Smooth muscle cell injury after cryopreservation of human thoracic aortas

Cryobiology 52 (2006) 309–316 www.elsevier.com/locate/ycryo

Brief communication

Smooth muscle cell injury after cryopreservation of human thoracic aortas 夽 G. Pasquinelli a,¤, L. Foroni b, M. Buzzi c, P.L. Tazzari c, C. Vaselli c, M. Mirelli b, M. Gargiulo b, R. Conte c, A. Stella b a

Clinical Pathology, Department of Experimental Pathology, University of Bologna, Policlinico S. Orsola-Malpighi, Bologna, Italy b Vascular Surgery, Department of Anesthesiological and Surgical Sciences, University of Bologna, Policlinico S. Orsola-Malpighi, Bologna, Italy c Cardiovascular Tissue Bank, Service of Transfusion Medicine, Policlinico S. Orsola-Malpighi, Bologna, Italy Received 2 August 2005; accepted 21 December 2005 Available online 3 February 2006

Abstract The cryopreservation protocol we use for arterial reconstructive surgery has been studied to evaluate smooth muscle cell (SMC) structural integrity and viability before implantation. Samples of human thoracic aortas (HTA) were harvested from Wve multi-organ donors. Sampling included unfrozen and cryopreserved specimens. Cryopreservation was performed using RPMI with human albumin and 10% Me2SO in a controlled-rate freezing apparatus. Thawing was accomplished by submerging bags in a water bath (39 °C) followed by washings in cooled saline. In situ cell preservation as investigated by light and transmission electron microscopy showed that SMCs from cryopreserved HTA had nuclear and cytoplasmic changes. A TUNEL assay, performed to detect DNA fragmentation in situ, showed increased SMC nuclear positivity in cryopreserved HTA when compared to unfrozen samples. 7-AAD Xow cytometry assay of cells derived from cryopreserved HTA showed that an average of 49 § 16% cells were unlabeled after cryopreservation. Organ cultures aimed to study cell ability to recover cryopreservation damage showed a decreasing number of SMCs from day 4 to day 15 in cryopreserved HTA. In conclusion, the cryopreservation protocol applied in this study induces irreversible damage of a signiWcant fraction of arterial SMCs. © 2005 Elsevier Inc. All rights reserved. Keywords: Cryopreservation; Human aortic homografts; Arterial banking; Transmission electron microscopy; Flow cytometry; Immunohistochemistry; Organ culture; TUNEL; Smooth muscle cell injury

The establishment of tissue banking facilities is expected to extend the clinical use of cryostored blood vessels. EVective cryopreservation of tissues is 夽

This work was funded by RFO (ex quota 60%) 2004—Università degli Studi di Bologna—Italy. * Corresponding author. Fax: +39051306861. E-mail address: [email protected] (G. Pasquinelli). 0011-2240/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cryobiol.2005.12.004

a very complex task that is even made more diYcult by the presence of possible tissue and species speciWc diVerences. Post-thawing functional studies performed on cryostored human arteries have documented a reduced contractile force and endothelial function in the graft [9,11,14,15]. The smooth muscle cells (SMCs), the eVectors of blood vessel contractile responses, are reported to be histologically

310

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

unaVected following cryopreservation procedures [15]. The present study was undertaken to evaluate the SMC structural integrity and viability in cryopreserved and unfrozen human aortas. Human thoracic aortas were harvested from Wve multi-organ donors (mean age 40 years; 4 male and 1 female). After collection, the arteries were kept in a sterile box with Celsior media (IMTIX SANGSTAT, Lyon, France), a Xushing and cold storage solution for solid organ preservation, and transferred to the Cardiovascular Tissue Bank in isothermal boxes Wlled with ice within 60 min since procurement. Under a laminar Xow, a 1-cm-long ring was cut from each artery and used as control. Each ring was further cut into three pieces for light (LM), transmission electron microscopy (TEM), and organ culture studies. Controls for the Xow cytometer analysis were not scheduled at the time of the study design. The arteries were prepared, classiWed, and transferred to a solution with RPMI 1640 (Cambrex, Bio Science Vierviers, Belgium) plus antibiotics; according to Strickett et al. [12] a mixture of Mefoxin 240 mg/ml, Lincomycin 120 mg/ml, Colimycin 100 mg/ml, and Vancomycin 50 mg/ml was used; Amphotericin B was not included due to its potential tissue toxicity. After decontamination (72 h at 4 °C), the homografts were transferred into sterile bags containing 100 ml of fresh cryoprotectant solution, i.e., RPMI 1640 with human albumin (Kedrion, Lucca, Italy) and Me2SO at a Wnal concentration of 10%. The solution was cooled at 4 °C for 30 min before its use. The bags were kept at 4 °C for 30 min to allow the Me2SO to penetrate into the tissues completely. The bags were labeled and cryopreserved in liquid nitrogen vapor in a controlled rate freezing system (IceCube 1860, Sy-Lab, Wien, Austria) using an electronically monitored program that allows to decrease the temperature at 1 °C/min to ¡45 °C and at a faster rate thereafter until ¡120 °C has been achieved. The cryopreserved arterial homografts were stored in the vapor phase of liquid nitrogen. Storing in liquid nitrogen was avoided since it induces microfractures as demonstrated by scanning electron microscopy [1] and may introduce the risk of both sample contamination and cross contamination with microbes [8]. Each freezing curve was checked and validated in the laboratory. To mimic the same conditions of clinical use, cryopreserved human thoracic aortas were thawed as follow: the bags were quickly submerged in a water bath at 39 °C and let thaw for 10 min. When thawing was achieved, the bags were cut

under sterile conditions and the samples were washed in cooled saline solution (NaCl 0.9%) for 3 times to remove the cryoprotectant that may be toxic at temperature above 10 °C. This is the method we routinely use for reconstructive vascular surgery. Pressure tests performed in the operating room did not reveal any leakage from the thawed graft. After thawing, a 2-cm-long ring was removed from the graft and longitudinally cut into a 2 £ 4 cm open square; sampling was performed by using a blade under sterile conditions, as follows: a 0.5 £ 4-cmlong tissue strip was used for LM, immunohistochemistry (IIC), and TUNEL assays, a series of 0.3 £ 0.3 cm cubic samples for TEM (1 sample) and organ culture studies (4 samples), the remaining tissue for isolating cells to be submitted to Xow cytometry. For LM, formalin-Wxed, paraYn-embedded 5 m thick sections were stained with hematoxylin and eosin. IIC was performed by using monoclonal antibodies to endothelial cell CD34 antigen (clone Qbend10, dilution 1:80, Dako, Copenhagen, Denmark) and smooth muscle actin (ASMA, clone 1A4, dilution 1:8000, Sigma–Aldrich, Milano, Italy). The antibodies were used to identify the number of speciWcally labeled cells in unfrozen and cryopreserved allografts. Following washing, sections were incubated with a byotinylated, aYnity puriWed horse anti-mouse secondary antibody (Vector Laboratories, dilution 1:500, Burlingame, CA, USA) for 30 min at rt and then with streptavidin:biotinylated peroxidase complexes (Biospa-Division, dilution 1:250, Milano, Italy) for 30 min at rt and Wnally treated with diaminobenzidine (Sigma–Aldrich). Counterstaining with hematoxylin was performed. Apoptotic cells were end-labeled in situ by TUNEL staining using the cell death detection kit, POD (Roche, Germany) according to the manufacturer’s instructions. TUNEL staining was performed on serial sections cut from the same samples processed for LM. The samples were analyzed quantitatively with a computer-assisted LM by using Image-ProPlus software (ver. 4.5; MediaCybemetics http://www.mediacy.com). For TEM, specimens were Wxed in 2.5% buVeredglutaraldehyde followed by 1% buVered-osmium tetroxide for 1 h at rt. After dehydration, the samples were embedded in epoxy resin; ultrathin sections were stained with uranyl acetate and lead citrate, and examined in a transmission electron microscope Philips 400T.

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

Flow cytometry was used to assess in vivo cell viability and to establish the phenotype of cells isolated from cryopreserved allografts. Cells were obtained by scraping and enzymatic digestion. BrieXy, the samples were treated with type VII collagenase (0.1%, Sigma–Aldrich) and scraped until reaching the adventitial layer. Minced tissue was incubated with the same concentration of collagenase at 37 °C in a 5% CO2 humidiWed atmosphere for 1 h. Then, cell isolates were Wltered, centrifuged in Hepes buVer, pH 7.4, plus 10% FCS (Seromed, Biochrom KG, Berlin, Germany) for 5 min 2£ at 1800 rpm. Pellets were suspended in 2 ml of PBS. Cell viability was assessed as follows: to 5 £ 105 cells in 2 ml volume of PBS, 20 ml 7-aminoactinomycin D (7-AAD) solution (Beckman-Coulter, Miami FL, USA) were added; after 20 min of incubation the cells were analyzed with a FC 500 Xow cytometer (Beckman-Coulter). To analyze CD34 expression, 5 £ 105 cells in a 100 ml volume of PBS were incubated for 20 min with 20 ml of CD34-PE (Beckman-Coulter). ASMA analysis was performed by intracytoplasmic stain. BrieXy, cells were Wxed with reagent 1 of the Intraprep kit (Beckman-Coulter) following manufacturer’s instruction. After two washes with PBS, the samples were permeabilized with reagent 2 and, after two additional washes, cells were Wrst incubated with 5 ml of ASMA (Sigma– Aldrich) and, subsequently, with FITC anti-Mouse IgG (Beckman-Coulter). Control samples were run with an irrelevant monoclonal antibody and FITC anti-mouse IgG. Organ culture was used to establish the ability of cells to recover freezing/thawing injury. The samples were carefully placed in contact with the 100 mm culture plates. RPMI 1640 with L-Glutamine (Cambrex, BioScience, Vierviers, Belgium), 10% FCS (Seromed, Biochrom KG, Berlin, Germany) and 1% Penicillin/Streptomycin 100£ solution (EuroClone, Life Science Division, Milano, Italy) was added. The culture plates were placed in a 37 °C incubator with 5% CO2. After culturing for 4, 8, and 15 days tissues were processed for LM and TEM as described above. Before cryopreservation, human thoracic aortas were normal at LM except for a patchy distribution of luminal endothelium; TEM revealed minimal features of SMC injury including vesicular nuclei and oedematous mitochondria. After thawing, the histological picture was similar to that of unfrozen arteries (Fig. 1A). Focal changes, including nuclear polymorphism and hypercromasia were seen in

311

SMCs; small bar-like areas of tissue clearings were found close to SMCs (Fig. 1A). By TEM SMCs had nuclei with condensed chromatin and dark stained cytoplasms (Fig. 1B); foci of contractile Wlament loss, small collections of lipid droplets and oedematous mitochondria were a common Wnding; the areas of tissue clearings observed at LM were located between the SMC plasma membrane and the innermost portion of the contractile cytoplasm; they appeared as empty cytoplasmic zones with granular and Wlamentous remnants (Fig. 1C). CD34 immunohistochemical staining showed a reduction in the number of luminal endothelial cells after cryopreservation; ASMA staining of cryopreserved aortas was similar to that of unfrozen arteries. The percentage of TUNEL stained SMCs ranged from 0.3 to 6.2% (average of 4.2% labeled SMCs) in unfrozen human thoracic aortas (Fig. 2A). After cryopreservation the percentage increased from 2.8 to 16.5% with an average of 8.6% (Fig. 2B). Flow cytometry 7-AAD staining of cryopreserved homograft isolates documented that an average of 49 § 16% cells were unlabeled. As shown in Table 1, in three cases the percentage of unlabeled cells was less than 50%, while two samples showed more satisfactory results, i.e., 56 and 75%. Flow cytometry immunophenotyping showed that SMCs were the major cell component present in the primary isolates (an average of 76 § 12% cells were ASMA positive); the CD34 positive endothelial cells were a minor fraction only (percentages of CD34+ positive cells ranged from 3.9 to 0.01%). Since ASMA staining requires cell permeabilization, we were not able to determine the absolute 7-AAD unstained fraction of ASMA-positive cells. LM and TEM did not reveal signiWcant diVerences between unfrozen and cryopreserved samples in organ cultures taken at 4 days. However, a progressive reduction in the number of SMCs from day 4 to day 15 associated with degenerative ultrastructural changes were found in cryopreserved samples when compared to controls (Figs. 3 and 4). The results of this study indicating that human arterial SMCs are injured following cryopreservation procedures contrasts with those from previous functional pharmacological studies on cryopreserved human internal mammary arteries [10]. This discrepancy may be related to the technical conditions applied in this study or tissue-speciWc responses to cryopreservation. Human thoracic aortas were harvested in cold Celsior media which belongs to a class of solutions

312

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

Fig. 1. Cryopreserved human thoracic aorta (HTA). (A) LM shows a normally looking media with focal perinuclear clearings (arrows); (B) a smooth muscle cell (SMC) with homogeneous chromatin clumping (single arrow), and dense cytoplasm. TEM, 3600£; (C) subplasmalemmal structureless area (M) of SMC cytoplasm corresponding to the clearings seen at LM. TEM, 6900£.

designed to physically restrict temperature-induced ionic imbalances in tissues [13]. Accordingly, LM and TEM of controls did not show signiWcant changes in the arterial wall whereas TUNEL assay revealed an average of 4.2% positive SMCs. As to disinfection we avoided the use of the potentially cytotoxic amphotericin B; the cryoprotectant solution, containing 10% Me2SO, was cooled before its use to minimize Me2SO

injury; it should be noted that no rinse was done before exposing the homografts to Me2SO; this could at least explain in part the decreased SMC preservation we found in the present study; in fact it was recently found that placing porcine heart valves directly in 10% Me2SO after 24 h (22 °C) exposure to antibiotics reduced valve-leaXet Wbroblast viability after freezing and thawing [4].

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

313

Fig. 2. TUNEL staining. An unfrozen HTA with unreactive SMCs (A) and a cryopreserved HTA showing many positive SMC nuclei (B). LM 400£.

As to the freezing method, we used a computerassisted controlled-rate freezer to guarantee a cooling rate of 1 °C/min to ¡45 °C and a faster rate thereafter until ¡120 °C was achieved; in our experience, this freezing method succeeds in providing adequate arterial substitutes for speciWc clinical settings [7]. Cryopreserved aortic allografts were thawed as currently done in the clinical use. After thawing, TEM revealed foci of cytoplasm disruption in SMCs

which were located between the plasma membrane and the innermost portion of the contractile cell cytoplasm. These Wndings can be a consequence of osmotic imbalance occurring when the rate of thawing in not well controlled; even if the rate of tissue rewarming is believed to be less critical than the controlled-rate of cooling, it seems likely that a slow warming allows better cell rehydration and gradual loss of intracellular accumulated solutes therefore minimizing osmotic injury; Wndings similar to ours,

314

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

Table 1 The table schematizes the mean percentages of 7-AAD unlabeled (line 1), ASMA positive (line 2), and CD34 positive (line 3) cells found in primary isolates recovered from cryopreserved human thoracic aortas (n. 5 samples harvested from heart beating donors, Wrst letters of each individual donor’ name [i.e., ZI, LU etc.] are indicated) DONOR

ZI (%) LU (%) DB (%) RU (%) NU (%)

7-AAD unlabeled cells 56 ASMA+ cells 66 3.9 CD34+ cells

37 68 2.28

75 93 0.01

37 84 1.5

43 71 2.1

Flow cytometry assay. 10,000 events per analyses. The percentage is a mean of a triplicate examination /each donor. 7-AAD dye unlabeles structurally viable cells; ASMA monoclonal antibody stains smooth muscle cells and CD34 monoclonal antibody stains endothelial cells.

i.e., empty pockets of oedema in close proximity to elastic tissue, were found by Buján et al. [5] in mini pig iliac arteries; in their study the pockets disap-

peared following gradual thawing of cryopreserved iliac arteries. Post-thawing treatments are also critical and may contribute to osmotic injury; in our study, the thawed aortic allografts were washed in cooled saline solution to remove the Me2SO; diluting the cryoprotectant incrementally in the presence of a non-permeating solute and protein support is expected to ameliorate this step. Although it is conceivable that the thawing method we used had contributed to damage SMCs, it is interesting to note that neither the rate of warming nor the rate of Me2SO dilution were shown to have any inXuence on the outgrowth of porcine valve-leaXet Wbroblasts in a recent experimental study [2]. SMCs are thought to be suYciently well resistant to cryo-injury with the possibility to fully recover their functionality after grafting [17]. Accordingly Müller-Schweinitzer et al. [10] found good recovery of function in human internal mammary arteries

Fig. 3. Organ culture of an unfrozen HTA. At LM no signiWcant decrease in the number of SMCs is seen during culture conditions (A, 4 days; B, 15 days); after 15 days SMCs shows peripheral chromatin clumping (arrow) (C, TEM, 3600£).

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

315

Fig. 4. Organ culture of a cryopreserved HTA. A reduction in the number of SMCs during culture conditions is seen by LM (A, 4 days; B, 15 days); TEM shows progressive loss of SMC structural integrity (C, 4 days, TEM, 3600£ and D, 15 days, TEM, 3600£).

subjected to diVerent freezing/thawing protocols. However, SMC contractibility was found reduced in human aorta [14,6], mesenteric, coronary [9], and femoral [11,15] arteries by using similar freezing protocols, thus suggesting that responses to cryopreservation could at least be in part tissue-speciWc.

Herein, we demonstrate that following the conditions applied in this study aortic SMCs are injured; after thawing chromatin condenzation and clumping was found by TEM; an increased nuclear staining was found by TUNEL assay which reveals “in situ” DNA strand ruptures; cytoplasmic changes

316

G. Pasquinelli et al. / Cryobiology 52 (2006) 309–316

including the presence of mitochondrial oedema, vacuoles, and lipid droplets were seen ultrastructurally. These results were extended by 7-AAD Xow cytometry staining of cell isolates. 7-AAD, a nuclear dye that is excluded by viable cells but penetrate cell membranes of dying or dead cell, is generally used to identify necrotic and late apoptotic cells; in the present study, we found that an average of 49 § 16% SD isolated cells from cryopreserved arteries were unlabeled thus suggesting that a reduced cell fraction was still alive at the time of analysis; immunophenotyping showed that 76 § 12% SD of cell isolates were ASMA-positive; this imply that SMCs are proportionally the most injured cells in the cryopreserved arterial wall. Bellón et al. [3] demonstrated that damage to cryopreserved rat iliac arteries cannot be reversed by organ culture; accordingly, we found that, unlike controls, the number of SMCs of cryopreserved aortas decreased from day 4 to day 15 of culture while appearing extensively damaged by ultrastructural examination. An additional point favoring the view that SMCs from cryopreserved arteries are unable to recover from damage was that, unlike fresh arteries, we did not succeed in establishing primary cell cultures from cryopreserved arteries (data not shown). In conclusion, we found that the cryopreservation protocol we use for clinical applications damages aortic SMCs; this provides a structural explanation to the reduced arterial contractibility found by using similar protocols and supports the fact that human cryopreserved homografts are converted to Wbrous acellular conduits after implantation [16]. References [1] M. Adam, J.F. Hu, P. Lange, L. WolWnbarger Jr., The eVect of liquid nitrogen submersion on cryopreserved human heart valves, Cryobiology 27 (1990) 605–614. [2] W.J. Armitage, W. Dale, E.A. Alexander, Protocols for thawing and cryoprotectant dilution of heart valves, Cryobiology 50 (2005) 17–20. [3] J.M. Bellón, M.J. Gimeno, G. Pascual, N. Garcia-Honduvilla, B. Dominguez, J. Buján, Arterial damage induced by cryopreservation is irreversible following organ culture, Eur. J. Vasc. Endovasc. Surg. 17 (1999) 136–143.

[4] V. Birtsas, W.J. Armitage, Heart valve cryopreservation: protocol for addition of dimethyl sulphoxide and amelioration of putative amphotericin B toxicity, Cryobiology 50 (2005) 139–143. [5] J. Buján, G. Pascual, R. López, C. Corrales, M. Rodríguez, F. Turégano, J.M. Bellón, Gradual thawing improves the preservation of cryopreserved arteries, Cryobiology 42 (2001) 256–265. [6] S.E. Langerak, M. Groenink, E.E. van der Wall, C. Wassenaar, E. Vanbavel, M.C. van Baal, J.A.E. Spaan, Impact of current cryopreservation procedures on mechanical and functional properties of human aortic homografts, Transplant. Int. 14 (2001) 248–255. [7] M. Mirelli, M. Buzzi, G. Pasquinelli, P.L. Tazzari, G. Testi, E. Ricchi, R. Conte, A. Stella, Fresh and cryopreserved arterial homografts: immunological and clinical results, Transplant. Proc. 37 (2005) 2688–2691. [8] G.J. Morris, The origin, ultrastructure, and microbiology of the sediment accumulating in liquid nitrogen storage vessels, Cryobiology 50 (2005) 231–238. [9] E. Müller-Schweinitzer, M.J. Mihatsch, M. Schilling, W. Haefeli, Functional recovery of human mesenteric and coronary arteries after cryopreservation at ¡196 °C medium, J. Vasc. Surg. 25 (1997) 743–750. [10] E. Müller-Schweinitzer, P. Stulz, H. StriVeler, W.E. Haefeli, Functional activity and transmembrane signalling mechanisms after cryopreservation of human internal mammary arteries, J. Vasc. Surg. 27 (1998) 528–537. [11] F. Stanke, D. Riebel, S. Carmine, J.L. Cracowski, F. Caron, J.L. Magne, et al., Functional assessment of human femoral arteries after cryopreservation, J. Vasc. Surg. 28 (1998) 273–283. [12] M.G. Strickett, B.G. Barratt-Boyes, D. McCulloch, Disinfection of human heart valve allografts with antibiotics in low concentration, Pathology 15 (1983) 457–462. [13] M.J. Taylor, A.M. Elrifai, J.E. Bailes, Hypothermia in relation to the acceptable limits of ischemia for bloodless surgery, in: P.K. Steponkus (Ed.), Advances in Low Temperature Biology, vol. 3, JAI Press, London, UK, 1996, pp. 1–64. [14] M.E.R. Vázquez, M. Rodriguez Carbarcos, M.V. Mart´nez Santos, R.O. Fernández Mallo, et al., Functional assessment of cryopreserved human aortas for pharmaceutical research, Cell and tissue banking 5 (2004) 119–123. [15] M.E.R. Vázquez, M. Rodriguez Carbarcos, R.O. Fernández Mallo, et al., Functional assessment of human femoral arteries after cryopreservation, Cryobiology 49 (2004) 83–89. [16] P.R. Vogt, T. Stallmach, U. Niederhäuser, J. Schneider, G. Zünd, M. Lachat, A. Künzli, M.I. Turina, Explanted cryopreserved allografts: a morphological and immunohistochemical comparison between arterial allografts and allografts heart valves from infants and adults, Eur. J. Cardio-Thoracic Sur. 15 (1999) 639–645. [17] M.C. Wusteman, D.E. Pegg, R.M. Warwick, The banking of arterial allografts in the United Kingdom. A technical and clinical review, Cell and tissue banking 1 (2000) 295–301.