Folia Microbiol. 44 (5), 567-578 (1999)

ht.t,p ://www .b•

e a s . ez/rabu/fo:l.•

Ultrastructure of Rat Aortic Grafts P. ROSSMANNa, J. L,~CHAb alnstitute of Microbiology, Academy of Sciences of the Czech Republic, 142 20 Prague bDepartment of Nephrology, TransplantCenter of the Institutefor Clinicaland ExperimentalMedicine. 140 21 Prague

e-mail [email protected] Received 22 February2000 ABSTRACT. This study compares the ultrastructure of three syngeneic and three allogeneic grafts of rat abdominal aorta (Lewis to Lewis and BN to Lewis, respectively); the tissue was sampled three months after transplantation (TPL). The endothelial plate was preserved and mononuclear cell adherence was absent. In syngeneic grafts the intima and media remained close to normal with well-preserved smoothmusclecells (SMC). The thickenedallograftneointimaconsistedof elongatedspindle cells and rich intercellularmatrix.

The cells weretypical SMC withoutapparentsigns of dystrophyor degeneration.On the other hand, most SMC of the media showed completedisruptionand disorganizationof membraneand organellessuggestiveof accomplishednecrosis. However,the frameworkof elastic lamellaewas preserved,without apparent ruptures or lytic changes. Intraintimal migration of medial SMC was not recorded while some cytoplasmic strips were seen to extend across the outer elastic lamella (possible rudimentary outgrowth of SMC?). Lymphocytesand histiomonocyticcells (macrophages)were found in the adventitia but not in the destroyedmedia. Thus electron microscopyelucidatedthe histologicalpictureof "anuclearallografimedia"and confirmed the predominanceof SMC in the thickened neointima. However,signs of the mediointimalSMC invasionwerenot apparentthreemonths postTPL. Abbreviations ASMA CHSPG CyA mAb PAS

a isoform of smooth muscle cell actin chondroitin sulfate-proteoglycan cyclosporin A monoclonal antibody periodic acid-Schiff staining

PCNA PDGF

SMC TPL

proliferatingcell nuclearantigen(cyclin) platelet-derivedgrowthfactor smooth musclecell(s) transplantation

The lesion of vessels, especially of small to medium calibre muscular arteries, is the morphological hallmark of human renal, cardiac, pulmonary and pancreatic allograRs; it exerts a substantial influence on the level and duration of graft function (Demetris et al. 1989; Russell et al. 1993). Hyperacute or "early accelerated" rejection, producing necrotic and thrombotic vascular lesions, is very rare. The early rejection damage with its typical intraarterial adherence of monocytes and lymphocytes and progressive obliteration of lumen has been significantly restricted and/or mitigated by modern immunosuppressive drugs (Rossmann et al. 1997). Thus the most important category of graft vascular lesion is at present the chronic, torpid and hitherto incurable wall thickening ("intimal fibrosis", "sclerosing transplant vasculopathy") of muscular arteries which leads to stenosis and obliteration of lumen within months or years post transplantation (TPL). Its long-term result is a progressive decline in graft function (kidney) or sudden graft failure (heart) (Russell et al. 1993; Isoniemi et al. 1994; Seron et al. 1994; Mihatsch et al. 1999). The pathogenesis of this "late graft endarteritis" remains far from elucidated: It is a probable combination of rejection and non-rejection insults, and seems to become autonomous, "self-progressive" at a certain stage of development (Tullius and Tilney 1995; H~iyry 1998; Tilney et al. 1998). Similar patterns of intimal thickening can be elicited in a series of experimental models based either on allogeneic TPL (Mennander et al. 1991) or on an injury to autologous vessels (e.g., by ballooning, cathetrization, angioplasty or ischaemia; Clowes 1999). The experimental research aims at suppression or mitigation of arterial damage with an effective clinical treatment or prevention in view. In our previous study (Rossmann et al. 1999) we compared the microscopic changes of rat aorta 2-5 months following a syngeneic or allogeneic TPL. The technically difficult microsurgery produced an inflammation with mononuclear infiltration (chiefly in the adventitia) which, however, largely subsided in the syngeneic grafts; the microscopic structure of these grafts returned to nearly normal after three months and later. On the other hand, the aliogeneic vessels displayed a prominent and persistent loss of nuclear staining in the smooth muscle cells (SMC) of the media; there was also a complete depletion of ct-SMC-actin (ASMA) from their cytoplasm. The allogeneic intima exhibited concentric intimal thickening by a highly cellular tissue consisting of longitudinally oriented spindle cells and a plentiful intercellular matrix devoid of elastic lamellae. The cells of this neointima strongly expressed ASMA and also the "tissue macrophage" (ED 2) surface epitope 4-5 months post TPL. The matrix substance gave an intense "/-metachromasia with toluidine blue, was stained with alcian blue and yielded a diffuse reaction with the anti-chondroitin sulfateproteoglycan (CHSPG) monoclonal antibody (mAb).

568 P ROSSMANN and J LACHA

Vol. 44

The present article addresses three problems: (i) the ultrastructural correlate of "anuclearity" of the allogeneic media (suspected complete destruction and disruption - i.e. accomplished SMC necrosis?); (ii) a closer characterization o f neointimal ceils (fine structure typical o f SMC?) and detection o f their possible discrete, microscopically undetectable or overseen dystrophic changes, and (iii) search for signs o f cellular spread from the media (mediointimal SMC invasion) which has been repeatedly described by others (Isoniemi et el. 1994; Hancock 1998) but was not apparent under the light microscope in our histological sections. A conclusive answer was obtained to the former two questions but not to the third one.

MATERIALS AND METHODS

The experimental series involved three allogeneic and three syngeneic TPL o f abdominal aortic segments (adult rat males, 200-220 g). The selection o f donors and recipients (BN RT-I n ---* Lewis RT-11 and Lewis ~ Lewis; Charles River, Germany) and the microsurgical technique matched those used in our previous study (Rossmann et al. 1999). No immunosuppressive medication was administered. The aortic grafts were harvested three months post TPL under general anesthesia (Sulfetanyl, Janssen). One half o f each sample was fixed in buffered 5 % formaldehyde and embedded in Histoplast S (Serva, Heidelberg, Germany). The 5-1am sections were stained with HE, PAS, toluidine blue and resorcinfuchsin for elastic fibers. For electron microscopy small blocks o f tissue (about 1 mm 3 ) were fixed in 2.5 % glutaraldehyde (0.15 mol/L cacodylate buffer-sucrose, pH 7.3; 1 h at 4 ~ rinsed three times in the same buffer (2 • 10 rain and overnight), postfixed in 1 % OsO4 (cacodylate buffer; 1 h at 4 ~ dehydrated in an ethanol-acetone series and embedded in Vestopal W polyester resin (Serva). Semithin sections (0.1 !am) were stained with alkaline toluidine blue. Selected areas o f intima and media were sectioned with a diamond knife (Ultracut E Microtome, Leica-Reichert, Vienna; silver to creamy interference color) and contrasted with uranyl acetate and lead citrate. The sections were examined under a Philips CM 100 microscope and photographed on Agfa Scientia 23D56 films.

RESULTS

Light microscopy The microscopic picture largely reproduced that seen in our previous study (Rossmann et al. 1999). In short, two syngeneic grafts displayed a thin intima of unmeasurable thickness, while discrete focal patches and cushions o f myointimal cells were recorded in the third one (20/am or less in thickness; Fig. 1 left). The nuclei o f medial SMC remained conspicuously preserved and uniform, as did the elastic lamellae which were faintly contoured by thin PAS-positive rims. Sporadic residual infiltrates (grade 0.5 o f semiquantitive 0-3.0 scale) were seen in the adventitia.

Fig. I. Left: Light microscopyof a syngeneicgraft. Well-preserved endothelial lining, normal architecture and cellularity of the tunica media (the ultrastructure of SMC is displayed in Figs 2 and 3). Minimal round-cell cellularity of the adventitia (bottom). Toluidine blue; Right: Pronouncedthickening of intima with a uniform population of transversally sectioned myointimal cells (ultrastructure in • Figs 4 and 5) embedded in a "hyaline matrix" expressing red-violet metachromasia. "Anuclear" appearance of media (lower center) except for scarce karyorrhectic remains of SMC nuclei (cf. Figs 6 and 7). On the bottom right is a cluster of SMC-like (elements) adjacent to the outer surface of the external elastic membrane (ultrastructure of a similar focus is displayed in Fig. 8). A persisting round-cell infiltrate is seen in the adventitia (bottom). Same staining as on the left; x230.

1999

ULTRASTRUCTURE OF RAT AORTIC GRAFTS 569

The allogeneic grafts were devoid of parietal thrombosis and leukocytic adherence. Tile characteristic neointima (60-100 lain in thickness) consisted of uniform elongated myointimal cells (most densely packed in the subendothelial area) and a strongly metachromatic and PAS-positive "hyaline" matrix enhanced in the deep portion (Fig. 1 right). Sporadic mononuclear leukocytes appeared in the same location whereas foam cells and atheromatous structures were lacking. The media displayed a typical "anuclearity" (focal rests of SMC nuclei persisted in only one sample) and gave a diffuse PAS staining. The elastic lamellae were preserved except for focal lifting of the external membrane which was undermined by adjacent layers of mononuclear elements. An intense focal to confluent cellularity (grade 2.0-2.5) was noted in the adventitia.

Electron microscopy of syngeneic grafts In the syngeneic grafts the endothelial layer was preserved and the individual cells were devoid of dystrophy or sloughing. Some smooth-walled vacuoles and vesicles were seen in the cytoplasm. The subendothelial space (i.e. between the endothelial cell membrane and the inner elastic membrane) was very narrow (1.0-1.5 lam or less in width); it consisted of a loose meshwork of delicate nonperiodic fibrils with interpolated translucent spaces and slits. No signs of intraintimal muscle cell penetration were seen in the inner elastic membrane. The medial smooth muscle cells (Fig. 2) expressed an uninterrupted cellular membrane with focal adjacent vesicles. The main structure of the cytoplasm was a dense network Of nonperiodic microfibrils with scattered patches of condensation, also apparent in the cell periphery ("fusiform densities"). Clusters of ribosomes and scattered ergastoplasmic vesicles were found in the perinuclear areas (Fig. 3). The mitochondria displayed focal eedema o f matrix without disruption of cristae. Vacuoles with dense material (lipid, lipopigment) remained sporadic, and phagolysosomes were absent. No chromatin clumping was seen in the nuclei and the perinuclear spaces were narrow. The intercellular spaces of the media were framed by translucent elastic lamellae with fine bufidles of collagen fibers. Small collections of dense granular material were devoid of cell debris or membrane fragments; myelin figures were exceptional. The collagenous component was accentuated in the adventitia close to the external elastic membrane.

Fig. 2. Tunica media of a syngeneic graft. Well-preserved leiomyofibrils and intact cell membranes of SMC. Elastic lamellae (translucent bars; top center and left) and bundles of collagen are seen in the intercellular spaces; • 12200.

570 P ROSSMANN and J. L,&CHA

Vol. 44

Fig. 3. The cytoplasmof SMC in syngeneicgraft media displays a dense regular meshworkof leiomyofibrils. The cytoplasmicmembrane and basement membrane(upper and lower left corners) have a normal ultrastructure. Ribosomesand rough endoplasmictubules are seen in tile perinuclearspace (lower right); • 100.

Electron microscopy of allografts The allografts also displayed an uninterrupted endothelial covering without signs of pronounced dystrophy or desquamation. Minute dense non-myelin corpuscles and pinocytotic vesicles were observed in the thickened cytoplasmic plates. The thickened (neo)intima featured multiple elongated spindle cells (Fig. 4) harboring a regular dense network of myofibrils typical of SMC, with well-formed fusiform densities (Fig. 5). Groups of ribosomes, some profiles of ergastoplasmic reticulum and mitochondria (partly with matrix oedema) were noted in the perinuclear spaces. The intercellular compartment of the neointima consisted mainly of translucent flocculent matrix, of"cedematous" appearance incorporating a loose texture of fine nonperiodic fibrils (about 50 nm in width). Besides, minute granules of very high density were scattered in the fibrillar mesh. Collagenous fibres, especially in the form of bundles or strands, were lacking, as were structures corresponding to (newly lbrmed) elastic lamellae seen in the media. The media was distinctly thinned. SMC exhibited a general disruption of ultrastructure with fragmentation, disorganization and clumping of membranes, organelles and nuclear residues (Fig. 6). Extensive areas of debris with some dense lipid vacuoles and myelin figures lay between the elastic membranes (Fig. 7). Dense rings rimmed with fine granularities (probably calcification) were focally seen. Areas of partly preserved (thinned) strips, patches and tongues of membrane-limited SMC with dense shrunken cytoplasm appeared in the outer (periadventitial) layers rather than in deep portions of the media. These remains of SMC enclosed some ribosomes or ergastoplasmic profiles, swollen translucent mitochondria and numerous dense lipid vacuoles of varying size. The myofibrils were highly restricted and reduced to the periphery of cytoplasm; they were absent from most cells. Phagolysosomal bodies were not found. Unlike SMC, the elastic membranes of the media were preserved and devoid of apparent ruptures or fragmentation. Bundles of adjacent collagen fibers were seen in some places. No elements corresponding to granulocytes, lymphocytes and macrophages were recorded in the destroyed media. Penetration of medial cells through the inner elastic membrane into the neointima (even in the form of slender cytoplasmic strips) was not seen in any of the localities examined. On the other hand, elongated cells were seen in the adventitia close to the

1999

ULTRASTRUC'FURE OF RAT AORTIC GRAFTS

5"/1

Fig. 4. The allograft neointima features uniform elongated cells devoid of dystrophic change or disruption They are incorporated in an abundant intcrcclhtlar matrix offlocculent appearance; x7000.

Fig. 5. The myointimal cell is a typical SMC exhibiting focally condensed cytoplasmic leiomyofibrils, a well-preserved cell membrane and a basement membrane. Minute granules of high density are scattered in the intercellular space (top left); x49200.

572

P t~,OSSMANN and J L,~CHA

Vol. 44

Fig. 6. lunica media of an allogeneic graft. General disruption and disintegration of SMC organelles and membranes Lipid vacuoles are seen i~1 the cell remains (bottom center). The elastic lamellac are preserved. The outer margin of the media is in the upper left comer: x 390(

Fig. 7. Detail of a necrotic SMC shows cell debris, myelin figures (bottom right) and a small calcification (top left). On the top is parl of a shrunken cell with dense cytoplasm. Collagen bundles are seen in the intercellular space; x20500.

1999

ULTRASTRUCTURE OF RAT AORTIC GRAFTS

573

Fig. 8. Fhc outer la~,er of allograft media (the top half is adjacent adventitia) Strips of cytoplasm (residual SMC?) are seen at the bottom C3toplasmic bridges are squeezing across the outer elastic lamella (upper left and lower right); x 10 I00.

Fig. 9. A small vessel m allograft adventitia (probably a venule) exhibits well-preserved SMC (upper right) alternating with thick bundles of collagen fibrils. The endothelial plate is not damaged (two nuclei are seen on the lower left and bottom right); x9200.

574

P. ROSSMANN and J. L,~.CHA

Vol. 44

outer elastic membrane. Their cytoplasm harbored plentiful ribosomes, partly also in form of typical ergastoplasmic profiles. Cytoplasmic fibres resembling leiomyofibrils remained scarce and if present, they only appeared close to the cytoplasmic membranes. These elements further exhibited clusters of smoothwalled vesicles and some mitochondria but no phagolysosomes. They had a fine structure of nuclear chromatin without peripheral condensation, and nucleoli were seen in some cells. Superficial microvilli or filopodia were lacking, as were collagen bundles in the pericellular space. Cytoplasmic strips of similar description appeared in the external elastic membrane and adjacent areas of the media (Fig. 8). Small blood vessels were recorded in the adventitia. One arteriole showed thickened but intact endothelium with numerous vesicles and ribosomes, another vessel (probably a venule) displayed a thin endothelial plate and inconsistent musculature with numerous bundles of collagen fibers. In both cases the" well-preserved SMC had a dense cytoplasmic net of myofibrils and intact cell membranes (Fig. 9). An extensive examination of the adventitial inflammatory infiltrate was not undertaken. The cell population consisted of lymphocytes and histiomonocytic cells - macrophages. The former cells displayed the usual condensation ofchromatin and a narrow cytoplasmic rim; "blastic" elements with plentiful (poly)ribosomes were not seen. Single mast cells were recorded whereas plasmocytes and granulocytes were lacking. The hallmarks of histiomonocytic cells wer~ plentiful (though not diffuse) surface microvillosities, fine chromatin structure in elongated nuclei, numerous small finely granular dense corpuscles and well-developed Golgi zones; there were no prominent phagolysosomes. A foreign-body giant-cell granuloma was encountered in one sample. The central surgical suture fiber consisted of homogeneous "hyaline" material of low density, encircled by a tightly adherent multinucleated cell. The microvillosities characteristic of macrophage (3-4 ~m long) were seen on the opposite "free" cell surface whereas the adherent portion was smooth or only emitted scarce filopodia (0.5 tam or less). The giant cell cytoplasm exhibited multiple organelles (smooth-membrane vesicles, mitochondria, granular microbodies, Golgi vesicles) but ribosomes were scarce and phagolysosomes were completely lacking.

DISCUSSION Ultrastructure o f the intima

The three-month post-operation interval chosen here represents a late post TPL period at the end of which the early nonspecific inflammation has become largely extinct and the damage produced by surgery has been repaired. Thus the parietal thrombosis, endothelial sloughing and leukocytic adherence were seen in neither syngeneic nor allogeneic aortic grafts. These lesions are likewise virtually absent from the "late sclerosing vasculopathy" or the "late endarteritis" of human renal allografts (Mihatsch et al. 1993; Rossmann et al. 1997). On the other hand, the intimal architecture of syngeneic aortae greatly differed from that of allografts. In the former case the intima, hardly recognizable by light microscopy, did not exceed 1-2 !am in width, as compared to the thick allograft neointima, which reached 60-100 ~m or even more (Geerling et al. 1994); this latter structure consisted of a dense mesh of spindle cells embedded in a highly metachromatic "hyaline" matrix, The electron microscopy confirmed that the neointima elements were ~pical SMC with a dense net of leiomyofibrils, which corresponds exactly to the massive expression of ASMA documented earlier by immunohistochemistry. These myointimal cells did not exhibit ultrastructural signs of dystrophy: They displayed a regular contractile cytoskeleton and well-preserved cytoplasmic and basement membranes. A distinct ribosomal and ergastoplasmic component, absent from the normal "sessile" SMC, points to a growth activity and correlates well with the positive proof of PCNA and Ki 67 epitopes (Rossmann et al. 1999). Unlike the genuine SMC of the media, the neointimal SMC did not produce elastic lamellae and collagen bundles, and were parallel to the long axis of the vessel. As noted below, the term "intimal fibrosis" does not seem appropriate. A similar longitudinal alignment of neointimal cells was also noted in the coronary arteries of heart allografts (Russell et al. 1993; Jenkins et al. 1997). The ultrastructure of the intercellular matrix displayed translucent "hydropic" slits alternating with delicate nonperiodic fibrils but (as mentioned above) lacking in fibrillar collagen. Scattered minute very dense granules were reminiscent of those seen in "mucoid oedema" of small arteries ascribed to the toxic (CyA-associated) lesion of renal allografts which also displays a massive metachromasia (Rossmann et al. 1991). This pattern points to a strong increase in acidic glycans rather than to the banal "collagenous scarring", and is fully in line with the strong positivity of chondroitin sulfate-proteoglycan recorded in our above-mentioned study. Consequently, the myointimal cells appear to be stimulated SMC. The growing SMC are normal medial elements in the foetal and early postnatal periods (Hfiyry 1996); they resemble fibroblasts in light microscopy and produce an

1999

ULTRASTRUCTURE OF RAT AORTIC GRAFTS

575

abundant matrix. They are sensitive to growth factors (e.g., PDGF) which in turn they can produce. Embryonal myosin and surface adhesion molecules were proved in the neointimal SMC of transplanted rabbit carotid arteries (Sai et al. 1999). The highly hydrated matrix of this "young" intima is a boardwalk for SMC and a reservoir of mediators (growth factors). In summary, the allograft neointima effects a progressive proliferative process without apparent signs of dystrophy, even if the proof of ASMA p e r se does not prove the leiomyocellular proliferation; its expression was also found in the interstitial fibrosis of chronic rejection (Pedagogos et al. 1997). Phenotypic transformation and intraintimal migration of medial SMC stimulated by growth factors released in the alloreactive inflammation, or following a mechanical injury, has been stressed as the starting event in the neointimal hyperplasia (Mennander et al. 1991). The chain of activation involves ET 1 endothelin discharged from the damaged endothelium (Tanabe et al. 1997), partial splitting of NFKB molecule with phosphorylation of the transcription factor, and subsequent shift in SMC phenotype (Hancock et al. 1998; Yamashita et al. 1999). This is an acute process; "SMC intimal colonization" has been noted as early as 1-2 d after a balloon catheter arterial injury (Hansson 1994; Clowes 1999). In our first study no mediointimal outgrowth could be detected by light microscopy, and no cellular structures extending across the inner elastic membrane were recorded electronmicroscopically 3 months post TPL in the present series. In fact, as will follow, the destroyed medial SMC seem incapable of any local extension. Ultrastructure o f the media and adventitia

The microscopic "anuclear appearance" (karyolysis) of allogeneic media, basically different from the well-preserved syngeneic SMC, raised a suspicion of severe cell damage; it was enhanced by the finding of total ASMA depletion signifying a breakdown of the cell contractile system. Two questions for electron microscopy ensued: (i) confirmation of the irreversible destruction of media, and (ii) detection of possible remains of intramedial leukocytic invasion. In the syngeneic grafts the almost normal SMC ultrastructure fully correlated with the light microscopic integrity and strong ASMA expression. It follows that syngeneic TPL can leave a morphologically (and probably also functionally) intact vessel, as also concluded by H~iyry et al. (1996). The allograft media displayed a radically different picture characterized by disruption of organelles and membranes, and virtual disappearance of leiomyofibrils suggestive of the complete destruction of SMC; however, we failed to document the presence of leukocytic invasion or its unequivocal remains as well as dense granular masses seen in immune aggregate deposits or "fibrinoid necrosis". Unlike SMC necrosis, the medial elastic membranes retained their integrity which contrasts with the elastolytic-elastoclastic changes known in many entities of human or experimental arteritis (aortitis). The preservation of the elastic framework explains the absence of graft rupture or aneurysm, as also recorded in our previous series of more than 90 grafts, in spite of persisting medial necrosis, paralysis and distension. However, aortic graft bulging had been observed by others (Schnitz-Rixen et al. 1988). As mentioned above, the disorganization of SMC in the given post TPL interval yields no morphological substrate for cellular replication and intimal outgrowth; even the ultrastructure revealed no signs of such a process. On the other hand, narrow cytoplasmic bridges and strands were seen in the outer elastic membrane. In view of (weak and focal) ASMA positivity at the medioadventitial borderline, a rudimentary expansion of SMC could have taken place in this area. However, the origin or lineage of these elements remained unclear; no cells identical with or similar to the myointimal SMC were identified in the adjacent adventitia. The damage to the allograft medial SMC was documented recently in the functional and morphological study by Bigaud et al. (1999). The authors tested the contractility of aortic wall samples challenged ex vivo with vasoconstrictors such as phenylepinephrine, endothelin 1 or KCI. The response of the syngeneic media was preserved (irrespective of endothelial abrasion) even 8 weeks post TPL. On the other hand, in the allografts it was largely restricted from the 3rd day on and completely extinct after 4 weeks; SMC paralysis distinctly preceded the microscopic hypo- or anuclearity of media seen two weeks and later but not one week post TPL. Remarkably, the neointimai thickening was only found close to the peripheral sutures after four weeks and appeared to be diffuse no sooner than two months post TPL. Thus necrosis of the media obviously took place much earlier than the myointimal proliferation which almost excludes a mediointimal SMC invasion and is in full agreement with our observations.

576 P ROSSMANN and J. L,/kCHA

Vol. 44

Conclusion, pathogenetic aspects

The exclusive occurrence of the above-mentioned intimal and medial changes in allografts must raise the suspicion of rejection origin. The correlation between the extent of aortic damage and the degree of histoincompatibility were stressed by others (Soots et al. 1998). The inflammatory origin of damage is also substantiated by the proof of oxygen radicals producing oxidized l ipids, very harmful to the stimulated SMC (Dimeny et al. 1995). Vascular damage of heart allografts can be mitigated by exclusion of the "/-interferon gene (Sokolowski et al. 1998) although this intervention was not effective in other studies (Russell et al. 1997). A complicated interplay of immune mechanisms emerges from the recent xenoTPL study carried out by Lorber el al. (1999). Segments of human arteries of appropriate calibre were transplanted into the abdominal aortae of immunodeficient SCID/beige mice. The implants were microscopically unaltered after 28 d whereas the passive transfer of human peripheral blood mononuclear leukocytes (done 14 d post TPL) elicited "endothelialitis", intimal leukocytic infiltration and proliferation of ASMA + cells in most grafts. Interestingly, these neointimal elements expressed human HLA epitopes; necrosis of medial SMC was not mentioned. Blood pressure was not monitored in our animals but a condition comparable to the focal necrosis of media in spontaneously hypertensive rats seems unlikely (Ryffel et al. 1986). As distinct from human beings, the aorta of rodents lacks the vasa vasorum. Thus the (initially) oedematous and infiltrated intima could raise a serious obstacle to the diffusion of nutrients, possibly Up to the start of ischaemic necrosis (Heemann et al. 1998). The prominent feature of late arteriopathies in both human and experimental allografts is slow progression with a probable point of no return (H~iyry et al. 1999) when the process becomes autonomous, and does not stop even after regrafting into a syngeneic recipient (Schmid et al. 1996). In the late stage there is a restriction or extinction of inflammatory infiltrate and the evidence of an ongoing allospecific humoral attack is scarce (if not absent) (H~iyi'y 1998; Wilhelm et al. 1998; Yamashita et al. 1999). In our earlier unpublished tests, rat Ig was detected in rat allografts with a FITC-conjugated swine antibody. The conspicuous negativity of the neointima contrasted with an intense diffuse positivity ofadventitia (possible nonspecific exudation in "periaortitis?") (Tilney et al. 1998). Plasma cells were sporadic or absent. The destroyed media revealed granular positivity of the outer layer. The anti-rat-C3 antibody (biotinavidin-peroxidase assay) gave a pronounced general (background?) staining. Elution tests of Ig and proof of in vitro tissue binding were not performed. The pattern of medial Ig binding differed from the well-known picture of "hypersensitivity angiitis" and could merely represent a non-immune sticking and intramedial penetration of the adventitial exudate. Necrosis of the media being absent from most long-functioning human allografts, the lesion of prime importance is (irrespective of its origin and mechanisms) the neointimal thickening - the cornerstone problem of today's TPL program. Its effective prevention or arrest has not yet been achieved. The series of measures aimed at the restriction of inflammatory activity and inhibition of myointimal proliferation have been enumerated in our previous article but new sophisticated, possibly promising manipulations are being undertaken. The recent attempts involve inhibition of 3-hydroxy reductase (Katznelson et al. 1999), competitive inhibitioll of PDGF receptor (Poon et al. 1999), modulation of TGF-131 levels by new immunosuppressive drugs or protocols (FellstrSm et al. 1999; Hutchinson et al. 1999), SMC-antiproliferative effects of Tripterygia glycosides (Hachida et al. 1999), and potentially promising inhibition of estrogen 13receptor or type 1,4 somatostatin receptors (H~iyry 1999). As the restitution of the stenosing neointima to the normal intimal thickness is hardly possible, especially in the renal grafts, all attempts at therapeutic intervention must be focused on the initial or early stage of the late graft vasculopathy. Some of the mAbs used in the histochemicalsection of this study (Rossmann et al. 1999) have been kindly supplied by Dr. B. I~i~ai~ov~t(lmmunotech-Coulter Ltd.) Prague,and by Prof. K. Matou~,ovic,MD., DSc. (Department ofNephrology, 2nd Medical Faculty. Charles University, Prague). The authorsthank Dr. V. Pokorn~,,PhD. (Institute of Microbiology, Academy of Sciences of the Czech Republic, Prague)for his expert help with the electronmicroscopyand gratefullyacknowledgeour skilled technicians: Mrs. J. Havli~kov~itbr performingthe aortic TPL, Mrs. M. Li~kovb.for the technical assistance in histology,and Mrs. B. Svobodov~ifor the preparationof the manuscript.The languagerevisionhas been kindly performedby Mr. J. Harries.

1999

ULTRASTRUCTURE OF RAT AORTIC GRAFTS

577

REFERENCES BIGAUD M., SCHRAAE.O., ANDRIAMBELOSONE.: Complete loss of functional smooth muscle cells precedes vascular remodeling in rat aorta allografts. Transplantation 68, 1701-1707 (1999). CLOWES A.W.: Regulation of smooth muscle cell proliferation and migration. Transplant.Proc. 31,810-811 (1999). DEMETRISA.J., ZERBE T., BANNERB.: Morphology of solid organ graft arteriopathy. Transplant.Proc. 21, 3667-3670 (1989). DIMENY E., FELLSTROMB., LARSSON E. et al.: The role of lipoprotein abnormalities in chronic vascular rejection a~er kidney transplantation~ Transplant.Proc. 27, 2036-2039 (1995). FELLSTROM B., BACKMANN U., LARSSON E. et HI.: Are there markers to initiate treatment of chronic rejection? Transplant.Proc. 31, 1796--1798 (I 999). GEERLING R.A., DE BRUIJN R.W.F., SCHERINGAM. et HI.: Suppression of acute rejection prevents graft arteriosclerosis after allogeneic aorta TPL in the rat. Transplantation 58, 1258-1263 (1994). HACHtDA M., LU H., ZHANG X. et HI.: Inhibitory effect of triptolide on PDGF-A and coronary arteriosclerosis after heart transplantation. Transplant. Proc. 31, 2719-2723 (1999). HANCOCK W.W.: Basic science aspects of chronic rejection - induction of protective genes to prevent development of transplant arteriosclerosis. Transplant.Proc. 30, 1585-1589 (1998). HANSSON G.K.: Cytokines regulate proliferation and cytoskeletal organization of vascular smooth muscle cells. Path.Res.Pract. 190, 891-894 (1994). HAYRY P.: Pathophysiology of chronic rejection. Transplant. Proc. 28 (Suppl.), 7-10 (1996). HAYRY P., MYLLARNIEMIM., AAVlK E. et HI,: Chronic rejection - potential new sites of therapy. Transplant.Proc. 28, 3225-3226 (1996). HAYRY P.: Chronic allografl vasculopathy - new strategies for drug development. Transplant.Proc. 30, 3989-3990 (1998). H~,YRY P., AAVlK E., SAVOLAINENH.: Mechanisms of Chronic rejection. Transplant.Proc. 31 (Suppl.), 5-8 (1999). HEEMANNU.W., AZUMA H., TULLIUSS.G. et HI.: Influence of renal mass on chronic kidney allografl rejection in rats. Transplant.Proc. 27, 549-550 (1995). HUTCHINSONI.V.: The role of transforming growth factor-[~ in transplant rejection. Transplant.Proc. 31 (Suppl.), 9-13 (1999). ISONIEMIH., TASKINENE., H,~YRY P.: Histological chronic allografi damage index accurately predicts chronic renal allografl rejection. Transplantation 58, 1195-1198 (1994). JENKINS J., BOYLEJ.J., MCPHADEN A.R., LINDOP G.BM.: Three-dimensional reconstruction of abnormal intramural coronary arteries in human heart allografi biopsy. J.Pathol. 181,247-250 (1997). KATZNELSONS.: lmmunosuppressive and antiproliferative effects of HMG-CoA reductase inhibitors. Transplant.Proc. 31 (Suppl.), 2224 (1999). LORBER MI., WILSON J.H., ROBERTM.E. et HI.: Human allogeneic vascular rejection after arterial transplantation and peripheral lymphoid reconstitution in severe combined immunodeficient mice. Transplantation 67, 897-903 (1999). MENNANDERA., TIISALA S., HALTTUNENJ. et al.: Chronic rejection in rat aortic allografis - an experimental model for transplant arteriosclerosis. Arterioscl.Thromb. l l , 671--680 (1991). MIHATSCHM.J., RYFFEL B., GUDATF.: Morphological criteria of chronic rejection. Transplant.Proc. 25, 2031-2037 (1993). MIHATSCHM.J., NICKELEITV., GUDAT F.: Morphological criteria of chronic rejection. Transplant.Proc. 31, 1295-1297 (1999). PEDAGOGOSE., HEWITSONT.D., WALKERR.G. et HI.: Myofibroblast - involvement in chronic transplant rejection. Transplantation 64, 1192-1197 (1997). POON M., COHEN J., SIDDIQUIZ. et HI.: Trapidil inhibits monocyte chemoattractant protein I and macrophage accumulation after balloon arterial injury in rabbits. Lab.lnvest. 79, 1369-1375 (1999). ROSSMANNP., JIRKA J., CHADIMOVAM. et HI.: Arteriolosclerosis of the renal allografi - morphology, origin, life history and relationship to Cy~, therapy. Virchows Arch.Path.Anat. 418, 129-141 (1991). ROSSMANNP., JIKKAJ., MATOU~OVICK.: Renal Allograft Biopsy - Image, Interpretation, Interventions. Academia Publishers, Prague 1997. ROSSMANNP., LACHA J., LODEREROVAA.: Morphology and immunohistochemistry of rat aortic grafts. Folia Microbiol. 44, 339-353 (1999). RUSSELLM.E., FUJITAM., MASEK MA. et HI.: Cardiac graft vascular disease. Transplantation 56, 762-763 (1993). RUSSEL M.E., RAtSANEN A., SOKOLOWSKIP. et al.: Knockout models of chronic cardiac rejection and graft arteriosclerosis. Transplant.Proc. 29, 2531-2532 (1997). RYFFEL B., SIEGELH, PETRIC R. et HI.: Nephrotoxicity of cyclosporine in spontaneously hypertensive rats - effects on blood pressure and vascular lesions. Clin.Nephrol. 25 (Suppl.), 193-198 (1986). SAI S., IWATAA., THOMAS R., ALLEN M.D.: Vascular smooth muscle cells express VCAM 1 and embryonic myosin isoforms in transplant arteriopathy. Transplant.Proc. 31,863-864 (1999). SCHMID C., HEEMANNU., TILNEY N.L.: Retransplantation reverses mononuclear infiltration but not myointimal proliferation in a rat model of chronic cardiac allograft rejection. Transplantation 61, 1695-1699 (1996). SERON D., MO~SO F.: Transplant vasculopathy in early protocol renal allograft biopsies. Transplant.Proc. 31, 2219-2220 (1999). SCHNITZ-RIXENT., MEGERMANJ., COLVIN R.B. et al.: Immunosuppressive treatment of aortic allografts. ,I. Vasc.Surg. 7, 82-92 (1998). SOKOLOWSKIAR., JENSENT.G., KOGLINJ.: Reduced transplant arteriosclerosis in murine heart allografts placed in interferon "/-knockout recipients. Amer.J.Pathol. 152, 359-365 (1998).

5711 P. ROSSMANN and J. LACHA

Vol. 44

SOOTS A., LAUTENSCHLAGERI., KROGERUS L.: An experimental model of chronic renal allograft rejection in the rat using triple drug

immunosuppression. Transplant.Proc. 65, 42~-6 (1998). TANABE S., UEDA M., HAN Y.S. et al.: Upregulation of endothelin-converting enzyme during the development of transplant renal arteriosclerosis in human renal allografts. Transplant. Proc. 29, 1517-1519 (I 997). TILNEY N+L., KUSAKA M., PRATSCHKE J., WILHELM M.J.: Chronic rejection. Transplant.Proc. 30, 1590-1594 (1998). TULLIUS S.G., TILNEY N.L+: Both alloantigen-dependent and -independent factors influence chronic allografl rejection. Transplant.Proc. 59, 313-318 (1995). WILHELM M.J., KUSAKA M., PRATSCHKE J., TILNEY N.L.: Chronic rejection - increasing evidence for the importance of alloantigenindependent factors. Transplant.Proc. 30, 2402-2406 (1998). YAMASHITA M., DIMAYUGA P++ KAUL S+ et al.: Phosphatase activity in the arterial wall after balloon injury - effect of somatostatin analog octreotide. Lab.lnvest. 79, 935-944 (1999).