Cyanoacrylate adhesive has been suggested as an alternative to suturing when repairing severed peripheral nerves. The authors examined the cytotoxic effect of ethyl-cyanoacrylate on the human neuroblastoma cell line SH-SY5Y and compared it with the effects of butyl-cyanoacrylate (Histoacryl ® ), an adhesive approved for skin closure. Ethyl-cyanoacrylate or butyl-cyanoacrylate was applied in confluent SH-SY5Y cultures. Immediately, at 24 h and at 7, 14, 21 and 28 days, cultures were photographed and analysed digitally. At corresponding intervals, cell death was quantified using a 51 Cr release assay. In cultures exposed to ethyl-cyanoacrylate or butyl-cyanoacrylate, cell death was observed predominantly in conjunction with the adhesive, causing a halo devoid of cells. Surviving cells showed neurodegenerative properties with loss of neuritis and reduction of body size up to 3 days post exposure. The inhibition halo diminished over time in both groups and at 28 days cells reached the margin of the adhesive in the ethyl-cyanoacrylate group. 51 Cr assay indicated significant cell death in exposed cultures, which rapidly decreased during the first 14 days. No significant differences were found between the adhesives. This study demonstrates that ethyl-cyanoacrylate and butyl-cyanoacrylate have a transient cytotoxic effect, which may explain the promising results when using cyanoacrylate for nerve repair.
Microsurgical suturing is the standard technique for repairing transected peripheral nerves. The technique requires surgical excellence and may be difficult to apply when surgical access is limited. Microsurgical suturing does not result in perfect coaptation of the internal fascicular structures, which hampers the growth of regenerating axons and, ultimately, complete nerve function recovery . There is a need for complementary surgical techniques that provide rapid and reliable primary repair of transected nerves.
Different strategies have been tried to readapt the nerve endings using tissue adhesive. Fibrin sealants are the most successful surgical adhesives and sealants, but they have limitations in their mechanical and biological properties. In peripheral nerve repair, the primary disadvantage is the low cohesive strength, which makes the anastomosis insecure during the healing phase . To avoid gaps at the repair site, a nerve anastomosis made with standard fibrin adhesive is often complemented, in practice, with microsutures. Other shortcomings preventing the acceptance of fibrin glue as a surgical tissue adhesive are: short-term persistence (<2 weeks) in vivo ; risk of infectious blood-borne disease transmission; and low viscosity prior to polymerization with thrombin, which makes the adhesive difficult to apply .
Cyanoacrylates (CAs) are synthetic adhesives that polymerize rapidly on contact with water or blood. The adhesives are inexpensive, relatively easy to apply, do not carry any risk of viral transmission and have sufficient strength to maintain a nerve anastomosis, even under tension . The cytotoxicity of CA adhesives in surgical applications is debated. When used as a tissue adhesive, CA has been shown to induce a stronger tissue reaction than non-resorbable sutures, resulting in a more pronounced foreign-body inflammatory reaction, while others have found it causes tissue necrosis in vivo . In some reports the inflammatory reaction around the repair site has been shown to be harmful and to cause a focal hindrance to the recreation of tissue . In other studies, the inflammatory reaction has been shown to require active communication between the injured tissue and the recruited macrophages needed for degradation and tissue regrowth and, consequently, to be an integral part of the healing process .
The advantages of CA as a tissue adhesive are described by authors in a variety of medical fields. In microsurgery, successful use of CA for microneural and microvascular anastomoses has been reported . The authors have compared ethyl-cyanoacrylate (ECA) with microsurgical suturing for microneural anastomosis in two different studies. The two techniques gave rise to equivalent recovery of motor and sensory conduction velocities as well as motor nerve action potentials when measured 6 months after surgery . 7 days after surgery, the adhesive stimulated an increased appearance of macrophages around the repair site, which could explain the accelerated Wallerian degeneration distal to the seam after nerve repair with ECA . The results are promising and have prompted the authors to investigate the toxicity of ECA further. They have used SH-SY5Y, a widely employed neuronal cell line for studying neuronal toxicity and neurocytoprotection . For comparison, the cytotoxicity of Histoacryl ® (butyl-2-cyanoacrylate) was studied, this is considered to be one of the least cytotoxic cyanoacrylate derivatives and it is one of the most frequently used CAs in general surgery.
Material and methods
SH-SY5Y culture and incubation conditions
SH-SY5Y cells, a human neuroblastoma cell line, obtained from ATCC (American Type Culture Collection, Manassas, VA, USA) were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% foetal bovine serum and 1% 200 mM l -glutamine (National Veterinary Institute of Sweden, Uppsala, Sweden) and seeded in T 75 cm 2 cell culture flasks (Nunc, Roskilde, Denmark). The culture medium was changed the following day and thereafter twice a week. When confluent, cultures were trypsinized (Trypsin 0.5% EDTA, Invitrogen Life Technologies, Paisley, UK), seeded in 24-well cell culture dishes and maintained in a 5% CO 2 /95% humidified air atmosphere at 37 °C. Plated cells were allowed to grow overnight prior to the treatment. Confluent cultures were used for experiments.
Cyanoacrylate exposure in culture
Cell cultures were exposed to the adhesive by adding 0.1 μl of ethyl- or butyl-cyanoacrylate to the centre of each well after the medium had been removed. The ECA used in this study is commercially available (Evobond ® , Tong Shen Enterprise Co., Ltd., Taiwan), whereas the butyl-cyanoacrylate (BCA) is a tissue adhesive produced for sutureless skin closure (Histoacryl ® , B. Braun Surgical GmbH, D-34209 Melsungen, Germany). The adhesive was set to polymerize on the surface of the confluent cell culture for 5 s, whereupon 500 μl of fresh medium was added to the well without letting the cell culture run dry. The culture medium was changed twice a week during the whole observation period. Eight wells in each 24-well dish were exposed to either ECA or BCA, whereas eight wells were unexposed to adhesive and used as controls.
Evaluation of the halo devoid of cells
After application, 24 h were allowed for full polymerization of each droplet of adhesive to occur. Cell cultures were followed up to 28 days ( n = 8 wells in each group for each time point). At 24 h and 7, 14, 21 and 28 days after incubation, cultures were examined by phase contrast microscopy (Nikon Diaphot 300, Nikon Corp., Tokyo, Japan). A calibration grid was photographed at the corresponding magnification.
The microscopic images were captured with a digital camera (Nikon CoolPix 990, Nikon, Tokyo, Japan). All micrographs were displayed on a computer monitor and analysed using morphometric software (ImageJ). The outer edge of the cell-free halo around the adhesive (see “Results”) was approximated to a circle and the distance from the adhesive border to the cell layer margin was measured. For each well, measurements were done in the four quadrants perpendicular to each other and mean values were calculated. In order to confirm the results obtained from cultures followed throughout the study period, complementary measurements were carried out in cultures intended for another analysis (i.e. adhesive-induced cell death measured using a 51 Cr assay; see below). These measurements were carried out at 24 h and 7 and 28 days.
Immunofluorescence staining and analysis of cell morphology
A cytoskeletal assessment was made on samples exposed to ECA or BCA as well as time-matching controls after 24 h and 3 and 7 days. The staining method for tubulin has been described in detail elsewhere . Briefly, cells were fixed with methanol for 5 min at −20 °C and permeabilized with acetone at −20 °C for 10 s. Cells were incubated with primary antibodies to tubulin (BioGenex Laboratories, Mainz, Germany) (dilution 1:20) and subsequently with FITC-conjugated secondary antibodies (Jackson Immuno Research Laboratories, West Grove, PA, USA) (dilution 1:20). The specimens ( n = 8 wells in each group for each time point) were examined by fluorescence microscopy (Nikon Eclipse E600, Nikon, Tokyo, Japan) and photographed with a digital camera (Nikon Digital Sight DS-U1). Images were captured from each culture in four different quadrants perpendicular to each other (total of four images per cell culture) from the area immediately outside the cell-free halo (see “Results”). Five separately located cells (displayed on a computer monitor) were selected randomly from each micrograph and cell sizes were analysed using morphometric software (ImageJ ® 1.33u, National Institutes of Health, USA) at 24 h and 7 days post exposure. Cell bodies and neurites were analysed descriptively with respect to shape and texture at all time points.
Detection of cell death with 51 Cr
Cytotoxicity designating necrotic cell death was assessed by the release of 51 Cr as described in detail elsewhere . Briefly, SH-SY5Y cells were grown to confluence in five 24-well dishes. Eight cell cultures in each 24-well dish were exposed to either ECA or BCA while eight wells were left unexposed to adhesive and used as controls. At 7, 14, 21 or 28 days the cultured cells (adhesive exposed), as well as time-matching controls, were labelled with 51 Cr (Perkin Elmer, Boston, MA, USA) for 24 h. Exceptions were cells analysed 24 h after exposure to adhesive, which were labelled prior to exposure. After labelling, cultures were washed and left in the cell incubator for 24 h. Supernatants were carefully aspirated and the remaining SH-SY5Y in each well was lysed with 500 μl 1 M NH 4 OH. The radioactivity of the supernatants and the remaining SH-SY5Y was measured in a gamma counter (Packard Cobra II from CIAB, Stockholm, Sweden). Cell death, expressed as percent of 51 Cr release, was calculated as follows: