|Home | About | Journals | Submit | Contact Us | Français|
The amphipathic molecule dimethyl sulphoxide (DMSO) is a solvent often used to dissolve compounds applied to the inner ear; however, little is known about its potential cytotoxic side effects. To address this question, we applied 0.1 to 6% DMSO for 24 h to cochlear organotypic cultures from postnatal day 3 rats and examined its cytotoxic effects. DMSO concentrations of 0.1% and 0.25% caused little or no damage. However, concentrations between 0.5 and 6% resulted in stereocilia damage, hair cell swelling and a dose-dependent loss of hair cells. Hair cell damage began in the basal turn of the cochlea and spread towards the apex with increasing concentration. Surprisingly, DMSO-induced damage was greater for inner hair cells than outer hair cell whereas nearby supporting cells were largely unaffected. Most hair cell death was associated with nuclear shrinkage and fragmentation, morphological features consistent with apoptosis. DMSO treatment induced TUNEL positive staining in many hair cells and activated both initiator caspase-9 and caspase-8 and executioner caspase-3; this suggests that apoptosis is initiated by both intrinsic mitochondrial and extrinsic membrane cell death signaling pathways.
Dimethyl sulphoxide (DMSO), an amphipathic solvent soluble in both water and organic substances, is often used to dissolve hydrophobic substances used in biological research. DMSO has a number of important properties that make it useful clinically. DMSO has been used successfully to treat urinary (McCammon et al., 1998), pulmonary (Iwasaki et al., 1994), rheumatic (Morassi et al., 1989) and dermatological disorders (Burgess et al., 1998). Because of its anti-inflammatory and antioxidant properties, DMSO has also been used with some success to treat gastrointestinal disorders (Salim, 1992a; Salim, 1992b). Since DMSO can act as a free radical scavenger and can cross the blood–brain barrier, it has also been used to treat brain edema (Broadwell et al., 1982; Ikeda et al., 1990). Although clinically beneficial in some situations, DMSO can have systemic side effects such as diarrhea, vomiting, bronchospasm, hypertension, and pulmonary edema (Davis et al., 1990; Hameroff et al., 1983; Smith et al., 1987; Stroncek et al., 1991). These effects appeared to be dose-dependent (Davis et al., 1990; Stroncek et al., 1991).
The effects of DMSO on cellular function have been studied in a large number of cell types, but with variable results. Some studies have shown that DMSO blocks the rise of intracellular calcium induced by different agents (Reynaud et al., 1999; Saldanha et al., 2002; Santos et al., 2002; Zhang et al., 1999). Others have reported that DMSO causes an increase in extracellular sodium, potassium and calcium (Santos et al., 2002). DMSO has diverse effects on ion transporters and pumps (Santos et al., 2002). In mouse lymphoma cells, 18 h treatment with 2.5% DMSO induced an apoptotic response consisting of a decline in Bcl-2, a decrease in the mitochondrial membrane potential, the release of cytochrome c from mitochondria, activation of initiator caspase-9 and executioner caspase-3 (Liu et al., 2001). Studies with lymphoma cells suggest that DMSO concentrations of 1−2% prevent apoptosis whereas higher concentrations induces apoptosis (Lin et al., 1995). DMSO enhances the suppressive effects on lidocaine on synaptic transmission (Somei et al., 1995). DMSO (10%) is often used as a cryoprotectant to optimize cell survival in vitro and in vivo (Davis et al., 1990; Trumble et al., 1992). However, when used in vivo DMSO did not protect peripheral nerves from cryoinjury and exacerbated functional recovery (Trumble et al., 1992).
Several studies have used DMSO to dissolve otoprotective compounds that are given systemically or applied locally to the middle ear or round window membrane of the cochlea. NMDA receptor antagonists, in the form of maleate and tartrate salts (metal chelators) dissolved in DMSO and administered systemically protected against aminoglycoside ototoxicity (Basile et al., 1996). However, when sodium maleate or tartaric acid dissolved in DMSO (40%) were given without the NMDA receptor antagonists, they also provided significant protection against aminoglycoside damage (Sha et al., 1998). These findings suggested that the metal-chelating and antioxidant properties of these agents acting alone or in combination were protecting against aminoglycoside ototoxicity. Another study infused Texas Red and DMSO into the middle ear space as a control in order to study the distribution of the tracer within the inner ear (Liu et al., 2006). The abstract did not mention the concentration and volume of DMSO applied to the round window or if DMSO caused any cochlear damage. Src protein tyrosine kinase inhibitors dissolved in DMSO (≤ 0.5%) have been placed on to the round window to determine if they would protect against noise induced hearing loss (Harris et al., 2005). No negative side effects from the solvent were reported.
Others have dissolved compounds in DMSO in order to facilitate the entry of these compounds into isolated hair cells (Canlon et al., 1993; Szonyi et al., 1999; Szonyi et al., 2001). One percent DMSO applied to acutely isolated outer hair cells (OHC) resulted in a significant decrease in OHC electromotility (Canlon et al., 1993); the reduction in electromotility was reversed by increasing the concentration of ionomycin, a calcium ionophore (Szonyi et al., 2001). These results indicate that low concentrations of DMSO can exert an immediate effect on OHC function.
Although DMSO is often used to dissolve compounds that are applied to the inner ear in vivo and in vivo, the physiologic and anatomical effects of DMSO on the cochlea are poorly understood. Since DMSO could conceivably cause negative side effects, we applied different concentration of DMSO to cochlear organotypic cultures to evaluate its cytotoxic effects on cochlear organotypic cultures. In contrast to in vitro studies, the cochlear culture preparation offers precise control of the amount of DMSO that can be applied to the inner ear.
Our procedures for preparing cochlear organotypic cultures have been described in detail in earlier publications (Corbacella et al., 2004; Ding et al., 2002; Ding et al., 2007; Lanzoni et al., 2005). Rat pups (CDF (F344), Charles River Laboratory Inc.) were decapitated on postnatal day 2−3. The cochlea was carefully removed, the lateral wall dissected away and the whole basilar membrane containing the organ of Corti and spiral ganglion neurons removed as a flat surface preparation. A drop (15 μl) of cool, rat tail collagen (Type 1, Collaborative Biomedical Products #40236, 3.76 mg/ml in 0.02 N acetic acid, 10×basal medium eagle (BME, Sigma B9638), 2% sodium carbonate, 9:1:1 ratio) was placed in the center of a 35 mm diameter culture dish (Falcon 1008, Becton Dickinson) and allowed to gel at room temperature. Then 1.2 ml of serum-free medium consisting of 2 g bovine serum albumin (BSA, Sigma A-4919), 2 ml Serum-Free Supplement (Sigma I-1884), 4.8 ml of 20% glucose (Sigma G-2020), 0.4 ml penicillin G (Sigma P-3414), 2 ml of 200 mM glutamine (Sigma G-6392), 190.8 ml of 1× BME (Sigma B-1522) were added to the dish. The cochlear tissues were placed on the surface of the collagen gel. The cochlear explants were placed in an incubator (Forma Scientific, #3029) and maintained at 37° C in 5% CO2 overnight. On day 2, the culture medium was removed, replaced with fresh culture medium without or with various concentrations of DMSO and cultured for 24 h.
For the dose-response studies, 60 cochleae were randomly divided into 10 groups (n = 6 per group). One group was used as normal control; the remaining eight groups were treated with DMSO (Sigma D-8418). DMSO stock solution was freshly made at a concentration of 20% (v/v) in serum-free medium. DMSO stock solution was diluted with serum-free medium and applied to the cochlear cultures at a final concentration of 0.1, 0.25, 0.5, 0.75, 1, 2, 3, 5 or 6%. Control and DMSO-treated cochlear explants were placed in the incubator for 24 h and then harvested for histological analysis.
Cochlear explants were fixed in 4% formalin in phosphate buffered saline (PBS) for 2 h. After being washed with 0.1 M PBS, specimens were stained with Alexa 488-conjugated phalloidin (Molecular Probes A12379; 1:200 in PBS, 30 minutes). Some cochlear cultures were mounted in glycerin on glass slides and observed with a confocal laser, scanning microscope (Zeiss LSM510, excitation 488, emission 518 nm). Images were processed with Advanced Imaging Microscopy (version 4.0, Carl Zeiss) and Adobe Photoshop (version 5.5) as described previously (Ding et al., 2002; Ding et al., 2007). For cochlear hair cell counting, cochlear cultures were mounted in glycerin on glass slides and observed under a fluorescent microscope (Zeiss Axioskop) equipped with appropriate filters (absorption 494 nm, emission 518 nm). Hair cells were evaluated along the entire length of the cochlea from apex to base. The stereocilia bundles, which are heavily labeled with Alexa 488-phalloidin, were clearly visible allowing for easy identification of hair cells that were present or missing (Corbacella et al., 2004). Hair cells were counted as missing if the stereocilia were missing and the cuticular plate was damage. A cochleogram showing the percent inner hair cell (IHC) and outer hair cell (OHC) loss as a function of percent distance from the apex was constructed. A mean cochleogram showing the percent IHC and OHC loss as a function of percent total distance from the apex was constructed for each group.
Six specimens were stained with To-Pro-3 (Invitrogen, Molecular Probes, T3605) to assess the morphological changes to the nucleus induced by DMSO. Cochlear cultures treated with 5% DMSO were fixed in 4% formalin in PBS for 30 minutes. Specimens were stained with fresh To-Pro-3 solution (1 mM To-Pro-3 in 0.75 μl DMSO dissolved in 1 ml H2O) for 30 minutes and rinsed with 0.1 mM PBS for 15 minutes. Tissues were then stained with Alexa 488-phalloidin for 30 minutes. Specimens were examined under a confocal laser, scanning microscope (Zeiss LSM510, To-Pro-3, excitation 642 nm, emission 661 nm). Images were processed with Advanced Imaging Microscopy (version 4.0, Carl Zeiss) and Adobe Photoshop as described previously (Ding et al., 2002; Ding et al., 2007).
Six cochlear cultures treated with 5% DMSO for 24 h were evaluated for TUNEL (terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling) labeling using the APO-BrdU TUNEL Assay Kit (A-23210, Molecular Probes, Inc, Eugene, OR) according to the manufacturer's protocol. Specimens were fixed with 10% formalin in PBS. Afterwards, the tissue were transferred to ethanol overnight (4° C), rinsed in washing buffer at room temperature and incubated with 50 μl of DNA-labeling solution overnight at room temperature. Specimens were washed twice in rinsing buffer and then incubated in 100 μl of freshly prepared antibody labeling solution for 1 h at room temperature. Afterwards, the samples were labeled with To-Pro-3 as described above.
Some cochlear explants were treated with 5% DMSO for 24 h, and then incubated for 1 h at 37° C with cell permeable, carboxyfluorescein caspases (Cell Technology Inc.) which preferentially labels activated caspase-8 (n=6), caspase-9 (n=6) or caspase-3 (n=6) in living cells. Afterwards, samples were fixed with 10% formalin in PBS for 20 minutes and then labeled with To-Pro-3 as described above. Cochlear cultures were mounted in glycerin on glass slides and examined under confocal microscope using appropriate filters to detect the carboxyfluorescein activated caspase (excitation 488 nm, emission 529 nm) and To-Pro-3 (excitation 642 nm, emission 661 nm) which labels the nuclei. Images were processed with Advanced Imaging Microscopy (version 4.0, Carl Zeiss) and Adobe Photoshop as described previously (Ding et al., 2002). These experiments were approved by the University at Buffalo's Animal Care and Use Committee.
Figure 1A shows a normal control cochlea maintained in culture for 24 h and then stained with Alexa 488-phalloidin to label the actin that is heavily expressed in the stereocilia and cuticular plate of hair cells. Three orderly rows of OHC and one row of IHC were present in normal control cochlear cultures (Figure 1A). Cochlear cultures treated with 0.1% DMSO were indistinguishable from normal control cultures (data not shown). However, when the concentration of DMSO was increased to 0.5%, mild stereocilia bundle damage was seen on a few IHC (Figure 1B). Increasing the concentration of DMSO to 2% resulted in stereocilia clumping and disarray that was more prominent on IHC than OHC (Figure 1C). Increasing the concentration to 3% resulted in mild to moderate swelling of hair cell bodies and disorderly rows of IHC and OHC (Figure 1D). In addition, the stereocilia bundle was partially or completely missing on some hair cells while on others the stereocilia bundle was fused and/or elongated. The two highest concentrations (5−6%) resulted in severe swelling of the hair cell bodies, disorganized hair cell rows and loss of stereocilia on the majority of hair cells (Figure 1 E-F). The most severely swollen hair cells consisted of a circumferential ring of actin and a dark central core devoid of actin (jagged arrows). The most severe swelling occurred with 6% DMSO (Figure 1F); some cells were extremely large compared to neighboring cells resulting in highly irregular hair cell rows.
To evaluate the location and degree of cochlear pathology as a function of DMSO concentration, mean cochleograms (N=6) were constructed showing the percent loss of OHC and IHC as a function of percent distance from the apex of the cochlea for each DMSO concentration. There was little or no IHC or OHC loss in cochlear cultures treated with 0.1% DMSO (Figure 2 A) or 0.25% DMSO (Figure 2B). Increasing the concentration to 0.5% DMSO resulted in a small loss of IHC near the base of the cochlea; however, the OHC were intact (Figure 2C). Although most IHC and OHC were present, a few showed signs of stereocilia damage (Figure 1B). When the concentration of DMSO was increased to 0.75%, there was widespread loss of IHC (Figure 2D). The mean IHC loss near the base of the cochlea was approximately 55% and declined to around 10% near the apex of the cochlea. The OHC loss, by contrast, declined from 45% in the extreme base to less than 10% in the apical three-fourths of the cochlea. With a DMSO concentration of 1%, the hair cell lesion increased in magnitude and spread further toward the apex (Figure 2E). The IHC loss ranged from approximately 80% in the extreme base and declined to around 20% near the apex while the OHC loss declined from 80% in the extreme base to around 5% in the apical third of the cochlea. The 5% concentration of DMSO caused nearly 100% IHC loss in the basal half of the cochlea and the loss declined to around 40% in the extreme apex (Figure 2F). This concentration resulted in nearly 100% missing OHC in the basal third of the cochlea; the loss rapidly declined to around 35% in the apical half of the cochlea.
Figure 2G shows mean (n=6/group, +1 SD) IHC and OHC loss over the length of the cochlea for the groups treated with 0.1, 0.25, 0.5, 0.75, 1 and 5% DMSO. The 0.1% group showed essentially no hair cell loss and served as the nominal control group. The DMSO concentrations of 0.25% (Figure 2G, solid arrow) caused a slight, but statistically significant increase in IHC loss relative to the control group (Kruskal-Wallis one-way ANOVA on ranks, p<0.001, 5 df, Student-Newman-Keuls pairwise comparisons (p<0.05). IHC loss increased rapidly once the concentration exceeded 0.5%. DMSO concentrations of 0.75% or higher (Figure 2F, dashed arrow) caused a statically significant increase in OHC loss relative to the 0.1% control group (Kruskal-Wallis one-way ANOVA on ranks, p<0.001, 5 df, Student-Newman-Keuls pairwise comparisons (p<0.05).
High concentrations of DMSO resulted in significant swelling and distortion of the hair cell soma (Figure 1 E-F), morphological features characteristic of oncosis (Jaeschke et al., 2003; Majno et al., 1995); however, it was unclear if the plasma membrane ruptured or if hair cell death (Figure 2) was mediated by apoptosis. To address this question, we treated cochlear cultures for 24 h with 5% DMSO, a concentration that causes significant hair cell swelling and cell death, and evaluated the status of the hair cells in the base of the cochlea (Fig. 3A-F, 2.5−3.5 mm from apex) for evidence of apoptosis. The TUNEL assay and To-Pro-3 nuclear staining were used to search for evidence of DNA fragmentation. Figure 3A shows a normal cochlear culture stained with To-Pro-3. The nuclei in the three row of OHC and one row of IHC are aligned in orderly rows; the nuclei of the Hensen cells can be seen alongside the third row of OHC. The nuclei in the hair cells and supporting cells are large, round, evenly stained and of similar size. Figure 3B shows a cochlear culture treated with 5% DMSO and then evaluated with the TUNEL assay and To-Pro-3. Many TUNEL-positive cells were seen in the hair cell region, but little or no labeling was seen in the supporting cells. In addition, many of the To-Pro-3 labeled nuclei in the hair cell region were condensed, a hallmark of apoptosis. In some cases, the TUNEL labeling overlapped (yellow-green) the condensed To-Pro-3 labeled nuclei. Figure 3C shows a cochlear culture treated with 5% DMSO and labeled with Alexa 488-phalloidin and To-Pro-3. Condensed To-Pro-3-positive nuclei can be seen beneath the green fluorescently labeled cuticular plate of the hair cells. Interestingly, most of the To-Pro-3 labeled nuclei in the adjacent Hensen cells appeared normal. Figure 3D shows a cochlear culture treated with 5% DMSO and labeled with caspase-9 and To-Pro-3. Many caspase-9 positive cells were present in the hair cell region; in some cases, a cloud of caspase-9 labeling surrounded a condensed To-Pro-3-labeled nucleus. Importantly, many of the supporting cell nuclei in the inner sulcus region were large and round and appeared normal. Figure 3E shows a cochlear culture treated with 5% DMSO and labeled with caspase-8 plus To-Pro-3. Many caspase-8 positive cells were present in the hair cell region. The specks of caspase-8 labeling frequently surrounded condensed To-Pro-3-labeled nuclei. Despite the fact that many of the nuclei in the hair cell region were condensed, the nuclei of nearby Hensen cells were large, round, normal looking. Figure 3F shows a cochlear culture treated with 5% DMSO and labeled with caspase-3 and To-Pro-3. There was widespread caspase-3 labeling in the hair cell region. The nuclei in the hair cells region tend to be highly condensed whereas the nuclei in the supporting cell regions tend to be large, round and reasonably normal in appearance.
DMSO is an amphipathic molecule with a polar domain and two apolar methyl groups making it soluble in both aqueous and organic media. It is one of the most common solvents for dissolving hydrophobic compounds employed in vivo or in vitro studies. DMSO has frequently been used in studies of the inner ear; however, its potential side effects are unknown and have been largely overlooked in previous studies (Basile et al., 1996; Guitton et al., 2003; Harris et al., 2005; Lee et al., 2002; So et al., 2005). Our in vitro results obtained with postnatal cochlear cultures showed little or no evidence of hair cell damage following 24 h treatment with 0.25% DMSO or less. These results are consistent with two previous reports that used 0.1% DMSO and found no negative side effects (Lee et al., 2001; Matsui et al., 2002). A 0.5% DMSO concentration caused a slight increase in IHC (Figure 2C, G). However, once the concentration reached 0.75% there were clear signs of damage to IHC over most of the cochlea as well as damage to OHC in the base of the cochlea. DMSO concentrations of 1% have been used with embryonic day 13 and 18 mouse organ cultures, but there was no report of damage in these cultures (Mueller et al., 2002). The absence of DMSO-induced damage could conceivably due to the addition of heparin, which has been reported to protect cells from damage (Han et al., 2005; Slofstra et al., 2005). Another factor that could play a role is the age of the cochlear cultures. Hair cells in embryonic day 13 and 18 cultures are much less mature that those in our postnatal day 2−3 cultures.
DMSO concentrations of 2−3% caused mild to moderate stereocilia damage and disarray and some swelling of the hair cell body. DMSO concentrations of 5−6% lead to the loss of stereocilia bundles, depletion of actin within the cuticular plate and significant swelling of the hair cells throughout cochlea. The pattern of hair cell loss induced by DMSO progressed from base to apex consistent with most other ototoxic drugs (Ding et al., 2002; Kamimura et al., 1999). In cochlear cultures, ototoxic drugs typically damage the OHC first followed by IHC (Ding et al., 2002; Zhang et al., 2003). However, a major difference in the damage pattern was that DMSO caused greater loss of IHC than OHC. This was clearly apparent at a concentration of 0.75% where IHC loss was evident over the entire length of the cochlea, yet OHC loss was present only in the extreme base.
Even though DMSO induced significant hair cell loss, the nearby Hensen cells had normal nuclei and were negative for TUNEL, caspase-3, caspase-8 and caspase-9 (Figure 3). These results clearly indicate that hair cells are more vulnerable to DMSO than the supporting cells and that IHC are more susceptible to DMSO damage than OHC. The reasons for these cellular differences in DMSO cytotoxicity are unclear. Other cell types appear to be more resistant to the cytotoxic effects of DMSO than hair cells. For example, DMSO concentration of 10% had no cytotoxic effects on Caco2/TC7 cells (Da Violante et al., 2002).
Oncosis, sometimes referred to as necrosis, is a form of cell death that is characterized by enlarged cellular organelles and swelling of the cell's membrane, which can cause the plasma membrane to rupture and release its contents into the extracellular environment (Bohne et al., 2007; Jaeschke et al., 2003; Majno et al., 1995). High doses of DMSO induced significant swelling of the hair cells suggesting that hair cells might be dying by oncosis. However, our morphological and biochemical assessment indicated that most hair cells in the DMSO treated cultures had shrunken and condensed nuclei, and stained positive for TUNEL, caspase-3, caspase-8, and caspase-9. These later results suggest that DMSO induced hair cell death was occurring predominantly by apoptosis (Jaeschke et al., 2003). Our caspase-9 results are in line with the apoptotic cell death pattern seen in murine lymphoma cells where DMSO caused a collapse of the mitochondrial membrane potential, release of cytochrome c from the mitochondria followed by activation of caspase-9 and caspase-3 (Liu et al., 2001). However, unlike cochlear hair cells, DMSO failed to activate the extrinsic caspase-8 pathway in lymphoma cells.
In summary, our results indicate that 24 h treatment with DMSO concentrations of 0.25% or less caused little or no damage to postnatal hair cells in cochlear organotypic cultures. DMSO is known to be a scavenger of the hydroxyl radical and it is possible that low concentrations of DMSO will be protective in cochlear cultures (Repine et al., 1981). However, this should be regarded as speculative at this time since there is no direct evidence for a protective effect in our data. However, our results clearly indicate that DMSO concentrations of 0.5% or higher are toxic to both IHC and OHC in postnatal rat cochlear cultures. DMSO-induced hair cell damage progressed from base to apex; however, unlike other ototoxic compounds IHC were more vulnerable than OHC. Thus, culture media with DMSO concentrations equal to or greater than 25% are likely to be damaging to postnatal cochlear cultures.
DMSO is sometimes used to dissolve drugs that are applied to the inner ear via the middle ear or round window (Harris et al., 2005; Liu et al., 2006); however, it is unclear if DMSO is toxic to adult hair cells in vivo. It is conceivable that adult hair cells are more resistant to the toxic effects of DMSO. However, given that high concentrations of DMSO are toxic in vitro, future studies should be carried out to assess the toxicity of DMSO in vivo. While in vivo studies are important clinically, determining the dose-response relationship for DMSO is challenging. It is technically difficult to control the volume of DMSO that is applied to and remains on the round window membrane. Moreover, it is difficult to know how long it takes DMSO to pass through the round window; how long it takes to diffuse from the base to the apex of the cochlea and what the actual concentration is at any location and point in time. Since the volume of fluid within the cochlea is relatively large compared to the volume of DMSO that can be applied to the round window, the actual concentration of DMSO reaching the hair cells in vivo may be quite low. An alternative method for assessing the vulnerability of IHC and OHC to DMSO in vivo would be to deliver DMSO into scala tympani using an osmotic pump with a specified flow rate and concentration. This method would seem to offer the greatest precision for establishing the dose-response toxicity of DMSO in vivo. With the increasing likelihood that therapeutic agents will be delivered directly to the cochlear, this clinically relevant question merits further study.
Research supported by NIH grant R01 DC00630
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.