| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Medicine,*
Pathology,
and
Surgery,
Brigham and Women's Hospital,
Harvard Medical School, Boston, Massachusetts
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For these reasons, defining the origin of these cells in long-term allografts of solid organs has considerable importance. There is older evidence from a rat experimental model that the endothelium of long-term aortic allografts contains some recipient-derived cells.3 In contrast, recent reports using immunohistochemical staining for donor-specific MHC class II molecules have demonstrated that endothelial cells (ECs) lining vessels of acutely rejecting allografts in rodents are of donor origin.4-7 Nevertheless, it is unclear whether the ECs that line vessels of longer-term allografts originate from the donor and/or recipient.
The availability of mouse strains of well defined histocompatibility haplotype, as well as allospecific antisera capable of selectively differentiating MHC molecules, should permit unambiguous determination of the animal of origin of all cell types. We have therefore used long-term total allogeneic-mismatched murine cardiac allografts to examine the pattern of donor versus recipient MHC class II expression in allograft ECs and vascular smooth muscle cells (SMCs) as well as allograft mononuclear inflammatory cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male BALB/c (B/c, H-2d) and C57BL/6 (B6, H-2b) mice were obtained from Taconic Farm (Germantown, NY) and were used at 10 to 12 weeks of age; body weight was approximately 25 g. B/c mice were used as allograft donors and B6 mice were used as recipients. These strains are disparate in major histocompatibility complex (MHC) class I, MHC class II, and multiple non-MHC alloantigens. B6 isografts were used as controls.
The mice were maintained in the Harvard Medical School animal facilities on acidified water. Sentinel animals surveyed serologically for viral pathogens were negative in the room in which these mice were housed. All experiments conformed to animal care protocols approved by the institutional review group.
Heterotopic Cardiac Transplantation
Heterotopic cardiac transplantation was performed using a modification of the method described by Corry et al.8 In brief, donors and recipients were anesthetized with Metofan (Pittman-Moore, Mundelein, IL). Donor hearts were perfused with chilled and heparinized saline via the inferior vena cava and harvested after ligation of the vena cava and pulmonary veins. The aorta and pulmonary artery of donor hearts were anastomosed to the abdominal aorta and inferior vena cava, respectively, of recipients using microsurgical techniques. Ischemic time was routinely approximately 25 minutes, with a success (long-term survival) rate of approximately 90%. The viability of the cardiac allograft was assessed by daily abdominal palpation. Cessation of the graft heartbeat was defined as the day of graft failure, typically indicating severe acute rejection as determined by histopathological examination.
Immunosuppressive Treatment and Experimental Groups
Table 1
shows the various
experimental groups. Immunosuppression was attained by weekly
intraperitoneal injections of 0.2 ml of ascites or comparably
concentrated antibody preparations containing anti-CD4 (GK1.5) and
anti-CD8 (2.43) monoclonal antibodies (MAbs) beginning 4 days after
transplantation.9
This protocol permits a single 4-day
early acute rejection episode, followed by complete CD4 and CD8 cell
ablation and therefore long-term suppression of any further
immune-specific response. The model permits long-term (at least 12
weeks) survival of complete allogeneic-mismatched allografts (not
possible in our hands with other protocols thus far published) and
yields graft arteriosclerosis lesions histologically identical to those
seen with conventional pretransplant MAb treatment.9,10
|
GK1.5 (anti-CD4) and 2.43 (anti-CD8) MAbs for immunosuppression
were prepared from hybridoma clones (American Type Culture Collection,
Rockville, MD) and used as ascites preparations or from threefold
concentrations of serum-free supernatants from an artificial capillary
system (Cellmax, Celluco, Rockville, MD); antibodies for CD45, B220,
Mac-3, and CD31 (PECAM-1) as well as biotinylated, fluorescein
isothiocyanate (FITC)-conjugated or nonconjugated isotype- and
species-specific secondary antibodies were purchased from PharMingen
(San Diego, CA). Mouse monoclonal anti-smooth muscle 
-actin antibody
(1A4) was obtained from Sigma Chemical Co. (St. Louis, MO).
Horseradish-peroxidase-conjugated anti-FITC antibodies were obtained
from Boehringer Mannheim (Indianapolis, IN).
Monoclonal Antibodies for MHC Class II Immunostaining
Biotinylated, FITC-conjugated and nonconjugated MHC class II
haplotype-specific MAbs as well as isotype-matched biotinylated,
FITC-conjugated, and nonconjugated IgG2a and
IgG2b control MAbs were purchased from PharMingen. The
anti-I-Ab antibody was clone AF6–120.1, specific for the
Ab chain, (IgG2a isotype, 
light chain); according to the supplier, this antibody is
cross-reactive with cells from H-2k and H-2u
mice and weakly reactive with cells from H-2p haplotype
mice but not with cells from H-2d animals. The
anti-I-Ad antibody was clone AMS-32.1 (IgG2b
isotype, light chain); according to the supplier, this antibody is
cross-reactive with cells from H-2f, H-2j, and
H-2v mice and weakly reactive with cells from
H-2k and H-2q haplotype mice but not with cells
from H-2b animals. To verify directly lack of
cross-reactivity under our immunostaining conditions, spleens from B6
and B/c mice were stained with both I-A-specific MAbs; as shown in
Figure 1
, each antibody specifically
stains cells from only the appropriate MHC II haplotype animals.
|
Grafts were explanted either at the time of cessation of heartbeat
or at 8 or 12 weeks and were transversely sectioned. One-half of each
heart was fixed in 10% buffered formalin for routine morphological
examination; paraffin-embedded sections of the fixed tissue were
stained with hematoxylin and eosin (H&E) or with an elastic fiber stain
using Weigert's method. The other half of each heart was frozen in OCT
compound (Ames Co., Division of Miles Laboratories, Elkhart, IN) and
stored at -80°C. Six-micron-thick sections were fixed in cold
acetone for 2 minutes, incubated at room temperature with 5% normal
goat serum in PBS for 20 minutes to block nonspecific binding sites,
and subsequently immunostained with rat MAbs for CD45, CD4, CD8, Mac-3,
or CD31 followed by biotinylated goat anti-rat IgG using an
avidin-horseradish peroxidase-biotin complex method11
(Vector Laboratories, Burlingame, CA) and an aminoethyl carbazole
substrate (Vector Laboratories); hematoxylin was used for
counterstaining. Controls included comparable concentrations of
nonspecific rat IgG in the first step. For immunohistochemical staining
for smooth muscle -actin, frozen sections were stained with mouse
1A4 antibody (IgG2a isotype) and then stained using
biotinylated goat anti-mouse IgG2a antibody. Controls
included comparable dilutions of nonspecific mouse IgG2a
antibody in the first step.
Immunohistochemical staining for MHC II was performed using directly biotin-conjugated antibodies to MHC II haplotypes I-Ab or I-Ad and the avidin-horseradish-peroxidase-biotin complex method.11 Controls included biotin-conjugated isotype-matched non-specific antibody in the first step.
Double-staining for MHC II and the EC marker CD31 was performed by sequential incubations with anti-CD31, biotin-conjugated anti-rat IgG, and avidin-alkaline-phosphatase-biotin complex method (Vector Laboratories) using the Vector Blue substrate kit (Vector Laboratories) and levamisole to block endogenous phosphatase activity. This treatment was followed by successive application of FITC-conjugated antibodies to MHC II haplotypes I-Ab or I-Ad and horseradish-peroxidase-conjugated anti-FITC; the slides were developed with the aminoethyl carbazole substrate. By this method, ECs stain blue, and MHC II expression is reflected by a red-brown stain; double-stained cells have a deep purple color. Controls included use of appropriately conjugated, isotype-matched irrelevant antibodies in the first step.
For immunofluorescent labeling of MHC class II molecules, frozen sections were incubated with biotin-conjugated antibodies to MHC II haplotypes I-Ab or I-Ad, followed by streptavidin-conjugated Texas Red (Amersham Life Science, Cleveland, OH), using standard techniques. Controls included comparable concentrations of isotype-matched, biotin-conjugated nonspecific antibodies in the first step.
Histological Evaluation
Grafts were analyzed by standard H&E and elastin stains, and the severity of parenchymal rejection versus graft arterial disease (GAD) was scored. Parenchymal rejection was graded using a scale modified from the International Society for Heart and Lung Transplantation12 (0, no mononuclear cell infiltrate; 1, mild interstitial or perivascular infiltrate without necrosis; 2, focal infiltrates with necrosis; 3, multifocal infiltrates with necrosis; 4, widespread infiltrates with hemorrhage and/or vasculitis), and a GAD score was calculated from the number and severity of involved vessels (0, <10% vascular occlusion; 1, 10 to 25% occlusion; 2, 25 to 50% occlusion; 3, 50 to 75% occlusion; 4, >75% occlusion).10 All results are expressed as the mean ± SD.
Numerical grades for the intensity and extent of staining for MHC class
II expression in both parenchymal inflammatory cells and vessels were
averaged from scores determined by three independent, blinded observers
(Table 2)
.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All isografts, whether recipients were given anti-CD4 and anti-CD8
MAbs or not, continued to beat until harvest at 12 weeks and showed
neither mononuclear cell infiltrates nor graft vascular lesions (Table 1)
. Allografts in nonimmunosuppressed animals ceased functioning at
approximately 7 days after transplantation and showed severe
parenchymal rejection (Table 1
; Figure 2A
). Long-term allografts (8 to 12 weeks)
in recipients immunosuppressed with weekly anti-CD4 and anti-CD8 MAbs
beginning 4 days after transplantation exhibited an ongoing mononuclear
cell infiltrate composed predominantly of macrophages.9
Arteries in these allografts exhibited intimal fibroproliferative
vascular lesions (Table 1
; Figure 2B
), resembling lesions seen in other
murine models of graft arteriosclerosis, as well as typical human graft
arteriosclerosis.2,9,10
|
Freshly explanted native B6 hearts showed no constitutive class II
expression in parenchymal or vascular wall cells (Figure 3A)
. Class II expression in isografts was
confined to infiltrating macrophages in the epicardium, and no vascular
or myocardial cells showed staining for class II (Table 3
and Figure 3, B and C
). The results
were the same regardless of whether isografted animals received weekly
anti-CD4 and anti-CD8 MAbs (Figure 3, B and C)
or not (not shown).
|
|
and Figure 4, A, B, E, and G
)
although scattered diffuse CD4- and CD8-positive T cells were also
present in the allografts (not shown). In contrast, ECs (CD31-positive)
and SMCs (smooth muscle -actin-positive) in the vessels showed
intense I-Ad (donor-specific) staining (Table 3
and Figure 4, C, D, and F
). Occasional inflammatory cells around vessels
also expressed I-Ad and are interpreted to represent
passenger leukocytes (Figure 4F)
.
|
and 12 weeks (Figure 6)
after transplantation demonstrated
I-Ab (recipient-specific) molecules on infiltrating
macrophages (Mac-3 positive) in the expanded intima and around vessels,
but not on ECs or medial SMCs (Table 3
; Figures 5 and 6
). In contrast,
I-Ad (donor-specific) molecules were expressed in the
vessel wall cells, including ECs and medial SMCs, but not in the
perivascular infiltrates (Table 3
; Figures 5 and 6
). Twelve-week
allografts in immunosuppressed recipients showed findings comparable to
8-week allografts but displayed overall lower levels of class II
expression (Table 3
; Figure 6
). No myocyte staining for MHC II of
either haplotype was ever seen, and in no case did allograft ECs or
SMCs express recipient MHC class II molecules.
|
|
show a rim of EC staining on
the vascular lumen and only background staining of the vessel wall and
surrounding interstitial tissue. Double staining for CD31 and recipient
MHC II (I-Ab) shows blue luminal staining of ECs as well as
strong I-Ab positivity in intramural and perivascular
inflammatory cells, demonstrating the host origin of the inflammatory
cells; ECs are not double stained (Figure 7B)
. In contrast, staining
for CD31 and donor MHC II (I-Ad) shows strong double
staining in the vessel lumen, demonstrating the donor origin of the ECs
(Figure 7C)
; as in Figure 6
, SMCs expressed relatively low levels of
donor MHC II.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,13
and MHC II on
murine macrophages may be increased by IFN-,14
tumor
necrosis factor-,15,16
and/or
interleukin-4.17
MHC class II expression on graft
endothelium has particular importance as a major determinant of graft
immunogenicity.2,13,18,19
Previous reports demonstrated
that ECs in acutely rejecting allografts express donor-specific class
II molecules.4-6
However, it was unclear whether donor or
recipient ECs populated the engrafted vessels in solid organs
transplanted for prolonged periods. This study used long-term (8- to
12-week) cardiac allografts in mice with MHC-mismatched donors and
recipients to establish the origin of the cells that express class II
molecules. Although MHC II expression is a critical component in the
stimulation of CD4+ T cells, it should also be emphasized
that a number of co-stimulator molecules, beyond the scope of this
study (eg, CD40, CD80, and CD86), likewise modulate the local immune
response. We have focused on MHC II expression because it plays a
central role in T cell stimulation by foreign tissues; persistent donor
MHC II would therefore be important in the pathology of long-term
allograft failure.
The immunosuppressive protocol used in these experiments permits a
single, early 4-day episode of untreated rejection, followed by
virtually complete CD4 and CD8 cell ablation and therefore long-term
suppression of any further immune-specific response.9
The
model permits long-term (at least 12 weeks) survival of complete
allogeneic-mismatched allografts. Moreover, this immunosuppression
protocol yields graft arteriosclerosis lesions histologically identical
to those seen with conventional pretransplant MAb
treatment.9,10
Although these long-term grafts lack CD4 and
CD8 T lymphocytes, they nevertheless exhibit an extensive infiltrate of
MHC-II-positive macrophages and levels of adhesion molecules (CD54
(ICAM-1) and CD106 (VCAM-1)) comparable to what we have described in
allografts with other immunosuppression protocols.9,10
Furthermore, the absence of IFN- abrogates graft arteriosclerosis
lesions in this model with early transient acute rejection, in a manner
analogous to what we have described using pretransplant
immunosuppression.9,10,20
We therefore believe that the
underlying pathogenic mechanisms of graft arteriosclerosis in this
model do not substantively differ from those described by ourselves and
others.9,10,20
In addition, the same persistence of donor
MHC-II-positive ECs and SMCs is seen in complete allogeneic-mismatched
allografts 12 weeks after transplantation, using rapamycin
immunosuppression where T cell ablation does not occur (S. Hasegawa H.
Nagano, P. Libby, and R. N. Mitchell, in preparation).
Normal nontransplanted mouse hearts in this study do not show any class II expression, although a previous study demonstrated the presence of class-II-positive dendritic cells in normal rat hearts.4 The class-II-positive cells in the rat heart study might reflect a low-level inflammatory stimulation due to pathogens. Our results support the contention that MHC class II expression in normal mouse heart tissue (ECs and SMCs) is inducible but not constitutive. Moreover, in agreement with others, we find no immunohistochemical evidence of MHC II expression on cardiac myocytes.21
In long-term isografts, only occasional macrophages in the epicardium express MHC class II molecules. In contrast, acutely rejecting allografts without treatment and allografts from long-term (8 to 12 weeks) immunosuppressed recipients both show intense expression of recipient-specific class II molecules in infiltrates around vessels but not in vascular ECs or SMCs. In the long-term allografts, mononuclear inflammatory cells in the expanded intima also express recipient-specific MHC class II molecules. The mononuclear cells expressing recipient MHC class II are likely macrophages,22,23 as they also stain for Mac-3; although B cells also bear MHC class II, they are generally sparsely represented in these allografts (not shown). Activated T cells in mice do not express class II molecules24; moreover, the immunosuppressive regimen used in these experiments (weekly anti-CD4 and anti-CD8 MAbs) deletes the vast majority of T cells from the periphery as well as the allograft.9 The findings suggest that sustained expression of host MHC class II molecules on infiltrating macrophages (host antigen-presenting cells) could enhance immune responses even in long-term allografts.
The ECs and medial SMCs of acutely rejecting allografts express donor-specific MHC class II molecules, in accord with previous work.4-7 Although other studies have reported that graft ECs express donor-specific MHC class II molecules in the first two weeks after transplantation, our study demonstrates that, in allografts up to 12 weeks, ECs and SMCs express predominantly donor-specific MHC class II molecules. This finding suggests that ECs and SMCs in long-term allografts largely derive from the donor and that ongoing, allo-restricted (direct) antigen presentation by these vascular wall cells may contribute to a sustained local immune response during the development of graft arteriosclerosis. We did not examine the origin of vascular wall cells beyond 12 weeks; however, the absence of any discernible recipient ECs or SMCs, even after substantial vascular pathological change has occurred, demonstrates the ability of donor cells to persist for long periods of time.
Although the medial SMCs predominantly derive from donor cells, it is much more difficult to establish the origin of the SMCs of the intimal fibroproliferative region. Despite strong MHC II immunoreactivity of ECs and medial SMCs, the SMCs accumulating in the intima consistently showed very poor staining for MHC II of either donor or host haplotype. This result agrees with previous work demonstrating an inverse correlation between SMC proliferation and MHC II expression.25 Accordingly, sites with the greatest growth, such as the intima of graft arteriosclerotic lesions, would be expected to have the lowest surface expression of MHC II.
Plissonnier et al,26
in a rat aortic allograft model,
showed that the SMCs in late (2 months) intimal hyperplastic lesions
may derive from the allograft recipient. However, in these aortic
allografts, the donor EC and SMC populations die within the first 1 to
2 weeks after transplantation and would be unable to repopulate the
graft. Because of the extensive early aortic wall necrosis, the later
predominance of recipient cells is therefore not surprising. Moreover,
these experiments did not determine how many of the intimal cells were
in fact infiltrating inflammatory cells. In the cardiac allografts, in
contrast, we do not see any substantial vascular wall necrosis, and we
can identify which cell type is present in the intimal hyperplastic
lesions. As shown in Figures 5 and 6
, infiltrating recipient
inflammatory cells can be readily demonstrated in the areas of intimal
hyperplasia. Likewise, we can demonstrate the presence of smooth muscle
-actin-positive cells in these areas but cannot establish their MHC
II haplotype. Consequently, the origin of the proliferating intimal
SMCs is uncertain. However, given that the allograft vascular media is
derived from the donor, it seems most likely that the bulk of the
intimal SMCs would also be largely donor derived.
Occasional cells around vessels in acutely rejecting allografts express donor-specific MHC class II molecules, most likely representing passenger leukocytes.27,28 A previous study in human allografts showed that dendritic cells were present in the vascular lesions.29 In our study, however, the class-II-expressing donor antigen-presenting cells decrease with time and have mostly disappeared by 12 weeks after transplantation. The role of passenger leukocytes in the sustained regional immune response in coronary arteries therefore appears limited, but it is conceivable that MHC class II expression by passenger leukocytes or donor dendritic cells might be implicated in long-term allograft acceptance. Tolerance induction by the mechanism of microchimerism has been suggested to reduce the development of graft arteriosclerosis.30,31
In conclusion, we have demonstrated in long-term cardiac allografts that infiltrating inflammatory cells largely express recipient-specific MHC class II molecules and that graft ECs and medial SMCs express donor-specific MHC class II molecules. These results suggest that donor-derived antigenic peptides binding to MHC molecules on host macrophages can contribute to indirect CD4+ T cell activation in long-term allografts. Concurrently, persistent allogeneic donor MHC class II molecules on graft ECs and SMCs can also contribute to a sustained direct activation of host CD4+ T cells. It is not possible to specifically quantitate the relative contribution of host versus donor MHC II to the immunological response; it is nevertheless consequential that numerous donor ECs and SMCs, capable of expressing foreign MHC antigens, persist long into the life of a solid-organ allograft.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by NIH grant HL 43364 to P. Libby and R.N. Mitchell.
S. Hasegawa's current address: Cardiothoracic Surgery, Tokyo Medical and Dental University, Tokyo, Japan.
H. Nagano's current address: Department of Surgery, Osaka University Medical School, Suita Osaka, Japan.
Accepted for publication April 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Am J Pathol 1998, 152:1187-1197[Abstract]
deficiency prevents coronary arteriosclerosis but not myocardial rejection in transplanted mouse hearts. J Clin Invest 1997, 100:550-557[Medline]
-interferon. J Exp Med 1983, 157:1339-1353, prostaglandin E2, and corticosterone on induced Ia expression on murine macrophages. J Immunol 1990, 145:1167-1175[Abstract]
and tumor necrosis factor- in a murine macrophage cell line. J Exp Med 1990, 171:1283-1299-Interferon regulates vascular smooth muscle proliferation and Ia antigen expression in vivo and in vitro. Circ Res 1988, 63:712-719This article has been cited by other articles:
 |
|
 |
|
  A. Valujskikh, Q. Zhang, and P. S. Heeger CD8 T Cells Specific for a Donor-Derived, Self-Restricted Transplant Antigen Are Nonpathogenic Bystanders after Vascularized Heart Transplantation in Mice J. Immunol., February 15, 2006; 176(4): 2190 - 2196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Furukawa, S. E Cole, R. V Shah, Y. Fukumoto, P. Libby, and R. N Mitchell Wild-type but not interferon-{gamma}-deficient T cells induce graft arterial disease in the absence of B cells Cardiovasc Res, August 1, 2004; 63(2): 347 - 356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Grazia, B. A. Pietra, Z. A. Johnson, B. P. Kelly, R. J. Plenter, and R. G. Gill A Two-Step Model of Acute CD4 T-Cell Mediated Cardiac Allograft Rejection J. Immunol., June 15, 2004; 172(12): 7451 - 7458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ii and D. W. Losordo Transplant Graft Vasculopathy: A Dark Side of Bone Marrow Stem Cells? Circulation, December 23, 2003; 108(25): 3056 - 3058. [Full Text] [PDF]
|
|
|
|
|
|
P. Libby Bone Marrow: A Fountain of Vascular Youth? Circulation, July 29, 2003; 108(4): 378 - 379. [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. Hillebrands, F. A. Klatter, and J. Rozing Origin of Vascular Smooth Muscle Cells and the Role of Circulating Stem Cells in Transplant Arteriosclerosis Arterioscler. Thromb. Vasc. Biol., March 1, 2003; 23(3): 380 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. V. Subramanian, R. J. Kelm Jr., J. A. Polikandriotis, C. G. Orosz, and A. R. Strauch Reprogramming of vascular smooth muscle {alpha}-actin gene expression as an early indicator of dysfunctional remodeling following heart transplant Cardiovasc Res, June 1, 2002; 54(3): 539 - 548. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Li, X. Han, J. Jiang, R. Zhong, G. M. Williams, J. G. Pickering, and L. H. Chow Vascular Smooth Muscle Cells of Recipient Origin Mediate Intimal Expansion after Aortic Allotransplantation in Mice Am. J. Pathol., June 1, 2001; 158(6): 1943 - 1947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Furukawa, D. A. Mandelbrot, P. Libby, A. H. Sharpe, and R. N. Mitchell Association of B7-1 Co-Stimulation with the Development of Graft Arterial Disease : Studies Using Mice Lacking B7-1, B7-2, or B7-1/B7-2 Am. J. Pathol., August 1, 2000; 157(2): 473 - 484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Heckenkamp and G. M. Lamuraglia Intimal Hyperplasia, Arterial Remodeling, and Restenosis: An Overview Perspectives in Vascular Surgery and Endovascular Therapy, January 1, 1999; 11(2): 71 - 94. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
