previously [10]. Briefly, the transgene
expression vector, which was used to generate CTLA4-Ig transgenic mice,
contains 2-kb AvaI fragment encompassing the K14 promoter and enhancers.
To stabilize transgene transcripts and enhance transgene expression, a fragment
containing rabbit b-globin 5‘
untranslated region (5‘ UTR) and an intron was inserted into the 5‘
terminus of CTLA4-Ig coding sequence, as well as the K14 3‘ UTR
and polyadenylation signal into 3‘ terminus. The transgene construct was
cut from plasmid and microinjected into fertilized mouse eggs to obtain
transgenic mice. The transgenic mice were engineered in the outbred strain of
KM, and the transgenic mouse line was established by mating with wild-type
mice.

RNA isolation and
reverse transcription-polymerase chain reaction (RT-PCR) analysis

Approximately 40 mg
tissue was cut from skin or other solid organs, such as liver, spleen, kidney
and intestine. Each was immediately immersed into 2 ml ice-cold TriPure
isolation reagent (Roche, Two-step RT-PCR was
performed to detect transgene expression. In a sterile RNase-free tube, the
mixture contained the following reagents (total 16 ml): 1–2 ml total RNA, 2 ml oligo-d(T)₂₃VN primers (50 mM), 4 ml dNTP
mixture ( 

Skin protein preparation
and Western blot analysis

Approximately 40 mg skin
tissue was collected from transgenic or wild-type mice, cut into tiny pieces
with clean scissors and immediately placed into liquid nitrogen. The frozen skin
tissue was crushed and homogenized thoroughly in cold tissue protein extraction
reagent (T-PER). After centrifuge at 4 ºC for 10 min (Approximately 40 mg skin protein was subjected to conventional
sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and
electrotransferred onto polyvinylidene fluoride membrane (Roche). The blot was
blocked with phosphate-buffered saline (PBS)+0.1% Tween 20 supplemented with 5%
(W/V) bovine
serum albumin (BSA) at 4 ºC for 16 h. The blocked blot was then incubated with
diluted (1:800) goat anti-human CTLA4 antibody (Santa-Cruz,  

Sandwich ELISA of
hCTLA4-Ig protein in transgenic mouse skins and sera

Transgenic mouse sera
and skin proteins were prepared from transgenic individuals randomly chosen
from F1–F7 generations (one mouse for each generation).
The sera were collected by eyeball enucleation and skin proteins were
extracted, as described above. The concentrations of skin protein samples were
measured with the To perform ELISA, the
goat anti-human CTLA4 antibody (Santa-Cruz) was diluted to 5 mg/ml with coating buffer ( 

Skin grafting

Wistar rats were used as
recipients for skin grafting. Burn wounds ( 

Immunohistochemical
staining

A tiny block of skin graft
was cut off and fixed in Bouin’s fixative overnight. The fixed skin tissue was
embodied in paraffin and a 5 mm section was
routinely prepared. The section was subjected to immunohistochemistry staining
using goat anti-human CTLA4 antibody as the primary antibody and HRP-conjugated
rabbit anti-goat IgG Fc antibody as the secondary antibody.

Results

Skin-specific CTLA4-Ig
expression through seven generations

To analyze the pattern
of CTLA4-Ig expression in transgenic mice through seven generations,
transgenic individuals were randomly chosen from F1–F7 generations (one mouse for each generation)
and total RNA was isolated from skins, livers, spleens, kidneys, muscle and
intestines and subjected to RT-PCR assay. To investigate if CTLA4-Ig was
ectopically expressed, RNA samples from liver, spleen, kidney or intestine of
different generations were mixed respectively and subjected to RT-PCR as a
pool. The transgenic skin RNA samples from different generations were also
mixed and subjected to RT-PCR without reverse transcriptase as a blank control.
Expression of the house-keeping gene, GAPDH, was detected at the same
RT-PCR process as an internal control and a measurement of the quality of RNA
samples. Results indicated that CTLA4-Ig was expressed skin-specifically
in transgenic mice through seven generations, and no ectopic expression was
observed (Fig. 1). GAPDH
expression was detected in all the RNA samples except the blank control,
suggesting that these RNA samples were qualified for RT-PCR with no genomic DNA
contamination.

Comparable abundance of
human CTLA4-Ig protein in transgenic skins of seven generations

To further confirm the
expression of CTLA4-Ig in transgenic mouse skin through generations, skin
protein samples were prepared from transgenic individuals at different ages
randomly chosen from F1–F7 generations (one mouse for each generation),
and subjected to Western blot using goat anti-human CTLA4 IgG as the primary
antibody. The constitutively expressing protein, AKT, was simultaneously
detected as an internal control and a measurement of protein sample quality.
Intensive Western blot bands were observed in all the transgenic skin protein
samples, and no Western blot band was detected in wild-type skin, suggesting
that the Western blot assay was specific [Fig. 2(A)]. The Western blot
band of CTLA4-Ig was even stronger than that of the internal control protein,
AKT, suggesting that the expression level of CTLA4-Ig protein was quite high,
which was consistent with previous Northern blot assay [10]. Relative hCTLA4-Ig
abundance was obtained by comparing the Western blot signal intensity of
hCTLA4-Ig to that of AKT using Geldoc 2000 software. The results showed that
the relative CTLA4-Ig abundance was comparable among all the seven transgenic
skin protein samples [Fig. 2(B)], suggesting that CTLA4-Ig expression
remained relatively constant through at least seven generations regardless of
the ages of transgenic individuals.

Comparable hCTLA4-Ig
concentration in transgenic mouse sera and skin proteins through seven
generations

Secretory protein can be
transported into the bloodstream by skin-specific expression [17]. CTLA4-Ig
protein, lacking transmembrane and intracellular region, is highly secretory
with the additional recombined Oncostatin M signal peptide at the N-terminus
flanking the original CTLA4 signal peptide [20]. Skin is an organ with the largest area in the body, so the constitutive and stable
expression of CTLA4-Ig in skin may result in the constant presence of CTLA4-Ig
in serum. To test this hypothesis, transgenic mice were randomly chosen from F1–F7 generations (one mouse for each generation),
and transgenic mouse sera were collected from the selected transgenic
individuals and subjected to ELISA (each sample had three duplicates). To
determine the serum hCTLA4-Ig concentration, purified hCTLA4-Ig protein samples
at different concentrations were simultaneously subjected to ELISA along with
transgenic sera. Non-transgenic sera collected from wild-type mice of the same
strain were used as a negative control. ELISA without adding antigen solution
was performed as a blank control. The mean serum hCTLA4-Ig concentration of all
the transgenic mouse serum samples was 117.0618.20 ng/ml. The A490 value of wild-type serum was 0.0640.006,
which was comparable to that of the blank control (0.0520.007), suggesting that
the ELISA was specific. The serum hCTLA4-Ig concentrations in transgenic mice
were comparable among all the seven generations (Fig. 3), indicating
that the serum CTLA4-Ig concentration was relatively constant through
generations, which was consistent with the results of Western blot assay.

To analyze the hCTLA4-Ig
concentration in transgenic mouse skin among different generations, skin
proteins were extracted from the transgenic individuals. The skin protein
samples were diluted to 0.01 mg/ml and subjected to ELISA along with different
concentrations of purified hCTLA4-Ig protein samples in the same process as
that of sera. The ELISA results (A490 values) of
non-transgenic skin protein samples were 0.0760.005, which was comparable to
that of non-transgenic sera or the blank control. The mean hCTLA4-Ig
concentration in transgenic skin was 5.230.67 mg/mg extracted total skin protein. The hCTLA4-Ig
concentration in transgenic skin (mg/mg skin
protein) was also comparable among different generations (Fig. 4), which
was consistent with the ELISA results of transgenic sera and further confirmed
the results of Western blot assay.

Comparable survival of
transgenic skin grafts from seven generations

To further test if the
stable transgenic expression of hCTLA4-Ig may result in comparable survival of
transgenic skin grafts, skin grafting was performed from mouse to rat, because
the cross-reactivity of CTLA4 protein between rat and human is relatively high
[19]. Transgenic individuals were randomly chosen from each generation (one
mouse for each generation) and their skins were grafted onto rat wounds from
one donor to three recipients. Non-transgenic skin was collected from wide-type
mouse of the same strain, and grafted using the same process as negative
controls. The mean survival time (MST) of transgenic skin grafts of F1–F7 generations was 13.74.8 d, On 7 d post-grafting,
most of the area of grafted non-transgenic skin was necrotic [Fig. 5(B)].
In contrast, grafted transgenic skin showed no signs of necrosis on 7 d
post-grafting, even though the transgenic skin was collected from a transgenic
individual from the seventh generation. On 15 d post-grafting, the transgenic
skin graft remained alive, and hCTLA4-Ig expression was detected by
immunohistochemical staining [Fig. 5(B)].

Discussion

Burn injuries are
examples of trauma that break down the protective barrier of skin. The chief management
of burn wounds is effective and long-term covering with proper dressing to
prevent infection and water loss. Due to the deeply limited supply of auto skin
grafts, heterologous (allo- or xeno-) skin, with natural skin structure,
physiological vapor transmission characteristics and plentiful supply, is still
a major and regularly used dressing to cover large burn wounds in clinical
practice, although different kinds of composite skin substitutes have been
developed [12]. The immune rejection against skin graft, which is a typical
acute cellular response mainly mediated by T cells [13–15], remains the main hurdle against the survival of
grafted heterologous skin.

hCTLA4-Ig is capable of
blocking T cell co-stimulation and further inhibiting T cell activation.
Therefore, CTLA4-Ig is a valuable therapeutic molecule to extend skin graft
survival [18]. In most previous experiments using CTLA4-Ig to extend graft
survival, purified hCTLA4-Ig protein or its recombinant adenovirus was directly
injected into recipients in which the protective skin barrier was functional
and the immune system was not in a shocked state, like that of burn patients,
and ready to be regulated. For patients with large burns, due to the loss of
skin barrier, the shock state and high vulnerability to exogenous infection, it
is not practical or reasonable to apply hCTLA4-Ig protein or its recombinant
viral vectors, which inhibit T cell activation by blocking the co-stimulation
signal in an antigen-non-specific manner, directly into
these patients.

To engineer a skin graft
to express hCTLA4-Ig protein instead may be more reasonable for burn patients.
It would be more practical, convenient and economic to cover burn wounds with
genetically manipulated skins autonomously expressing hCTLA4-Ig protein, than
with wild-type skins by injection of purified hCTLA4-Ig protein or its viral
vectors. With local expression of hCTLA4-Ig in grafted skin, the activated T
cells infiltrating the skin graft would be more ready to be regulated by
hCTLA4-Ig protein produced by grafted skin in situ, leaving the
remaining T cells of the recipient exposed to a relatively lower concentration
of hCTLA4-Ig protein diffused from a transgenic skin graft. As a result, the
survival of the skin graft would be further extended and the antigen
non-specific inhibitory efficacy of hCTLA4-Ig protein on the immunity of burn
patients would be further reduced.

Our previous data showed
that hCTLA4-Ig expression remained relatively constant through at least
three generations at the transcriptional [10] and translational levels (data
not shown). Using RT-PCR assay, we found that human CTLA4-Ig expression
remained skin-specific through the seven generations. By Western blot assay,
CTLA4-Ig protein was detected in the skins of transgenic individuals randomly
chosen from the seven generations, with comparable abundance to that of the
internal control protein, AKT. Using phosphoimage analysis to obtain the
relative expression of CTLA4-Ig compared to that of the internal control, AKT,
we found that the expression of CTLA4-Ig remained relatively constant through
all seven generations. The concentrations of hCTLA4-Ig protein in transgenic
mouse skin were also determined by ELISA, and the results further demonstrated
the stable expression of hCTLA4-Ig in transgenic mice through generations. No
immuno-compromise phenotype was observed in the transgenic individuals of any
generation, except that some transgenic individuals had fragmented tails,
consistent with our previous report [10], suggesting that mice have a rather
good tolerance to the overexpression of exogenous CTLA4-Ig protein. Because the
cross-reactivity of CTLA4 protein between pig and human was even lower than
that between mouse and human [19], pig may have an even better tolerance to hCTLA4-Ig
expression. These results provide a straight demonstration of the efficacy of
the K14 promoter to drive hCTLA4-Ig
to overexpress in skin stably through generations, and rather good tolerance of
mice (or pig) to hCTLA4-Ig overexpression. This lays the foundation to
further establish a transgenic pig line with stable and skin-specific
expression of hCTLA4-Ig.

Skin grafting
demonstrated that the survival of transgenic skins grafted on rat wounds was
significantly longer than that of non-transgenic skins derived from the same
mouse strain. The expression of hCTLA4-Ig was still detected in grafted
transgenic skin even on 15 d post-grafting, suggesting that the transgenic skin
graft was still bioactive at this time point. The MST of transgenic skin was comparable
among all seven generations, which was consistent with the stable and
skin-specific overexpression of CTLA4-Ig through generations, indicating that
the transgenically expressed CTLA4-Ig protein was functional. This result
suggests that once a transgenic line overexpressing hCTLA4-Ig skin-specifically
and stably is established, transgenic skin capable of extended survival would
be obtained through the natural animal breeding process, and there would be no
need to prepare purified CTLA4-Ig protein or its recombinant adenovirus.

Serum hCTLA4-Ig protein
was detected in transgenic mice and its concentration was relatively higher
than that of serum hGH, resulting from skin-specific hGH overexpression
in mice driven by the K14 promoter [17]. This indicated that the CTLA4-Ig
protein expressed in skin was transported into the bloodstream. Serum hCTLA4-Ig
concentration also remained constant through generations, suggesting that the
transgenic line could be used as a reproducible resource to provide transgenic
sera. This result suggests that the transgenic skin overexpressing CTLA4-Ig
protein can be used not only as a wound dressing, but also as a bioreactor to
produce CTLA4-Ig protein for therapeutic purposes. In addition to using the
transgenic skin directly to cover burn wounds, the keratinocytes in transgenic
skin, with extraordinary proliferative capacity in culture and well-documented
values in burn grafting operation [16,17], can also be used as a vehicle to
deliver immunoregulating molecules or other factors in burn grafting. The
transgenic animal line, skin-specifically expressing valuable therapeutic
molecules in vivo, can be used as the reproducible source of such
genetically manipulated vehicle keratinocytes.

References

 1   Linsly
PS, Wallence PM, Iohnson J. Immuno-suppression in vivo by a soluble form
of the CTLA4 T cell activation molecule. Science 1992, 257:792–797

 2   Schwarts
RH. A cell culture model for T lymphocyte clonal anergy. Science 1990, 248:
1349–1356

 3   Linsley
PS, BradyW, Urnes M, Grosmaire LS, Damle NK, Ledbetter JA. CTLA-4 is a second
receptor for the B cell activation antigen B7. J Exp Med 1991, 174: 561–569

 4   Jin
YZ, Xie SS. Bicistronic adenovirus-mediated gene transfer of CTLA4Ig gene and
CD40Ig gene result in indefinite survival of islet xenograft. Transplant Proc
2003, 35: 3165–3166

 5   Wang
YF, Xu AG, Hua YB, Wu WX. Effect of local CTLA4Ig gene transfection on acute
rejection of small bowel allografts in rats. World J Gastroenterol 2004, 10:
885–888

 6   Stell
D,  7   Jin
Y, Zhang Q, Hao J, Gao X, Guo Y, Xie S. Simultaneous administration of a
low-dose mixture of donor bone marrow cells and splenocytes plus adenovirus
containing the CTLA4Ig gene result in stable mixed chimerism and long-term
survival of cardiac allograft in rats. Immunology 2003, 110: 275–286

 8   Guo
Z, Wang J, Dong Y, Adams AB, Shirasugi N, Kim O, Hart J et al. Long-term
survival of intestinal allografts induced by costimulation blockade, busulfan
and donor bone marrow infusion. Am J Transplant 2003, 3: 1091–1098

 9   Guo
L, Fujino M, Kimura H, Funeshima N, Kitazawa Y, Harihara Y, Tezuka K et al.
Simultaneous blockade of co-stimulatory signals, CD28 and ICOS, induced a
stable tolerance in rat heart transplantation. Transpl Immunol 2003, 12: 41–48

10  Wang
Y, Wang FC, Wei H, Ni Y, Wu J, Gao X. Establishment of transgenic mouse line
skin-specifically expressing hCTLA4-Ig. Yi Chuan Xue Bao 2005, 32: 916–922

11  Ge
LP, Wei H. A preliminary study on rat model of diabetic ulcers. Acta
Laboratorium Animalis Scientia Sinica 2005, 13: 88–80

12  Robert
L, Sheridan, Ronald G. Tompkins. Skin substitutes and burns. Burns 1999, 25:97–103

13  Simeonovic
S, Basadomn G. Effect of GK1.5 monoclonal antibody dosage on survival of pig
proislet exnografts in CD4+ depleted mice. Transplantation 1990, 49: 489–512

14  Zhan
Y, Corbett AJ, Brady JL, Sutherland RM, Lew AM. Delayed rejection of
fetal pig pancreas in CD4 cell deficient mice was correlated with residual
helper activity. Xenotransplantation 2000, 7: 267–289

15  Pierson
RN, Winn HJ, Russell PS. Xenogeneic skin graft rejection is especially
dependent on CD4+ T cells. J ExpMed 1989, 170: 991–1002

16  Ni
Ao, Yang ZC. Therapeutics of Burns. 17  Wang
X, Zinke S, Polonsky K, Fuchs E. Transgenic studies with a keratin
promoter-driven growth hormone transgene: Prospects for gene therapy. Proc Natl
Acad Sci USA 1997, 94: 219–226

18 Luo
GX, Wu J, Yi SX, Chen XW. Prolonging the survival of transplanted alloskin by
transferring CTLA4-Ig locally with recombinant adenovirus vector in mice. Chinese
Journal of Burns 2000, 16: 37–39

19  Lui
VC, Tam PK, Leung MY, Lau JY, Chan JK, Chan VS, Dallman M et al. Mammary
gland-specific secretion of biologically active immunosuppressive agent
cytotoxic-T-lymphocyte antigen 4 human immunoglobulin fusion protein (CTLA4Ig)
in milk by transgenesis. J Immunol Methods 2003, 277: 171–183

20  Yi
SX, Wu J, Jiang Q. Cloning of extracellular region of human CTLA4 and
construction of Ig fusion protein expression vector. Acta Academia Medicinae
Militaris Teritia 2000, 22: 834–837