CRYOBIOLOGY
28, 210-215 (1991)
Quantitation of Dimethyl Sulfoxide in Solutions and Tissues by High-Performance Liquid Chromatography JOHN F. CARPENTER AND PATTI E. DAWSON CryoLife, Inc., 2211 New Market Parkway, Suite 142, Marietta, Georgia 30067 We have developed a rapid and simple method to determine the level of dimethyl sulfoxide (Me,SO) in both solutions and tissue samples. For analysis of Me,SO in a cryopreservation medium, the solution is simply diluted in 10% (vol/vol) methanol and centrifuged. Then an aliquot of the supematant is assayed by high-performance liquid chromatography. For tissue samples, the wet weight is measured and the intact sample is extracted with 10% (vollvol) methanol (e.g., 10 ml/g wet wt) in a seated vial. The extract is then diluted and centrifuged, and an aliquot of the supematant is assayed. The dry weight of the tissue is measured after the methanol-extracted sample is placed into ether for 2 h and air-dried overnight. The water content of the tissue is calculated as the difference between the wet and the dry weights. The concentration of Me,SO in the aqueous compartment of the tissue can then be calculated by taking into account the concentration of Me,SO in the extract and the dilution factor, based on the tissue water volume and the volume of methanol used to extract the Me,SO. The calculated values for porcine myocardium samples correlated 1:l with the actual Me,SO concentrations in the solutions in which the tissue samples were equilibrated. Finally, we present results documenting the usefulness of this assay by following the time course of Me,SO penetration into core versus peripheral regions of l-cm3 samples of porcine myocardium. 0 1991 Academic press, Inc.
In the development of freeze-thawing protocols utilizing dimethyl sulfoxide (Me,SO), it is advantageous to determine the distribution of the cryoprotectant within the experimental sample, especially with relatively bulky tissues. For example, the time-course of penetration of the cryoprotectant prior to freezing and the kinetics of elution during post-thaw washes can greatly influence the design of the cryopreservation method. In the former case, it is critical that sufficient cryoprotectant penetrate into the sample prior to freezing. In the latter example, monitoring the Me,SO concentration during step-wise washing steps is prudent to avoid potentially damaging osmotic stress to the cells (cf. (3)). To obtain these measurements without the use of radiolabeled Me,SO it is necessary to have a sensitive and reliable chemical method to determine the concentration
Received July 30,199O; accepted September 4, 1990
of Me,SO. Previous workers have employed gas chromatography to measure Me$O in blood and serum samples (1, 2). Ivey and Haddad (4) used a high-performance liquid chromatographic (HPLC) assay employing ion-exchange chromatography to measure Me,SO in seawater. For the present study we have developed a HPLC assay to measure Me$O in complex tissue culture media, employing reversephase chromatography (modified after (5) ), that is rapid and simple. In addition, we have devised a straightforward method to extract Me,SO from tissue samples. This extraction technique used in combination with the HPLC assay allows one to calculate accurately the concentration of Me+0 in a sample. As a demonstration of the utility of this method, we have determined the levels of Me+0 in porcine myocardium as a function of the position within the tissue sample and the length of incubation in a 1 M Me$O solution.
210 001 l-2240/91 $3.00 Copyright B 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
DIMETHYL
MATERIALS
AND
SULFOXIDE
METHODS
Materials. High purity methanol and Me,SO (Burdick and Jackson Brand) were purchased from Baxter. Dulbecco’s modified Eagle medium with 10% fetal calf serum (DMEM) was purchased from GIBCO and pig hearts were obtained from a local meat processing plant. HPLC apparatus. The apparatus consisted of a LDC/Milton Roy Model CM4000 multiple solvent delivery system coupled to a Milton Roy SpectroMonitor 3000 variable wavelength detector. Data were acquired and analyzed with a Milton Roy Model CI 4000 plotting integrator, linked to a Unisys personal computer which ran LDC Analytical’s LCadvantage 1.22 software. The column used was a LDCA4ilton Roy Spherisorb ODS2 C,, (5 pm) measuring 4.6 x 25 cm. The injector was a Rheodyne Model 7125 with a 20-p,l injection loop. A Brownlee Labs Spheri-18 Cis (5 pm> guard cartridge was placed between the injector and the column. HPLC assay method. The HPLC assay method was a modification of that described by Rao et al. (5). The mobile phase contained 10% methanol (voh’vol) in highpurity deionized water (i.e., >18 MfI resistance) and was degassed by helium sparging. The flow rate was 1.0 mYmin, and the detector was set at 214 nm. All samples were diluted as needed in 10% methanol and centrifuged for 10 min in a Beckman microfuge prior to injection. Extraction of MezSO from porcine myocardium. Porcine myocardium was cut into pieces and placed into DMEM containing Me,SO. After the requisite time, the tissue sample was removed and blotted dry on a Kimwipe, and tissue wet weight was measured. The tissue sample was placed in a glass vial and 10 ml of 10% methanol was added per gram of tissue wet weight. The vial was then tightly capped. The tissue was extracted for at least 6 h at room temperature. It was found that for tissue samples of
QUANTITATION
BY
HPLC
211
no more than about 300 mg, this was sufficient time to allow equilibration of Me,SO with the tissue and the surrounding methanol solution (data not shown). Longer extraction times are needed for larger tissue samples (i.e., >I g). An aliquot of the extract was removed, diluted l/10 with 10% methanol, and assayed for Me,SO according to the HPLC method outlined above. A known concentration of Me,SO prepared in DMEM was used as a standard. The standard was diluted l/11 with 10% methanol to simulate the dilution arising upon addition of the methanol to the tissue. This preparation was then further diluted l/10 with 10% methanol and centrifuged and assayed as described above. Determination of tissue dry weight. The 10% methanol was removed from the vial containing the tissue and replaced with 100% methanol. After a minimum of 6 h the methanol was removed and replaced with ether. The tissue was incubated in the ether for approximately 2-3 h and then the ether was aspirated from the vial. The vial containing the tissue was placed in a venting fume hood to evaporate the residual ether. The dry weight of the tissue was then measured. Calculation of Me$O concentration in tissue samples. To determine the concentration of Me,SO in the aqueous compartment of a tissue sample, the water volume within the tissue must be calculated. This value is the difference between the wet and dry weights, in grams, multiplied by 1 ml/g. Once this value is known, then the actual dilution of the tissue’s aqueous compartment during the extraction procedure can be calculated. For example, a tissue sample with a l-g wet weight may have a dry weight of 0.15 g. Thus, the water content of the tissue would be 0.85 g or 0.85 ml. Since the l-g tissue sample would be extracted with 10 ml of 10% methanol, the actual dilution of the tissue’s aqueous compartment would be l/12.8, instead of l/l 1. How this value is used in the calculation of the con-
212
CARPENTER
AND
centration of Me,SO in the tissue is shown in Eq. 14. Tissue wet wt. - Tissue dry wt. = Tissue water wt. [l] Tissue water wt. X 1 ml/g = Tissue water volume PI Tissue water volume + (10 ml/g x Tissue wet wt.) Tissue water volume = Tissue dilution factor Peak area extract x Tissue dilution factor Peak 11 area standard Std. [Me$O] = Tissue [MezSO].
[31
DAWSON
placed into 1 M Me+0 (in DMEM). At timed intervals, cubes were removed from the solution and blotted dry on a Kimwipe. Then, 2-3 mm was cut from each edge of the block. The first section removed was blotted dry and used as the sample from the “periphery.” The remaining central region was blotted dry and used as the sample from the “core.” Me,SO concentrations in the samples were determined as described above. RESULTS
X
AND
DISCUSSION
Chromatographic Resolution Tissue Culture Medium
of MetSO
in
In the initial stages of the assay development, chromatography conditions were ad[41 justed to give clear resolution of the Me,SO in The tissue wet and dry weights are deter- peak from the peaks for constituents complex tissue culture media. Tissue culmined as described above. The difference (Eq. 1) in these values is assumed to be ture solutions are often used in the cryorepresentative of the weight of water in the preservation of cells and tissues and thus sample, which is converted to volume in there is a need to be able to monitor Me$O Eq. [2] by dividing the water weight by the concentrations in these systems. For examdensity of water. The tissue dilution factor ple, the time course of elution of Me,SO calculated in Eq. [3] is the actual dilution of from a tissue sample after thawing and the tissue aqueous compartment that arises transfer to a diluting medium could be folwhen 10 ml (10% methanol) per gram of lowed by assaying for Me,SO concentratissue weight wet is used to extract the tis- tion in the diluting medium as a function of sue. The peak areas referred to in Eq. [4] time. Figure 1A shows a typical chromatogram are those for Me,SO peaks on the HPLC for DMEM containing 10% fetal calf serum. chromatograms for the tissue extract and Figure 1B is a chromatogram for a sample the Me,SO standard. The second term in treated identically, except that it contained the equation is a correction for the differ1.1 mg/ml Me,SO. The peak for Me,SO is ence between the dilution factor for the tiswell separated from the peaks for the consue and the standard. The denominator stituents in DMEM, and the assay is rapid value of 11 is based on the addition of 1 since the elution time for Me$O is less volume of the standard to 10 volumes of than 4 min. These results indicate that this 10% methanol, which approximates the diassay can be used to quantitate Me,SO raplution of aqueous compartment of the tissue idly in complex mixtures such as tissue culsample during extraction. Std. [Me,SO] and ture media. Tissue [Me,SO] refer to the concentrations of Me,SO in the standard and the tissue Calibration of the Assay aqueous compartment, respectively. Determination of temporal and spatial Plots of the Me,SO chromatograph peak distribution of Me,SO in tissue. Porcine area or peak height versus the concentramyocardium was cut into 1 cm3 cubes and tion (0.5-5.5 l&ml) of Me,SO were linear
DIMETHYL
lime
SULFOXIDE
QUANTITATION
BY
213
HPLC
Time
(Min.)
(Min.)
1. Typical chromatograms for DMEM (with 10% fetal calf serum) alone (A) and DMEM (with 10% fetal calf serum) plus 1.1 mg’ml Me,SO (B). Both samples were diluted l/200 with 10% (vol/vol) methanol and centrifuged in a Beckman microfuge for 5 mitt prior to injection. The detector was set at 0.01 absorbance units full scale. FIG.
(Fig. 2). These results indicate that either peak parameter can be used to calibrate the assay. Furthermore, once the linearity of the plots for a range of standards is established, a single standard can be used for routine analysis of samples, provided that the samples have Me,SO contents falling within the linear range. Accuracy of Calculated Values for Me,SO Concentration in Tissue Samples
A more critical application of an assay for Me,SO is the determination of the penetration of the cryoprotectant into a tissue or the elution of the cryoprotectant from a
sample. With relatively bulky tissue samples it is difficult to predict the time needed for the cryoprotectant to reach either a desired concentration in the tissue or an equilibrium between the tissue and the bathing medium. In order to determine the kinetics of cryoprotectant penetration and distribution, it is essential that the assay method provide an accurate determination of the actual concentration of Me,SO in the aqueous compartment of the tissue. To determine the accuracy of the calculated values for Me,SO concentration in tissue samples, porcine myocardium was cut into pieces of approximately 200-300 mg wet wt. Tissue samples were incubated in
90
0 0
1.0
2.0
3.0
4.0
[Me901(w/m0
5.0
6.0
0
1.0
2.0
3.0
4.0
5.0
6.0
[MQOI (w/m0
FIG. 2. Calibration plots for HPLC assay of dimethyl sulfoxide. Me,!30 was prepared as described in FIG. 1. The 1.1 mg/ml Me,SO stock solution was then diluted with DMEM (with 10% fetal calf serum) to give a series of standards ranging in Me,SO concentration from 0.1 l-l. 1 mg/ml Me,SO. All samples were then diluted l/200 with 10% (voYvo1) methanol and centrifuged in a Beckman microfuge for 5 min prior to injection. The chromatographic peak area (A) and height (B) are plotted against the concentration of Me+30 in the methanol-diluted samples.
214
CARPENTER
AND DAWSON
20 ml of DMEM (with 10% fetal calf serum) containing 0.2, 0.4, 0.6, 0.8, or 1.0 M Me,SO. Three individual tissue samples were prepared for each concentration of Me,SO. The preparations were placed in tightly capped vials and stored at 4°C for 24 h to allow equilibration of Me+0 between the tissue and the solution. The tissue samples were processed further as described above. The calculated values are plotted as a function of the concentration of Me,SO in the bathing solution in Fig. 3. There is a 1: 1 correlation between the calculated value in the tissue and the actual value in the bathing solution. These findings indicate that the Me,SO concentrations determined with the current method accurately reflect the actual concentration in the aqueous compartment in the tissue. Thus, this method can be used to quantitate the concentration of Me,SO in tissue samples.
posure to 1 M Me,SO, in the peripheral edge and central core of l-cm3 samples of porcine myocardium. The results show that, as expected, the penetration of Me,SO into the core region lags behind that for the periphery (Fig. 4). For both regions the kinetics of Me,SO penetration display a biphasic nature, with an initial rapid influx followed by a more gradual increase in Me,SO concentration. This experiment demonstrates that the current protocol would have utility for assessing the timecourse and spatial distribution of Me,SO penetration into tissue samples. Such information is critical for the design of human tissue cryopreservation protocols. Conversely, the same methods can be used for following the elution of Me,SO from tissue during cryoprotectant dilution procedures.
Temporal and Spatial Distribution of Me,,!30 in Tissue Samples
High-performance liquid chromatography can be used to determine rapidly the concentration of Me,SO in complex solutions, such as tissue culture media. When this assay is used in combination with the described Me,SO extraction procedure, accurate measurement of Me,SO concentration in tissue samples can be obtained. This method should be of use to investigators who need to follow either the penetration of Me,SO into a tissue sample prior to cryo-
Once the accuracy of the quantitation method was validated, the next step was to measure the Me,SO concentration in tissue samples under conditions where the cryoprotectant had not reached an equilibrium between the tissue and the bathing medium. We determined the concentration of Me$O, as a function of time of tissue exf
g F .c
0.B
6
0.6
CONCLUSIONS
1.0. 1.0
Periphery 0-o-O
0.8 3 zo.4. E P
!j
”
0’
8 1___1
0.q
rz=
/
ol/ 0
Slope
.
0.4
0.2 [Me,SO]
0.6
(M) in Bathing
3 3 E
= 1 ,016 0.999
0.8
0’ 0.6
0-o
/
0.4 -0 O/O
/
1.0
Solution
FIG. 3. Correlation of calculated concentration of dimethyl sulfoxide in porcine myocardium with dimethyl sulfoxide concentration in bathing solution. Tissue samples were prepared and processed as described in the text. The data points plotted are the means 2 SD for three separate tissue samples.
0
20
40 Incubation
60
SO
100
120
Time (Min.)
FIG. 4. Concentration of dimethyl sulfoxide in periphery and central core of l-cm3 samples of porcine myocardium as a function of time of incubation in 1 M dimethyl s&oxide. Tissue samples were prepared and processed as described in the text.
DIMETHYL
SULFOXIDE
preservation or the elution of the cryoprotectant during post-thaw washing of the tissue. REFERENCES 1. Garretson, S. E., and Aitchison, J. P. Comparison of dimethyl sulfoxide levels in whole blood and serum using an autosampler-equipped gas chromatograph. Ann. NY Acad. Sci. 411, 328331 (1983). 2. Garretson, S. E., and Aitchison, J. P. Determination of dimethyl sulfoxide in serum and other body fluids by gas chromatography. J. Anal. Toxicol. 6, 76-81 (1982).
QUANTITATION
BY HPLC
215
3. Hempling, H. G. Mass transfer of liquids across biological barriers. In “The Biophysics of Organ Preservation” (D. E. Pegg and A. M. Karow, Eds.), pp. 51-78, Plenum Press, New York, 1987. -~ 4. Ivey, J. P., and Haddad, P. R. Determination of dimethylsulphoxide using ion-exclusion chromatography with ultraviolet absorption detection. J. Chromatogr. 391, 309-314 (1987). 5. Rao, P. S., Rujikam, N., and Luber, J. M. Highperformance liquid chromatography method for the direct quantitation of oxy radicals in myocardium and blood by means of 1,3-dimethylurea and dimethyl sulfoxide. .Z. Chromatogr. 459, 269-273 (1988).
28, 210-215 (1991)
Quantitation of Dimethyl Sulfoxide in Solutions and Tissues by High-Performance Liquid Chromatography JOHN F. CARPENTER AND PATTI E. DAWSON CryoLife, Inc., 2211 New Market Parkway, Suite 142, Marietta, Georgia 30067 We have developed a rapid and simple method to determine the level of dimethyl sulfoxide (Me,SO) in both solutions and tissue samples. For analysis of Me,SO in a cryopreservation medium, the solution is simply diluted in 10% (vol/vol) methanol and centrifuged. Then an aliquot of the supematant is assayed by high-performance liquid chromatography. For tissue samples, the wet weight is measured and the intact sample is extracted with 10% (vollvol) methanol (e.g., 10 ml/g wet wt) in a seated vial. The extract is then diluted and centrifuged, and an aliquot of the supematant is assayed. The dry weight of the tissue is measured after the methanol-extracted sample is placed into ether for 2 h and air-dried overnight. The water content of the tissue is calculated as the difference between the wet and the dry weights. The concentration of Me,SO in the aqueous compartment of the tissue can then be calculated by taking into account the concentration of Me,SO in the extract and the dilution factor, based on the tissue water volume and the volume of methanol used to extract the Me,SO. The calculated values for porcine myocardium samples correlated 1:l with the actual Me,SO concentrations in the solutions in which the tissue samples were equilibrated. Finally, we present results documenting the usefulness of this assay by following the time course of Me,SO penetration into core versus peripheral regions of l-cm3 samples of porcine myocardium. 0 1991 Academic press, Inc.
In the development of freeze-thawing protocols utilizing dimethyl sulfoxide (Me,SO), it is advantageous to determine the distribution of the cryoprotectant within the experimental sample, especially with relatively bulky tissues. For example, the time-course of penetration of the cryoprotectant prior to freezing and the kinetics of elution during post-thaw washes can greatly influence the design of the cryopreservation method. In the former case, it is critical that sufficient cryoprotectant penetrate into the sample prior to freezing. In the latter example, monitoring the Me,SO concentration during step-wise washing steps is prudent to avoid potentially damaging osmotic stress to the cells (cf. (3)). To obtain these measurements without the use of radiolabeled Me,SO it is necessary to have a sensitive and reliable chemical method to determine the concentration
Received July 30,199O; accepted September 4, 1990
of Me,SO. Previous workers have employed gas chromatography to measure Me$O in blood and serum samples (1, 2). Ivey and Haddad (4) used a high-performance liquid chromatographic (HPLC) assay employing ion-exchange chromatography to measure Me,SO in seawater. For the present study we have developed a HPLC assay to measure Me$O in complex tissue culture media, employing reversephase chromatography (modified after (5) ), that is rapid and simple. In addition, we have devised a straightforward method to extract Me,SO from tissue samples. This extraction technique used in combination with the HPLC assay allows one to calculate accurately the concentration of Me+0 in a sample. As a demonstration of the utility of this method, we have determined the levels of Me+0 in porcine myocardium as a function of the position within the tissue sample and the length of incubation in a 1 M Me$O solution.
210 001 l-2240/91 $3.00 Copyright B 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
DIMETHYL
MATERIALS
AND
SULFOXIDE
METHODS
Materials. High purity methanol and Me,SO (Burdick and Jackson Brand) were purchased from Baxter. Dulbecco’s modified Eagle medium with 10% fetal calf serum (DMEM) was purchased from GIBCO and pig hearts were obtained from a local meat processing plant. HPLC apparatus. The apparatus consisted of a LDC/Milton Roy Model CM4000 multiple solvent delivery system coupled to a Milton Roy SpectroMonitor 3000 variable wavelength detector. Data were acquired and analyzed with a Milton Roy Model CI 4000 plotting integrator, linked to a Unisys personal computer which ran LDC Analytical’s LCadvantage 1.22 software. The column used was a LDCA4ilton Roy Spherisorb ODS2 C,, (5 pm) measuring 4.6 x 25 cm. The injector was a Rheodyne Model 7125 with a 20-p,l injection loop. A Brownlee Labs Spheri-18 Cis (5 pm> guard cartridge was placed between the injector and the column. HPLC assay method. The HPLC assay method was a modification of that described by Rao et al. (5). The mobile phase contained 10% methanol (voh’vol) in highpurity deionized water (i.e., >18 MfI resistance) and was degassed by helium sparging. The flow rate was 1.0 mYmin, and the detector was set at 214 nm. All samples were diluted as needed in 10% methanol and centrifuged for 10 min in a Beckman microfuge prior to injection. Extraction of MezSO from porcine myocardium. Porcine myocardium was cut into pieces and placed into DMEM containing Me,SO. After the requisite time, the tissue sample was removed and blotted dry on a Kimwipe, and tissue wet weight was measured. The tissue sample was placed in a glass vial and 10 ml of 10% methanol was added per gram of tissue wet weight. The vial was then tightly capped. The tissue was extracted for at least 6 h at room temperature. It was found that for tissue samples of
QUANTITATION
BY
HPLC
211
no more than about 300 mg, this was sufficient time to allow equilibration of Me,SO with the tissue and the surrounding methanol solution (data not shown). Longer extraction times are needed for larger tissue samples (i.e., >I g). An aliquot of the extract was removed, diluted l/10 with 10% methanol, and assayed for Me,SO according to the HPLC method outlined above. A known concentration of Me,SO prepared in DMEM was used as a standard. The standard was diluted l/11 with 10% methanol to simulate the dilution arising upon addition of the methanol to the tissue. This preparation was then further diluted l/10 with 10% methanol and centrifuged and assayed as described above. Determination of tissue dry weight. The 10% methanol was removed from the vial containing the tissue and replaced with 100% methanol. After a minimum of 6 h the methanol was removed and replaced with ether. The tissue was incubated in the ether for approximately 2-3 h and then the ether was aspirated from the vial. The vial containing the tissue was placed in a venting fume hood to evaporate the residual ether. The dry weight of the tissue was then measured. Calculation of Me$O concentration in tissue samples. To determine the concentration of Me,SO in the aqueous compartment of a tissue sample, the water volume within the tissue must be calculated. This value is the difference between the wet and dry weights, in grams, multiplied by 1 ml/g. Once this value is known, then the actual dilution of the tissue’s aqueous compartment during the extraction procedure can be calculated. For example, a tissue sample with a l-g wet weight may have a dry weight of 0.15 g. Thus, the water content of the tissue would be 0.85 g or 0.85 ml. Since the l-g tissue sample would be extracted with 10 ml of 10% methanol, the actual dilution of the tissue’s aqueous compartment would be l/12.8, instead of l/l 1. How this value is used in the calculation of the con-
212
CARPENTER
AND
centration of Me,SO in the tissue is shown in Eq. 14. Tissue wet wt. - Tissue dry wt. = Tissue water wt. [l] Tissue water wt. X 1 ml/g = Tissue water volume PI Tissue water volume + (10 ml/g x Tissue wet wt.) Tissue water volume = Tissue dilution factor Peak area extract x Tissue dilution factor Peak 11 area standard Std. [Me$O] = Tissue [MezSO].
[31
DAWSON
placed into 1 M Me+0 (in DMEM). At timed intervals, cubes were removed from the solution and blotted dry on a Kimwipe. Then, 2-3 mm was cut from each edge of the block. The first section removed was blotted dry and used as the sample from the “periphery.” The remaining central region was blotted dry and used as the sample from the “core.” Me,SO concentrations in the samples were determined as described above. RESULTS
X
AND
DISCUSSION
Chromatographic Resolution Tissue Culture Medium
of MetSO
in
In the initial stages of the assay development, chromatography conditions were ad[41 justed to give clear resolution of the Me,SO in The tissue wet and dry weights are deter- peak from the peaks for constituents complex tissue culture media. Tissue culmined as described above. The difference (Eq. 1) in these values is assumed to be ture solutions are often used in the cryorepresentative of the weight of water in the preservation of cells and tissues and thus sample, which is converted to volume in there is a need to be able to monitor Me$O Eq. [2] by dividing the water weight by the concentrations in these systems. For examdensity of water. The tissue dilution factor ple, the time course of elution of Me,SO calculated in Eq. [3] is the actual dilution of from a tissue sample after thawing and the tissue aqueous compartment that arises transfer to a diluting medium could be folwhen 10 ml (10% methanol) per gram of lowed by assaying for Me,SO concentratissue weight wet is used to extract the tis- tion in the diluting medium as a function of sue. The peak areas referred to in Eq. [4] time. Figure 1A shows a typical chromatogram are those for Me,SO peaks on the HPLC for DMEM containing 10% fetal calf serum. chromatograms for the tissue extract and Figure 1B is a chromatogram for a sample the Me,SO standard. The second term in treated identically, except that it contained the equation is a correction for the differ1.1 mg/ml Me,SO. The peak for Me,SO is ence between the dilution factor for the tiswell separated from the peaks for the consue and the standard. The denominator stituents in DMEM, and the assay is rapid value of 11 is based on the addition of 1 since the elution time for Me$O is less volume of the standard to 10 volumes of than 4 min. These results indicate that this 10% methanol, which approximates the diassay can be used to quantitate Me,SO raplution of aqueous compartment of the tissue idly in complex mixtures such as tissue culsample during extraction. Std. [Me,SO] and ture media. Tissue [Me,SO] refer to the concentrations of Me,SO in the standard and the tissue Calibration of the Assay aqueous compartment, respectively. Determination of temporal and spatial Plots of the Me,SO chromatograph peak distribution of Me,SO in tissue. Porcine area or peak height versus the concentramyocardium was cut into 1 cm3 cubes and tion (0.5-5.5 l&ml) of Me,SO were linear
DIMETHYL
lime
SULFOXIDE
QUANTITATION
BY
213
HPLC
Time
(Min.)
(Min.)
1. Typical chromatograms for DMEM (with 10% fetal calf serum) alone (A) and DMEM (with 10% fetal calf serum) plus 1.1 mg’ml Me,SO (B). Both samples were diluted l/200 with 10% (vol/vol) methanol and centrifuged in a Beckman microfuge for 5 mitt prior to injection. The detector was set at 0.01 absorbance units full scale. FIG.
(Fig. 2). These results indicate that either peak parameter can be used to calibrate the assay. Furthermore, once the linearity of the plots for a range of standards is established, a single standard can be used for routine analysis of samples, provided that the samples have Me,SO contents falling within the linear range. Accuracy of Calculated Values for Me,SO Concentration in Tissue Samples
A more critical application of an assay for Me,SO is the determination of the penetration of the cryoprotectant into a tissue or the elution of the cryoprotectant from a
sample. With relatively bulky tissue samples it is difficult to predict the time needed for the cryoprotectant to reach either a desired concentration in the tissue or an equilibrium between the tissue and the bathing medium. In order to determine the kinetics of cryoprotectant penetration and distribution, it is essential that the assay method provide an accurate determination of the actual concentration of Me,SO in the aqueous compartment of the tissue. To determine the accuracy of the calculated values for Me,SO concentration in tissue samples, porcine myocardium was cut into pieces of approximately 200-300 mg wet wt. Tissue samples were incubated in
90
0 0
1.0
2.0
3.0
4.0
[Me901(w/m0
5.0
6.0
0
1.0
2.0
3.0
4.0
5.0
6.0
[MQOI (w/m0
FIG. 2. Calibration plots for HPLC assay of dimethyl sulfoxide. Me,!30 was prepared as described in FIG. 1. The 1.1 mg/ml Me,SO stock solution was then diluted with DMEM (with 10% fetal calf serum) to give a series of standards ranging in Me,SO concentration from 0.1 l-l. 1 mg/ml Me,SO. All samples were then diluted l/200 with 10% (voYvo1) methanol and centrifuged in a Beckman microfuge for 5 min prior to injection. The chromatographic peak area (A) and height (B) are plotted against the concentration of Me+30 in the methanol-diluted samples.
214
CARPENTER
AND DAWSON
20 ml of DMEM (with 10% fetal calf serum) containing 0.2, 0.4, 0.6, 0.8, or 1.0 M Me,SO. Three individual tissue samples were prepared for each concentration of Me,SO. The preparations were placed in tightly capped vials and stored at 4°C for 24 h to allow equilibration of Me+0 between the tissue and the solution. The tissue samples were processed further as described above. The calculated values are plotted as a function of the concentration of Me,SO in the bathing solution in Fig. 3. There is a 1: 1 correlation between the calculated value in the tissue and the actual value in the bathing solution. These findings indicate that the Me,SO concentrations determined with the current method accurately reflect the actual concentration in the aqueous compartment in the tissue. Thus, this method can be used to quantitate the concentration of Me,SO in tissue samples.
posure to 1 M Me,SO, in the peripheral edge and central core of l-cm3 samples of porcine myocardium. The results show that, as expected, the penetration of Me,SO into the core region lags behind that for the periphery (Fig. 4). For both regions the kinetics of Me,SO penetration display a biphasic nature, with an initial rapid influx followed by a more gradual increase in Me,SO concentration. This experiment demonstrates that the current protocol would have utility for assessing the timecourse and spatial distribution of Me,SO penetration into tissue samples. Such information is critical for the design of human tissue cryopreservation protocols. Conversely, the same methods can be used for following the elution of Me,SO from tissue during cryoprotectant dilution procedures.
Temporal and Spatial Distribution of Me,,!30 in Tissue Samples
High-performance liquid chromatography can be used to determine rapidly the concentration of Me,SO in complex solutions, such as tissue culture media. When this assay is used in combination with the described Me,SO extraction procedure, accurate measurement of Me,SO concentration in tissue samples can be obtained. This method should be of use to investigators who need to follow either the penetration of Me,SO into a tissue sample prior to cryo-
Once the accuracy of the quantitation method was validated, the next step was to measure the Me,SO concentration in tissue samples under conditions where the cryoprotectant had not reached an equilibrium between the tissue and the bathing medium. We determined the concentration of Me$O, as a function of time of tissue exf
g F .c
0.B
6
0.6
CONCLUSIONS
1.0. 1.0
Periphery 0-o-O
0.8 3 zo.4. E P
!j
”
0’
8 1___1
0.q
rz=
/
ol/ 0
Slope
.
0.4
0.2 [Me,SO]
0.6
(M) in Bathing
3 3 E
= 1 ,016 0.999
0.8
0’ 0.6
0-o
/
0.4 -0 O/O
/
1.0
Solution
FIG. 3. Correlation of calculated concentration of dimethyl sulfoxide in porcine myocardium with dimethyl sulfoxide concentration in bathing solution. Tissue samples were prepared and processed as described in the text. The data points plotted are the means 2 SD for three separate tissue samples.
0
20
40 Incubation
60
SO
100
120
Time (Min.)
FIG. 4. Concentration of dimethyl sulfoxide in periphery and central core of l-cm3 samples of porcine myocardium as a function of time of incubation in 1 M dimethyl s&oxide. Tissue samples were prepared and processed as described in the text.
DIMETHYL
SULFOXIDE
preservation or the elution of the cryoprotectant during post-thaw washing of the tissue. REFERENCES 1. Garretson, S. E., and Aitchison, J. P. Comparison of dimethyl sulfoxide levels in whole blood and serum using an autosampler-equipped gas chromatograph. Ann. NY Acad. Sci. 411, 328331 (1983). 2. Garretson, S. E., and Aitchison, J. P. Determination of dimethyl sulfoxide in serum and other body fluids by gas chromatography. J. Anal. Toxicol. 6, 76-81 (1982).
QUANTITATION
BY HPLC
215
3. Hempling, H. G. Mass transfer of liquids across biological barriers. In “The Biophysics of Organ Preservation” (D. E. Pegg and A. M. Karow, Eds.), pp. 51-78, Plenum Press, New York, 1987. -~ 4. Ivey, J. P., and Haddad, P. R. Determination of dimethylsulphoxide using ion-exclusion chromatography with ultraviolet absorption detection. J. Chromatogr. 391, 309-314 (1987). 5. Rao, P. S., Rujikam, N., and Luber, J. M. Highperformance liquid chromatography method for the direct quantitation of oxy radicals in myocardium and blood by means of 1,3-dimethylurea and dimethyl sulfoxide. .Z. Chromatogr. 459, 269-273 (1988).
