(Circulation. 1995;92:1919-1926.)
© 1995 American Heart Association, Inc.
Articles |
From the Division of Vascular Surgery of the General Surgical Services (G.M.L., F.A.), Department of Pathology (T.J.F.), Department of Dermatology (N.A.M., T.J.F.), and Wellman Laboratories of Photomedicine (G.M.L., F.A., N.A.M., T.J.F.), Massachusetts General Hospital and Harvard Medical School, Boston, and the Department of Surgery, University of Cologne (T.S.-R.), Germany.
Correspondence to Glenn M. LaMuraglia, MD, Division of Vascular Surgery, Massachusetts General Hospital, Boston, MA 02114.
| Abstract |
|---|
|
|
|---|
Methods and Results Shortly after impregnation with the photosensitizer drug chloroaluminum sulfonated phthalocyanine, infrarenal aortas of ACI rats were PDT-treated and orthotopically grafted in Lewis rats (PDT). The transplanted grafts were sequentially analyzed at 2, 4, and 8 weeks by morphometry, immunohistochemistry, and scanning electron microscopy. Of 25 untreated allografts, 4 (16%) developed aneurysms compared with 0 of 33 in PDT or untreated isografts (ISO, P<.001). PDT treatment of allografts significantly inhibited intimal hyperplasia (P<.001) and resulted in intimal areas comparable to those in ISO. However, medial thickness in both control allografts and PDT grafts was markedly decreased compared with ISO. External graft diameters of control allografts at 8 weeks were significantly enlarged (P<.02) compared with PDT or ISO. At all time points, T lymphocytes were found in a substantially larger number in untreated control grafts than in PDT or ISO. Scanning electron microscopy at 4 weeks confirmed complete repopulation with endothelial cells in PDT, which was not seen in the control allografts.
Conclusions Our findings suggest that local PDT treatment of arterial allografts inhibits inflammatory infiltration, aneurysmal dilatation, and development of intimal hyperplasia and may be used to develop vascular bioprostheses for use in humans.
Key Words: photodynamic therapy arteries free radicals allografts rejection
| Introduction |
|---|
|
|
|---|
Transplanted arteries have been part of the vascular surgeon's armamentarium for generations. Some of the earliest work in vascular reconstruction used tissue of both human and nonhuman origin.2 3 Bovine heterografts4 and cryopreserved human arterial and venous allografts5 6 have been tested as arterial bypass conduits. However, studies on arterial allografts revealed significant histological changes resulting from chronic rejection7 8 : shortly after implantation, infiltration of T cells and macrophages caused a severe inflammatory reaction, with breakdown and degeneration of the collagen-elastin network in the media. This event resulted in a loss of biomechanical stability with graft dilatation, aneurysm formation, and rupture. Furthermore, endothelial cell damage and later myointimal proliferation led to thrombus formation, anastomotic narrowing, and graft occlusion.
Problems relating to the elimination of graft antigens and preservation of vessel function are not as complex as those involved in organ transplantation. Viability of the implanted graft, although possibly desirable, is not a principal requirement. Therefore, vascular grafts of biological origin have been subjected to a wide variety of preservation or preparation techniques before implantation. These techniques included chemical stabilization,9 cryopreservation,4 5 6 lyophilization, and UV10 and gamma irradiation. However, there is no convincing evidence that any of these approaches provide a satisfactory long-term outcome.
Photodynamic therapy (PDT) is a process that uses light to activate otherwise relatively inert photosensitizers for the production of free-radical moieties. After absorption of light, the photosensitizer is activated into an excited triplet state from which it can either directly form a free radical (type 1 reaction) or transfer its energy to molecular oxygen to generate singlet oxygen (type 2 reaction), both of which cause cell injury. These free radicals exert cytotoxic effects by damaging cellular membranes and organelles and cross-linking proteins.11 12
In experimental studies, PDT has been shown to inhibit intimal hyperplasia (IH) by cell depletion of the vessel wall without provoking an inflammatory response.13 14 Because vascular graft rejection is triggered primarily by major histocompatibility complex (MHC) antigens in cells residing in the vessel wall,7 8 15 PDT was postulated as an innovative method to blunt or obviate immunologic responses by eliminating these cells. Therefore, this work was undertaken to investigate the effects of PDT on graft rejection in an allotransplantation model using rats of two histocompatibly disparate, high-responder, inbred strains.
| Methods |
|---|
|
|
|---|
The rats were assigned to five different experimental groups: in the treatment group, PDT-treated ACI aortas were grafted in Lewis rats (PDT). Controls included nontreated (NT) and those treated with photosensitizer only (PO) and light irradiation only (LO) aortas. Syngeneic isotransplantations of untreated aortic grafts (ISO) were made from Lewis to Lewis rats, verified surgical changes, and served as negative control for the transplantation model. Rats from every group except NT and PO were killed sequentially at 2, 4, and 8 weeks. To minimize the number of animals, NT and PO rats were studied only at 8 weeks. For the whole study, 59 transplantations were performed.
Photodynamic Therapy
The photosensitizer chloroaluminum
sulfonated phthalocyanine
(CASPc, Ciba Geigy) was diluted to a concentration of 5 mg/mL in PBS
and administered intravenously at a dose of 5 mg/kg 24
hours before explantation of PDT and PO grafts. Rats in the ISO, LO,
and NT groups received an equivalent volume of saline solution. Rats
receiving the photosensitizer without preimplantation laser irradiation
(PO) had the aorta resected in low ambient light. Implantation of these
grafts was performed under filtered illumination at a band pass of 440
to 600 nm to avoid photoactivation of CASPc, which was already
present in the tissue of the allograft.
For irradiation of PDT and LO grafts, an argon-pumped dye laser (Coherent INNOVA I 100 and Coherent CR 599) was tuned to emit light at 675 nm and coupled to a 1-mm core silica fiber. With a 5-mm-focal-length lens, the output from the fiber optic was magnified to provide a uniform 2-cm spot and calibrated to an irradiance of 100 mW/cm2. The grafts were mounted on a bridged double microclamp (Pilling), placed on a reflective surface, and submerged in 5 µg/mL CASPc in PBS. Irradiations were performed at a fluence of 100 J/cm2 and included turning the graft over after 50% of the total energy was delivered. After PDT, the aortic grafts were gently rinsed with PBS and stored in 4°C Hanks' solution before implantation.
Harvest
Under anesthesia, the iliac artery was cannulated
for blood pressure monitoring and perfusion fixation. Midgraft
diameters were determined in the donor rat before explantation and in
the recipient before euthanasia by use of a calibrated eyepiece in the
operating microscope. After euthanasia, the aorta was flushed with
saline and subsequently perfusion-fixed in situ at 80 mm Hg with 10%
buffered formalin for light microscopy or 1.5%
glutaraldehyde in cacodylate buffer for electron
microscopy. The grafts were excised and placed in fresh 10% formalin
or 4% glutaraldehyde, respectively. For
immunohistochemistry, fresh grafts were excised, rinsed thoroughly with
saline, and placed in Tissue Tek OCT compound (Miles Inc) before
storage at -70°C.
Immunohistochemistry
Monoclonal anti-rat OX 47 antibody
against T lymphocytes (Sera
Lab) was used for immunohistochemical characterization of the
inflammatory infiltrates. Cryosections (4 µm) were stained with the
two-layer indirect immunoperoxidase technique.7 Briefly,
the sections were air-dried for 30 minutes and subsequently fixed in
acetone for 8 minutes. After incubation with primary antibody (1:1000),
endogenous peroxidase activity was blocked with 0.3%
H2O2 in PBS. After incubation with normal horse
serum and rat-adsorbed, biotinylated horse anti-mouse immunoglobulin
(Vector Laboratories), the sections were treated with diaminobenzidine
(DAB Substrate Kit, Vector) and counterstained with nuclear fast red
(Sigma Chemical Co). Normal arteries were used as negative controls;
thymus tissue was used as positive control.
Morphological Studies
Formalin-fixed sections from three
different graft segments
(proximal, mid, and distal) were prepared with hematoxylin and eosin
stain and Verhoeff's elastin stain. Morphometric evaluation was
performed with a digitizing measurement system (Sigma Scan, Jandel
Scientific) coupled to a camera lucida.13 The inflammatory
response was assessed by scoring at least four microscopic fields for
cellular infiltration at x400 magnification with the following
arbitrary score: 0 (<5 cells per field), 1 (5 to 25 cells per field),
or 2 (>25 cells per field).
Scanning Electron Microscopy
To determine differences in
endothelialization
and structure of the intimal surface between PDT and untreated
allografts, scanning electron microscopy of longitudinal sections was
performed at 4 weeks. After overnight fixation in buffered 4%
glutaraldehyde, the specimens were dehydrated in
increasing concentrations of ethanol. After critical-point drying and
coating with a thin film of gold palladium, the sections were examined
with a scanning electron microscope (Amray 1400).
Statistics
Data are expressed as mean±SEM. An ANOVA
with Bonferroni's
correction was used to compare morphometric parameters of
the three untreated allograft controls: NT, LO, and PO. The
inflammatory scores were analyzed with the
2 test. Statistical analysis of the
morphometric data was performed with a two-tailed Student's
t test. Values of P<.05 were considered
significant.
| Results |
|---|
|
|
|---|
Histological and Morphometric
Analyses
No histological differences were observed between
the NT, LO, and PO groups. Peaking at 4 weeks, LO grafts displayed a
large number of lymphocytes infiltrating the adventitia and, to a
lesser degree, the media and intima (Fig 1
), which
demonstrated positive labeling with the monoclonal antibody OX47 to T
cells. At 2 and 4 weeks, an infiltration of white blood cells was also
observed in both PDT and ISO (Figs 1
and 2
).
Unlike NT,
LO, and PO, however, the infiltrate in PDT and ISO consisted primarily
of polymorphonuclear cells. At 8 weeks, this infiltrate persisted
in the adventitia of PDT, whereas ISO presented without
significant inflammation at this time point. There were no noticeable
differences between the number of inflammatory cells in the proximal,
mid, or distal graft segments.
|
|
Inflammatory degeneration of the media
with fragmentation of the
internal elastic lamina was noted at 2 weeks and was highest at 8 weeks
in LO. Elastic laminae were no longer in parallel orientation, and the
space between the laminae was widened (Fig 3
).
Regardless of treatment or the time interval, all allografts
demonstrated thinner medial layers than their respective isograft
counterparts (Table 1
). Smooth muscle cells in the media
of LO diminished over time and were almost undetectable by 8 weeks. In
PDT, however, smooth muscle cells were eliminated entirely from the
media and could not be seen at any time (Fig 1
). The media
appeared to
be compacted and acellular. An amorphous eosinophilic material was
observed between the elastic laminae, and the internal elastic lamina
remained intact. Although medial areas of non-PDT and PDT grafts were
equivalent at 8 weeks (Table 1
), external diameters of non-PDT
grafts
increased significantly, by approximately 47% versus 13% in PDT
(P<.02, Table 2
). In contrast, medial smooth
muscle cells appeared to be viable in syngeneic controls, and medial
areas and external diameters remained almost unchanged.
|
|
|
Intimal hyperplasia was present in LO at all time intervals, often unevenly distributed and mostly confined to graft segments near the proximal and distal anastomoses. The IH was hypercellular at 2 and 4 weeks, whereas by 8 weeks it was less cellular and contained mostly extracellular matrix. At 8 weeks, the intimal area was notably increased in LO with 0.23±0.02 mm2 versus 0.09±0.02 mm2 in PDT and 0.06±0.03 mm2 in ISO (P<.001). No significant IH was seen in PDT at 2 and 4 weeks.
Scanning electron microscopy at 4 weeks was used to examine
endothelial morphology and repopulation of the luminal
surfaces in PDT and untreated allografts. PDT demonstrated complete
covering with normal-appearing endothelial cells,
whereas untreated allografts presented with abundant
platelets and rare erythrocytes adhering to denuded luminal
surfaces (Fig 4
).
|
| Discussion |
|---|
|
|
|---|
To date, two avenues have been followed to overcome immunologic rejection in transplanted blood vessels: attenuation of immune responses in the host through immunosuppressive drugs22 23 24 and reduction of graft antigenicity with different preservation and preparation techniques.4 5 6 9 10 However, clinical trials applying either approach have not yet demonstrated an unequivocal reduction in the incidence of rejection. The use of immunosuppressants like cyclosporine may considerably reduce leukocyte infiltration or delay the development of intimal hyperplasia but cannot entirely suppress the rejection process.22 23 These drugs also pose additional risks and side effects and are poorly tolerated by the often polymorbid cardiovascular patients.24 Cryopreservation techniques, using liquid nitrogen vapor in conjunction with dimethyl sulfoxide, are the currently favored methods of maintaining viability and structural integrity of vascular grafts.5 6 Nevertheless, significant antigenicity of cryopreserved veins and arteries still represents a major obstacle to their widespread use.25 Application of fixatives like glutaraldehyde appears to induce nonspecific inflammatory responses in the adventitia and structural degradation and calcification of the graft. Additional cross-linking of extracellular matrix proteins leads to anastomotic compliance mismatch and subsequent intimal hyperplasia.9 Moreover, fixatives have not been able to homogeneously penetrate the full thickness of the arterial wall, resulting in nonuniform fixation.9
Another approach is decellularization of arterial allografts and xenografts induced by the detergent SDS.15 It reduced the immune injury with preservation of medial elastin in the presence of very few adventitial inflammatory cells. The presence of inflammatory cells, despite complete decellularization, was attributed to minor antigenicity of extracellular matrix components, which are not targeted by the detergent.
Our previous studies,13 which demonstrated PDT-induced depletion of cells and additional modification of the remaining extracellular matrix in the arterial wall, prompted this investigation by suggesting that PDT may significantly decrease graft antigenicity. PDT can be considered a specialized and targeted form of chemotherapy in which the drug itself is harmless unless activated by exposure to wavelength-specific light. Alternatively, PDT may also be considered an extension of radiotherapy in which, because of prior sensitization of the tissue, cytotoxicity is induced at much lower radiation energies than with conventional radiotherapy. Similar to ionizing radiation, the PDT effect occurs within a very short time frame and is localized to the irradiated field. Therefore, PDT appeared to be highly suitable for an ex vivo treatment under standardized conditions, right before graft implantation.
For the present study, the potent, water-soluble photosensitizer CASPc was used. In previous vascular PDT studies, this drug has been shown to possess advantageous properties for this application.13 14 26 Donor animals were injected with CASPc 24 hours before harvest, when they are still known to have elevated photosensitizer concentrations in the vessel wall.26 Loading of the allograft with in vivo exposure to CASPc in the donor was intended to induce an expeditious, uniform transmural distribution of the photosensitizer in the artery. In fact, a pilot study of PDT allografts (data not shown) that were pretreated with in vivo loading and harvested 1 day after transplantation displayed complete decellularization. However, it is unknown whether this result could have also been achieved by ex vivo immersion of the arteries in CASPc solution.
The 675-nm light used for PDT has excellent penetration in tissue and minimal attenuation in the millimeter range.27 Its low absorption and higher scattering coefficients render it highly suitable for the production of vascular grafts even in larger arteries.
MHC antigens in endothelial and medial smooth muscle cells are the primary targets of the rejection process in vascular grafts.7 Removal of the graft from the donor and acute rejection are responsible for the rapid destruction of the blood vessel endothelium.7 28 Hence, the remaining smooth muscle cells are the principal sites of chronic rejection in transplanted arteries.9 Because PDT leads to depletion of all cells in the vessel wall within 24 hours, potential histocompatibility antigenbearing targets were thought to be absent shortly after implantation. It can therefore be inferred that infiltration of mononuclear cells, which would have indicated an ongoing rejection process, was obviated as a consequence of total smooth muscle cell eradication after PDT. However, assuming that the cells were still viable immediately after the PDT treatment and probably not completely eliminated at the time of implantation, one would have to expect many of the soluble products of cell lysis to be altered by PDT, dispersed before initiation of an immunologic response, or most likely both. At 8 weeks after implantation, a small number of polymorphonuclear cells still persisted in the adventitia of PDT grafts. The definite etiology and significance of this observation remain unclear. It can only be hypothesized that possible antigens in the extracellular matrix or cellular debris, which had not been targeted by the preceding PDT treatment, may have contributed to this inflammatory response.15
The structural integrity of the elastic laminae in PDT grafts remained
intact at all time points, whereas untreated allografts demonstrated
focal fragmentation (Fig 3
). Although medial thickness in PDT
and
control allografts was equivalent, significant graft dilatation was
seen only in control allografts. These data confirm previous findings
that acknowledge the importance of mechanical stress to a thinner
arterial wall but attribute aneurysmal dilatation
primarily to immunologically induced destruction of medial
elastin.15 Moreover, other studies on the effect of
elastase and concomitant inflammation verified temporal correlation
between inflammatory infiltrate, destruction of elastic laminae, and
dilatation.29 30 Nevertheless, at 8 weeks, external
diameters of PDT grafts increased by approximately 13%, whereas ISO
remained almost unchanged. Although it cannot be ruled out that graft
dilatation after PDT occurred on the basis of an insidious rejection
process, previous studies corroborate intrinsic, PDT-induced diameter
increases31 32 : by maintaining a "crimp"
geometry of
elastic fibers, smooth muscle cells reduce direct stresses on similarly
aligned collagen. When smooth muscle cell tone is eliminated (eg, after
PDT treatment), the vessel diameter increases, stiffer collagen fibers
are directly stressed, and circumferential compliance is reduced.
In the present study, the area of IH in PDT at 2 weeks was
significantly less compared with LO and at 4 weeks was equivalent to
ISO and LO (Table 1
). At 8 weeks, however, IH in PDT grafts,
although
almost unchanged to 4 weeks, was increased compared with ISO. A
possible compliance mismatch between graft and adjacent
arterial segments, caused by the loss of smooth muscle cell
tone31 32 and the increased structural compactness of
adventitial collagen after PDT,33 may contribute to the
comparably larger intimal areas in PDT grafts by 8 weeks. The degree of
compliance mismatch present in the graft body affects the amount of
intimal hyperplasia that develops.34 Although compliance
of the allografts was not evaluated, the inferred compliance
differences may explain the development of moderate IH in PDT
grafts.
As indicated by scanning electron microscopy, the luminal surface of PDT grafts was completely reendothelialized at 4 weeks. This finding suggests that PDT provides a nonthrombogenic luminal surface, which allows migration of repopulating endothelial cells and may help inhibit subsequent development of IH. The resulting decellularized graft, lined by endothelial cells that probably minimize thrombus formation, represents a prerequisite for satisfactory long-term graft performance. On the other hand, untreated control allografts, which present with no coherent endothelial cell lining but significant adhesion of platelets, are most likely prone to thrombus formation, intimal hyperplasia, and finally graft failure.
In conclusion, PDT pretreatment of allografts quantitatively suppressed histological markers of arterial wall immune injury: adventitial inflammation, aneurysmal dilatation, and the development of intimal hyperplasia. PDT seems to be a safe method of producing a biocompatible and nonthrombogenic arterial scaffold for use as a bypass graft. However, further studies of PDT and graft handling are needed to determine which role this novel approach will ultimately play in the development of vascular bioprostheses.
| Acknowledgments |
|---|
The formulas to correct midgraft diameters for animal growth are as follows:
![]() |
where CAn is the correction factor for midgraft diameter in one animal, D1p is the diameter of the host artery proximal to the graft at implant, D1d is the diameter of the host artery distal to the graft at implant, D2p is the diameter of the host artery proximal to the graft at harvest, and D2d is the diameter of the host artery distal to the graft at harvest.
![]() |
where CExp is the mean correction factor for each experimental group (eg, PDT, LO, or ISO) and n is the number of observations in one experimental group.
![]() |
where Dc is the corrected midgraft diameter at harvest and D2Mid is the measured midgraft diameter at harvest.
Received December 19, 1994; revision received March 6, 1995; accepted April 5, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. J. Granville, B. A. Cassidy, D. O. Ruehlmann, J. C. Choy, C. Brenner, G. Kroemer, C. van Breemen, P. Margaron, D. W. Hunt, and B. M. McManus Mitochondrial Release of Apoptosis-Inducing Factor and Cytochrome c During Smooth Muscle Cell Apoptosis Am. J. Pathol., July 1, 2001; 159(1): 305 - 311. [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] |
||||
![]() |
T. Tojo, K. Niwaya, N. Sawabata, K. Kushibe, K. Nezu, S. Taniguchi, and S. Kitamura Tracheal replacement with cryopreserved tracheal allograft: experiment in dogs Ann. Thorac. Surg., July 1, 1998; 66(1): 209 - 213. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |