We thank the referees for the helpful comments. We have incorporated all the referees’ suggestions in the revised manuscript. Below we list responses in the sequence in which they were raised in the referees’ reports.
Response to Referee #1:
Reviewer #1: This is a straightforward study of the biophysical properties of a pair of cell types important to clinical studies. The biophysical portion is a simple and straightforward analysis of 2 parameter transport in ideal dilute solutions under a number of simplifying assumptions. The manuscript is clearly written and the experiments and analysis are, for the most part, well described. The impact of this manuscript may be in the design of future protocols, but in the absence of a number of key parameters, this paper seems to be a cataloguing of some of the key data. The scope would be considerably extended if, for example, osmotic tolerance data and/or CPA toxicity data were coupled to the BVH and permeability data to produce actual predictions of cell damage during CPA exposure instead of the vague comment in line 207; "CPA addition and removal should be relatively slow..." The use of a single temperature also limits the possibility of cooling rate prediction. In short, this manuscript leaves the reader wanting more than it provides at present.
We thank for the reviewer’s comments. We agree with the reviewer on the manuscript’s limited scope. This is a fundamental cryobiological study of the human vaginal mucosal immune cells’ membrane permeabilities to water and different cryoprotective agents (the four most widely used ones) at room temperature investigated with the developed microfluidic perfusion channel. A thorough complete cryobiological study is far beyond of this scope, such as osmotic tolerance limits of the cells, CPA toxicity to the cells, intracellular ice formation temperatures, cell membrane permeabilities at lower temperatures (especially at subzero temperatures), and others. As suggested by the reviewer, osmotic tolerance limit and CPA toxicity test data were included in the revised manuscript (Page xx, Line xx). In order to keep the manuscript in focus and with reasonable length, some of the other studies are planned to be presented in separated manuscripts (e.g., the measurement of cell membrane permeabilities at subzero temperatures with differential scanning calorimetry and cell water transport during cooling). Some work is still on going in our laboratory (such as cryopreservation of the immune cells, mass diffusivity of CPA in mucosal tissues, cryopreservation and vitrification of mucosal tissues, and others.). A comprehensive and intensive study of the cryopreservation of vaginal mucosal immune cells and mucosal tissues will be finished and presented in the near future.
Additionally, I have several key complaints with this manuscript beyond its limited scope.
1) The assessment of the BVH relationship is limited. First, no hyposmotic data were used and a minimum of points were used for a regression in the hyperosmotic range. This is important to assure that cells are linear osmometers throughout the range in which their volumes may be predicted during CPA equilibration. Second, the details of the regression are missing: was the regression forced to fit through the isosmotic volume point? Third, there are no errors on the V_0 and/or V_b values. Fourth, the value of V_0 is not published.
We thank for the reviewer’s comments. We agree with the reviewer that the more hypo- and/or hypertonic data points, the more accurate for the curve-fitted Vb values. In our tests, the mucosal immune cell osmotic response to two hypertonic solutions was tested and plotted with the isotonic volume by Boyle-van’t Hoff linear regression. We also totally agree with the reviewer that in our analysis, it was assumed the immune cells are linear osmometers throughout the range in which their volumes may be predicted during CPA equilibration (generally in hypertonic range!), although this may be not true in large osmolality range including both hypo- and hypertonic concentrations (Igor I. Katkov, Cryobiology 62 (2011), 232–241.).
One major reason for this experiment plan was due to the very limited population of the cells we could get. The vaginal mucosal tissues were obtained as bypass from surgery, and then digested, labelled and sorted for T cells and macrophages. After treatment, usually the total population of the cells was about 10,000. In addition, Boyle van’t Hoff plots with only three data points, or plots only in hypo- or hypertonic range were used in some papers to predict the osmotically inactive cell volumes, such as:
H. Chen, H. Shen, S. Heimfeld, K.K. Tran, J. Reems, A. Folch, D. Gao, A microfluidic study of mouse dendritic cell membrane transport properties of water and cryoprotectants, International Journal of Heat and Mass Transfer 51 (2008) 5687-5694. (1x PBS, 2xPBS, 3xPBS)
H. Chen, J.J.P. Purtteman, S. Heimfeld, A. Folch, D. Gao, Development of a microfluidic device for determination of cell osmotic behavior and membrane transport properties, Cryobiology 55 (2007) 200-209. (1x, 1.5x, 2x, 3x PBS)
K. Edashige, M. Tanaka, N. Ichimaru, S. Ota, K. Yazawa, Y. Higashino, M. Sakamoto, Y. Yamaji, T. Kuwano, D.M. Valdez, F.W. Kleinhans, M. Kasai, Channel- Dependent Permeation of Water and Glycerol in Mouse Morulae 1, Biol. Reprod. 74 (2006) 625-632. doi:10.1095/biolreprod.105.045823.( 3 hypertonic sucrose solutions)
M. Hagedorn, J. Ricker, M. McCarthy, S.A. Meyers, T.R. Tiersch, Z.M. Varga, F.W. Kleinhans, Biophysics of zebrafish (Danio rerio) sperm, Cryobiology 58 (2009) 12-19. doi:10.1016/j.cryobiol.2008.09.013. (200, 300, and 600 mOsm/kg)
J. Liu, S. Mullen, Q. Meng, J. Critser, A. Dinnyes, Determination of oocyte membrane permeability coefficients and their application to cryopreservation in a rabbit model, Cryobiology 59 (2009) 127-134. doi:10.1016/j.cryobiol.2009.06.002. (isosmotic and 3 hypertonic soutions)
Y. Yamaji, D.M. Valdez Jr., S. Seki, K. Yazawa, C. Urakawa, B. Jin, M. Kasai, F.W. Kleinhans, K. Edashige, Cryoprotectant permeability of aquaporin-3 expressed in Xenopus oocytes, Cryobiology 53 (2006) 258-267. doi:10.1016/j.cryobiol.2006.06.008. (isosmotic and 3 hypotonic solutions)
As suggested by the reviewer, the cell volumes and error bars in isosmotic solution for T cells and macrophages are added in the revised manuscript (Page xx, Line xx). The V0 of T cells was xxx, and V0 of macrophages was xxx+xxx. There were no restraints on the linear regression.
2) More important is the dependence of the volume estimation using a spherical cell assumption. Figure 1 shows the cross sectional areas used in the spherical assumption. Using ImageJ, I looked at the first image from Figure 1 and found its circularity to be ~90%. This metric is the ratio of 2 pi r_equiv (the equivalent radius of a disc using your method) and the actual perimeter length. This 10% error dramatically affects the volume estimates in ways that are difficult to predict, but importantly, changes as a function of volume, making accurate estimates of the error during the protocol nearly impossible. I suggest a careful analysis of the volume as a function of cross sectional area given non-circular cross sectional morphologies. Even if we assumed the volume to be well modeled by this spherical assumption, the surface area of a nonspherical object will be at least 10% higher (given the rough estimate from the ImageJ analysis). Dramatically affecting Lp and Ps estimates that depend on this value.
We thank for the reviewer’s comments.
Reviewer #2:
This manuscript adeptly describes the use of a microfluidics chamber to determine the biophysical parameters of vaginal mucosa derived T cells and macrophages. It is a very well written manuscript of interest to the readership of Cryobiology. I have a few concerns and comments/questions that I think could improve the manuscript.
The data presented to determine osmotically inactive fraction is not adequate for extrapolation. There are only two data points aside from the normalization value. This doesn't yield a reliable extrapolation. There should minimally be another data set added at a different concentration, or otherwise some attempt needs to be made to bracket the range of error in the determination of osmotically inactive fraction. The former approach is preferred.
We thank for the reviewer’s comments. We agree with the reviewer that the more hypo- and/or hypertonic data points, the more accurate for the curve-fitted Vb values. In our tests, the mucosal immune cell osmotic response to two hypertonic solutions was tested and plotted with the isotonic volume by Boyle-van’t Hoff linear regression. One major reason for this experiment plan was due to the very limited population of the cells we could get. The vaginal mucosal tissues were obtained as bypass from surgery, and then digested, labelled and sorted for T cells and macrophages. After treatment, usually the total population of the cells was about 10,000.
Boyle van’t Hoff plots with only three data points, or plots only in hypo- or hypertonic range were used in some papers to predict the osmotically inactive cell volumes, such as:
H. Chen, H. Shen, S. Heimfeld, K.K. Tran, J. Reems, A. Folch, D. Gao, A microfluidic study of mouse dendritic cell membrane transport properties of water and cryoprotectants, International Journal of Heat and Mass Transfer 51 (2008) 5687-5694. (1x PBS, 2xPBS, 3xPBS)
H. Chen, J.J.P. Purtteman, S. Heimfeld, A. Folch, D. Gao, Development of a microfluidic device for determination of cell osmotic behavior and membrane transport properties, Cryobiology 55 (2007) 200-209. (1x, 1.5x, 2x, 3x PBS)
K. Edashige, M. Tanaka, N. Ichimaru, S. Ota, K. Yazawa, Y. Higashino, M. Sakamoto, Y. Yamaji, T. Kuwano, D.M. Valdez, F.W. Kleinhans, M. Kasai, Channel- Dependent Permeation of Water and Glycerol in Mouse Morulae 1, Biol. Reprod. 74 (2006) 625-632. doi:10.1095/biolreprod.105.045823.( 3 hypertonic sucrose solutions)
M. Hagedorn, J. Ricker, M. McCarthy, S.A. Meyers, T.R. Tiersch, Z.M. Varga, F.W. Kleinhans, Biophysics of zebrafish (Danio rerio) sperm, Cryobiology 58 (2009) 12-19. doi:10.1016/j.cryobiol.2008.09.013. (200, 300, and 600 mOsm/kg)
J. Liu, S. Mullen, Q. Meng, J. Critser, A. Dinnyes, Determination of oocyte membrane permeability coefficients and their application to cryopreservation in a rabbit model, Cryobiology 59 (2009) 127-134. doi:10.1016/j.cryobiol.2009.06.002. (isosmotic and 3 hypertonic soutions)
Y. Yamaji, D.M. Valdez Jr., S. Seki, K. Yazawa, C. Urakawa, B. Jin, M. Kasai, F.W. Kleinhans, K. Edashige, Cryoprotectant permeability of aquaporin-3 expressed in Xenopus oocytes, Cryobiology 53 (2006) 258-267. doi:10.1016/j.cryobiol.2006.06.008. (isosmotic and 3 hypotonic solutions)
As suggested by the reviewer, the cell volumes and error bars in isosmotic solution for T cells and macrophages are added in the revised manuscript (Page xx, Line xx). The V0 of T cells was xxx, and V0 of macrophages was xxx+xxx. There were no restraints on the linear regression.
T cells isolated from blood are often cryopreserved within 4 hours to prevent deterioration in diagnostic potential. In this study the tissue samples were stored overnight before isolation of cells. The lengthy storage and processing time is likely to alter the biophysical properties. Were freshly isolated samples analyzed? (ie. samples not stored overnight). Is the method to collect and prepare cells for this study the standard assay method used in clinical practice at the associated facilities of the authors? Also what effect is labeling likely to have on the biophysical properties? Has this comparison been made previously? (assuming there are ways to isolate cells without labels)
We thank for the reviewer’s comments. The overnight rest of the tissue samples was necessary for logistical reasons. Mucosal immune cells isolated after overnight rest retain the ability to produce cytokines in response to stimulation with antigen or superantigen, as well as to proliferate in response to stimulation, both key functions of immune cells. Additionally, viability after isolation was typically in excess of 75%. During the sorting procedure, we isolated exclusively viable cells, so the cells used in the assays in the microfluidic device were all viable.
The enzymatic digestion procedure is in common use for isolation of immune cells from the cervicovaginal or colorectal mucosa.
It is not possible to isolate these cells without labeling. The cells of interest are a minority population (T cells 5-10% and macrophages 1-5%) of all mucosal cells. There is no way to purify them without labeling. The labeling method we used involved binding antibodies with fluorescent labels to the cells and purifying them by flow cytometric sorting. An alternative possibility would have been to use antibodies with magnetic beads attached and purify the cells with a magnetic column. In both cases, the cells have to be labeled. Sorting has the advantage of producing a purer population. In [1] the authors compared gene expression profiles of cells isolated by magnetic bead purification or sorting and found little difference in the cellular gene expression based on purification method.
Beliakova-Bethell, N. et al. The effect of cell subset isolation method on gene expression in leukocytes. Cytometry. A (2013). doi:10.1002/cyto.a.22352
Minor: Table 4 contains references to other cell types. Have T cells or macrophages been characterized previously from any origin? This would be a useful comparison. Similarly putting the data in the context of current cryopreservation methods for T cells and macrophages from vaginal mucosa or other origins would be important (including PBMCs).
We thank for the reviewer’s comments.
Note: While submitting the revised manuscript, please double check the author names provided in the submission so that authorship related changes are made in the revision stage. If your manuscript is accepted, any authorship change will involve approval from co-authors and respective editor handling the submission and this may cause a significant delay in publishing your manuscript.
We thank the referees for the helpful comments. We have incorporated all the referees’ suggestions in the revised manuscript. Below we list responses in the sequence in which they were raised in the referees’ reports.
Response to Referee #1:
Reviewer #1: This is a straightforward study of the biophysical properties of a pair of cell types important to clinical studies. The biophysical portion is a simple and straightforward analysis of 2 parameter transport in ideal dilute solutions under a number of simplifying assumptions. The manuscript is clearly written and the experiments and analysis are, for the most part, well described. The impact of this manuscript may be in the design of future protocols, but in the absence of a number of key parameters, this paper seems to be a cataloguing of some of the key data. The scope would be considerably extended if, for example, osmotic tolerance data and/or CPA toxicity data were coupled to the BVH and permeability data to produce actual predictions of cell damage during CPA exposure instead of the vague comment in line 207; "CPA addition and removal should be relatively slow..." The use of a single temperature also limits the possibility of cooling rate prediction. In short, this manuscript leaves the reader wanting more than it provides at present.
We thank for the reviewer’s comments. We agree with the reviewer on the manuscript’s limited scope. This is a fundamental cryobiological study of the human vaginal mucosal immune cells’ membrane permeabilities to water and different cryoprotective agents (the four most widely used ones) at room temperature investigated with the developed microfluidic perfusion channel. A thorough complete cryobiological study is far beyond of this scope, such as osmotic tolerance limits of the cells, CPA toxicity to the cells, intracellular ice formation temperatures, cell membrane permeabilities at lower temperatures (especially at subzero temperatures), and others. As suggested by the reviewer, osmotic tolerance limit and CPA toxicity test data were included in the revised manuscript (Page xx, Line xx). In order to keep the manuscript in focus and with reasonable length, some of the other studies are planned to be presented in separated manuscripts (e.g., the measurement of cell membrane permeabilities at subzero temperatures with differential scanning calorimetry and cell water transport during cooling). Some work is still on going in our laboratory (such as cryopreservation of the immune cells, mass diffusivity of CPA in mucosal tissues, cryopreservation and vitrification of mucosal tissues, and others.). A comprehensive and intensive study of the cryopreservation of vaginal mucosal immune cells and mucosal tissues will be finished and presented in the near future.
Additionally, I have several key complaints with this manuscript beyond its limited scope.
1) The assessment of the BVH relationship is limited. First, no hyposmotic data were used and a minimum of points were used for a regression in the hyperosmotic range. This is important to assure that cells are linear osmometers throughout the range in which their volumes may be predicted during CPA equilibration. Second, the details of the regression are missing: was the regression forced to fit through the isosmotic volume point? Third, there are no errors on the V_0 and/or V_b values. Fourth, the value of V_0 is not published.
We thank for the reviewer’s comments. We agree with the reviewer that the more hypo- and/or hypertonic data points, the more accurate for the curve-fitted Vb values. In our tests, the mucosal immune cell osmotic response to two hypertonic solutions was tested and plotted with the isotonic volume by Boyle-van’t Hoff linear regression. We also totally agree with the reviewer that in our analysis, it was assumed the immune cells are linear osmometers throughout the range in which their volumes may be predicted during CPA equilibration (generally in hypertonic range!), although this may be not true in large osmolality range including both hypo- and hypertonic concentrations (Igor I. Katkov, Cryobiology 62 (2011), 232–241.). One major reason for this experiment plan was due to the very limited population of the cells we could get. The vaginal mucosal tissues were obtained as bypass from surgery, and then digested, labelled and sorted for T cells and macrophages. After treatment, usually the total population of the cells was about 10,000. In addition, Boyle van’t Hoff plots with only three data points, or plots only in hypo- or hypertonic range were used in some papers to predict the osmotically inactive cell volumes, such as: H. Chen, H. Shen, S. Heimfeld, K.K. Tran, J. Reems, A. Folch, D. Gao, A microfluidic study of mouse dendritic cell membrane transport properties of water and cryoprotectants, International Journal of Heat and Mass Transfer 51 (2008) 5687-5694. (1x PBS, 2xPBS, 3xPBS) H. Chen, J.J.P. Purtteman, S. Heimfeld, A. Folch, D. Gao, Development of a microfluidic device for determination of cell osmotic behavior and membrane transport properties, Cryobiology 55 (2007) 200-209. (1x, 1.5x, 2x, 3x PBS) K. Edashige, M. Tanaka, N. Ichimaru, S. Ota, K. Yazawa, Y. Higashino, M. Sakamoto, Y. Yamaji, T. Kuwano, D.M. Valdez, F.W. Kleinhans, M. Kasai, Channel- Dependent Permeation of Water and Glycerol in Mouse Morulae 1, Biol. Reprod. 74 (2006) 625-632. doi:10.1095/biolreprod.105.045823.( 3 hypertonic sucrose solutions) M. Hagedorn, J. Ricker, M. McCarthy, S.A. Meyers, T.R. Tiersch, Z.M. Varga, F.W. Kleinhans, Biophysics of zebrafish (Danio rerio) sperm, Cryobiology 58 (2009) 12-19. doi:10.1016/j.cryobiol.2008.09.013. (200, 300, and 600 mOsm/kg) J. Liu, S. Mullen, Q. Meng, J. Critser, A. Dinnyes, Determination of oocyte membrane permeability coefficients and their application to cryopreservation in a rabbit model, Cryobiology 59 (2009) 127-134. doi:10.1016/j.cryobiol.2009.06.002. (isosmotic and 3 hypertonic soutions) Y. Yamaji, D.M. Valdez Jr., S. Seki, K. Yazawa, C. Urakawa, B. Jin, M. Kasai, F.W. Kleinhans, K. Edashige, Cryoprotectant permeability of aquaporin-3 expressed in Xenopus oocytes, Cryobiology 53 (2006) 258-267. doi:10.1016/j.cryobiol.2006.06.008. (isosmotic and 3 hypotonic solutions) As suggested by the reviewer, the cell volumes and error bars in isosmotic solution for T cells and macrophages are added in the revised manuscript (Page xx, Line xx). The V0 of T cells was xxx, and V0 of macrophages was xxx+xxx. There were no restraints on the linear regression. 2) More important is the dependence of the volume estimation using a spherical cell assumption. Figure 1 shows the cross sectional areas used in the spherical assumption. Using ImageJ, I looked at the first image from Figure 1 and found its circularity to be ~90%. This metric is the ratio of 2 pi r_equiv (the equivalent radius of a disc using your method) and the actual perimeter length. This 10% error dramatically affects the volume estimates in ways that are difficult to predict, but importantly, changes as a function of volume, making accurate estimates of the error during the protocol nearly impossible. I suggest a careful analysis of the volume as a function of cross sectional area given non-circular cross sectional morphologies. Even if we assumed the volume to be well modeled by this spherical assumption, the surface area of a nonspherical object will be at least 10% higher (given the rough estimate from the ImageJ analysis). Dramatically affecting Lp and Ps estimates that depend on this value. We thank for the reviewer’s comments.
Reviewer #2: This manuscript adeptly describes the use of a microfluidics chamber to determine the biophysical parameters of vaginal mucosa derived T cells and macrophages. It is a very well written manuscript of interest to the readership of Cryobiology. I have a few concerns and comments/questions that I think could improve the manuscript.
Boyle van’t Hoff plots with only three data points, or plots only in hypo- or hypertonic range were used in some papers to predict the osmotically inactive cell volumes, such as: H. Chen, H. Shen, S. Heimfeld, K.K. Tran, J. Reems, A. Folch, D. Gao, A microfluidic study of mouse dendritic cell membrane transport properties of water and cryoprotectants, International Journal of Heat and Mass Transfer 51 (2008) 5687-5694. (1x PBS, 2xPBS, 3xPBS) H. Chen, J.J.P. Purtteman, S. Heimfeld, A. Folch, D. Gao, Development of a microfluidic device for determination of cell osmotic behavior and membrane transport properties, Cryobiology 55 (2007) 200-209. (1x, 1.5x, 2x, 3x PBS) K. Edashige, M. Tanaka, N. Ichimaru, S. Ota, K. Yazawa, Y. Higashino, M. Sakamoto, Y. Yamaji, T. Kuwano, D.M. Valdez, F.W. Kleinhans, M. Kasai, Channel- Dependent Permeation of Water and Glycerol in Mouse Morulae 1, Biol. Reprod. 74 (2006) 625-632. doi:10.1095/biolreprod.105.045823.( 3 hypertonic sucrose solutions) M. Hagedorn, J. Ricker, M. McCarthy, S.A. Meyers, T.R. Tiersch, Z.M. Varga, F.W. Kleinhans, Biophysics of zebrafish (Danio rerio) sperm, Cryobiology 58 (2009) 12-19. doi:10.1016/j.cryobiol.2008.09.013. (200, 300, and 600 mOsm/kg) J. Liu, S. Mullen, Q. Meng, J. Critser, A. Dinnyes, Determination of oocyte membrane permeability coefficients and their application to cryopreservation in a rabbit model, Cryobiology 59 (2009) 127-134. doi:10.1016/j.cryobiol.2009.06.002. (isosmotic and 3 hypertonic soutions) Y. Yamaji, D.M. Valdez Jr., S. Seki, K. Yazawa, C. Urakawa, B. Jin, M. Kasai, F.W. Kleinhans, K. Edashige, Cryoprotectant permeability of aquaporin-3 expressed in Xenopus oocytes, Cryobiology 53 (2006) 258-267. doi:10.1016/j.cryobiol.2006.06.008. (isosmotic and 3 hypotonic solutions) As suggested by the reviewer, the cell volumes and error bars in isosmotic solution for T cells and macrophages are added in the revised manuscript (Page xx, Line xx). The V0 of T cells was xxx, and V0 of macrophages was xxx+xxx. There were no restraints on the linear regression.
The enzymatic digestion procedure is in common use for isolation of immune cells from the cervicovaginal or colorectal mucosa.
It is not possible to isolate these cells without labeling. The cells of interest are a minority population (T cells 5-10% and macrophages 1-5%) of all mucosal cells. There is no way to purify them without labeling. The labeling method we used involved binding antibodies with fluorescent labels to the cells and purifying them by flow cytometric sorting. An alternative possibility would have been to use antibodies with magnetic beads attached and purify the cells with a magnetic column. In both cases, the cells have to be labeled. Sorting has the advantage of producing a purer population. In [1] the authors compared gene expression profiles of cells isolated by magnetic bead purification or sorting and found little difference in the cellular gene expression based on purification method.
Minor: Table 4 contains references to other cell types. Have T cells or macrophages been characterized previously from any origin? This would be a useful comparison. Similarly putting the data in the context of current cryopreservation methods for T cells and macrophages from vaginal mucosa or other origins would be important (including PBMCs). We thank for the reviewer’s comments.
Note: While submitting the revised manuscript, please double check the author names provided in the submission so that authorship related changes are made in the revision stage. If your manuscript is accepted, any authorship change will involve approval from co-authors and respective editor handling the submission and this may cause a significant delay in publishing your manuscript.