Cell-cell contact affects membrane integrity after intracellular freezing

The response of cells to freezing depends critically on the presence of an intact cell membrane. During rapid cooling, the cell plasma membrane may no longer be an effective barrier to ice propagation and can be breached by extracellular ice resulting in the nucleation of the supercooled cytoplasm. In tissues, the formation of intracellular ice is compounded by the presence of cell-cell and cell-surface interactions. Three different hamster fibroblast model systems were used to simulate structures found in organized tissues. Samples were supercooled to an experimental temperature on a cryostage and ice nucleated at the constant temperature. A dual fluorescent staining technique was used for the quantitative assessment of the integrity of the cell plasma membrane. A novel technique using the fluorescent stain SYTO was used for the detection of intracellular ice formation (IIF) in cell monolayers. The cumulative incidence of cells with a loss of membrane integrity and the cumulative incidence of IIF were determined as a function of temperature. Cells in suspension and individual attached cells showed no significant difference in the number of cells that formed intracellular ice and those that lost membrane integrity. For cells in a monolayer, with cell-cell contact, intracellular ice formation did not result in the immediate disruption of the plasma membrane in the majority of cells. This introduces the potential for minimizing damage due to IIF and for developing strategies for the cryoprotection of tissues during rapid cooling.     

Water transport and IIF parameters for a connective tissue equivalent

Understanding the biophysical processes that govern freezing injury of a tissue equivalent (TE) is an important step in characterizing and improving the cryopreservation of these systems. TEs were formed by entrapping human dermal fibroblasts (HDFs) in collagen or in fibrin gels. Freezing studies were conducted using a Linkam cryostage fitted to an optical microscope allowing observation of the TEs cooled under controlled rates between 5 and 130 degrees C/min. Typically, freezing of cellular systems results in two biophysical processes that are both dependent on the cooling rate: dehydration and/or intracellular ice formation (IIF). Both these processes can potentially be destructive to cells. In this study, the biophysics of freezing cells in collagen and fibrin TEs have been quantified and compared to freezing cells in suspension. Experimental data were fitted in numerical models to extract parameters that governed water permeability, E(Lp) and L(pg), and intracellular ice nucleation, omega(o) and kappa(o). Results indicate that major differences exist between freezing HDFs in suspension and in a tissue equivalent. During freezing, 55% of the HDFs in suspension formed IIF as compared to 100% of HDFs forming IIF in collagen and fibrin TE at a cooling rate of 130 degrees C/min. Also, both the water permeability and the IIF parameters were determined to be higher for HDFs in TEs as compared to cell suspensions. Between the TEs, HDFs in fibrin TE exhibited higher values for the biophysical parameters as compared to HDFs in collagen TE. The observed biophysics seems to indicate that cell-cell and cell-matrix interactions play a major role in ice propagation in TEs.     

Kinetics and mechanism of intercellular ice propagation in a micropatterned tissue construct

Understanding the effects of cell-cell interaction on intracellular ice formation (IIF) is required to design optimized protocols for cryopreservation of tissue. To determine the effects of cell-cell interactions during tissue freezing, without confounding effects from uncontrolled factors (such as time in culture, cell geometry, and cell-substrate interactions), HepG2 cells were cultured in pairs on glass coverslips micropatterned with polyethylene glycol disilane, such that each cell interacted with exactly one adjacent cell. Assuming the cell pair to be a finite state system, being either in an unfrozen state (no ice in either cell), a singlet state (IIF in one cell only), or a doublet state (IIF in both cells), the kinetics of state transitions were theoretically modeled and cryomicroscopically measured. The rate of intercellular ice propagation, estimated from the measured singlet state probability, increased in the first 24 h of culture and remained steady thereafter. In cell pairs cultured for 24 h and treated with the gap junction blocker 18beta-glycyrrhetinic acid before freezing, the intercellular ice propagation rate was lower than in untreated controls (p < 0.001), but significantly greater than zero (p < 0.0001). These results suggest that gap junctions mediate some, but not all, mechanisms of ice propagation in tissue.     

Kinetics of intracellular ice formation in one-dimensional arrays of interacting biological cells

Although cell-cell interactions are known to significantly affect the kinetics of intracellular ice formation (IIF) during tissue freezing, this effect is not well understood. Progress in elucidating the mechanism and role of intercellular ice propagation in tissue freezing has been hampered in part by limitations in experimental design and data analysis. Thus, using rapid-cooling cryomicroscopy, IIF was measured in adherent cells cultured in micropatterned linear constructs (to control cell-cell interactions and minimize confounding factors). By fitting a Markov chain model to IIF data from micropatterned HepG2 cell pairs, the nondimensional rate of intercellular ice propagation was found to be alpha = 10.4 +/- 0.1. Using this measurement, a new generator matrix was derived to predict the kinetics of IIF in linear four-cell constructs; cryomicroscopic measurements of IIF state probabilities in micropatterned four-cell arrays conformed with theoretical predictions (p < 0.05), validating the modeling assumptions. Thus, the theoretical model was extended to allow prediction of IIF in larger tissues, using Monte Carlo techniques. Simulations were performed to investigate the effects of tissue size and ice propagation rate, for one-dimensional tissue constructs containing up to 100 cells and nondimensional propagation rates in the range 0.1 < or = alpha < or = 1000.     

Effects of solution composition on the theoretical prediction of ice nucleation kinetics and thermodynamics

Predictions by various mathematical models of intracellular ice formation (proposed by Mazur, Pitt, Toner, and Karlsson, respectively) were compared to the known thermodynamic and kinetic behavior of ice formation in supercooled aqueous systems. The older models (Mazur, Pitt, and Toner) significantly underestimated the magnitude of colligative nonequilibrium freezing point depression in response to increased concentration of solutes, such as salts or cryoprotectants. Furthermore, kinetics predicted using phenomenological models (by Mazur and Pitt) exhibited implausible temperature-dependence, with the probability of intracellular ice formation being allowed to increase even at temperatures below the glass transition point. The Toner model, on the other hand, produced invalid results at temperatures below −48 °C. The Karlsson model was the only model that consistently yielded realistic predictions over a wide range of temperatures and solute concentrations, especially in the presence of cryoprotectant additives. To facilitate adoption of the Karlsson model of intracellular ice nucleation, the complete set of model equations has been collected and described in detail.     

Effects of Intercellular Junction Protein Expression on Intracellular Ice Formation in Mouse Insulinoma Cells

The development of cryopreservation procedures for tissues has proven to be difficult in part because cells within tissue are more susceptible to intracellular ice formation (IIF) than are isolated cells. In particular, previous studies suggest that cell-cell interactions increase the likelihood of IIF by enabling propagation of ice between neighboring cells, a process thought to be mediated by gap junction channels. In this study, we investigated the effects of cell-cell interactions on IIF using three genetically modified strains of the mouse insulinoma cell line MIN6, each of which expressed key intercellular junction proteins (connexin-36, E-cadherin, and occludin) at different levels. High-speed video cryomicroscopy was used to visualize the freezing process in pairs of adherent cells, revealing that the initial IIF event in a given cell pair was correlated with a hitherto unrecognized precursor phenomenon: penetration of extracellular ice into paracellular spaces at the cell-cell interface. Such paracellular ice penetration occurred in the majority of cell pairs observed, and typically preceded and colocalized with the IIF initiation events. Paracellular ice penetration was generally not observed at temperatures >−5.65°C, which is consistent with a penetration mechanism via defects in tight-junction barriers at the cell-cell interface. Although the maximum temperature of paracellular penetration was similar for all four cell strains, genetically modified cells exhibited a significantly higher frequency of ice penetration and a higher mean IIF temperature than did wild-type cells. A four-state Markov chain model was used to quantify the rate constants of the paracellular ice penetration process, the penetration-associated IIF initiation process, and the intercellular ice propagation process. In the initial stages of freezing (>−15°C), junction protein expression appeared to only have a modest effect on the kinetics of propagative IIF, and even cell strains lacking the gap junction protein connexin-36 exhibited nonnegligible ice propagation rates.     

Effects of Intercellular Junction Protein Expression on Intracellular Ice Formation in Mouse Insulinoma Cells

The development of cryopreservation procedures for tissues has proven to be difficult in part because cells within tissue are more susceptible to intracellular ice formation (IIF) than are isolated cells. In particular, previous studies suggest that cell-cell interactions increase the likelihood of IIF by enabling propagation of ice between neighboring cells, a process thought to be mediated by gap junction channels. In this study, we investigated the effects of cell-cell interactions on IIF using three genetically modified strains of the mouse insulinoma cell line MIN6, each of which expressed key intercellular junction proteins (connexin-36, E-cadherin, and occludin) at different levels. High-speed video cryomicroscopy was used to visualize the freezing process in pairs of adherent cells, revealing that the initial IIF event in a given cell pair was correlated with a hitherto unrecognized precursor phenomenon: penetration of extracellular ice into paracellular spaces at the cell-cell interface. Such paracellular ice penetration occurred in the majority of cell pairs observed, and typically preceded and colocalized with the IIF initiation events. Paracellular ice penetration was generally not observed at temperatures >−5.65°C, which is consistent with a penetration mechanism via defects in tight-junction barriers at the cell-cell interface. Although the maximum temperature of paracellular penetration was similar for all four cell strains, genetically modified cells exhibited a significantly higher frequency of ice penetration and a higher mean IIF temperature than did wild-type cells. A four-state Markov chain model was used to quantify the rate constants of the paracellular ice penetration process, the penetration-associated IIF initiation process, and the intercellular ice propagation process. In the initial stages of freezing (>−15°C), junction protein expression appeared to only have a modest effect on the kinetics of propagative IIF, and even cell strains lacking the gap junction protein connexin-36 exhibited nonnegligible ice propagation rates.     

Mitochondria and membrane cryoinjury in micropatterned cells: Effects of cell-cell interactions

The maintenance of cell membrane integrity is an absolute minimum criterion for the selection of a successful cryopreservation process; however, it is often used as the sole determinant of cell “viability”. Membrane stresses and strains that develop with cell volume fluctuations are only one component of the overall cellular response to freezing. Damage to organelles resulting from excessive concentration of intracellular solutes and/or the alternation of molecular signalling events may affect post-thaw outcomes. As the low temperature response of cells is affected by the presence of cell-cell interactions, the cryopreservation of tissues and tissue model systems would benefit from a more detailed understanding of the sites and mechanisms of cryoinjury. The purpose of this study was to examine the relationship between mitochondria and plasma membrane damage in frozen micropatterned cells and to identify the role of cell-cell interactions. Madin Darby Canine Kidney cells (MDCK) were micropatterned using a polydimethylsiloxane (PDMS) elastomeric stamp to create non-adhesive regions of agarose on untreated glass substrates. Five different cell arrangements were used to examine the effect of cell-cell contact: single cells, cell doublets, linear arrangement of cells, randomly arranged cells and confluent monolayers. Cells were cooled in a programmable alcohol bath at 1 °C/min to −40 °C after extracellular ice nucleation at −5 °C. Post-thaw plasma membrane integrity and mitochondria depolarization were determined using trypan blue and the lipophilic, cyanine derivative JC-1, respectively. alamarBlue was used to assess the post-thaw metabolic activity of the cell arrangements. We found that the incidence of plasma membrane damage and mitochondria integrity increased with decreasing temperature and was dependent on the degree of cell-cell interaction. Mitochondria damage was evident in cells that displayed intact plasma membranes, however this injury could be reversed in the micropatterned cells that are exposed to suprazero temperatures. The results from this study suggest that the exclusive use of membrane integrity as a measure of cell “viability” does not consider subcellular injury that may contribute to delayed recovery and/or cell death following low temperature exposures.     

Protective effect of intracellular ice during freezing?

Injury results during freezing when cells are exposed to increasing concentrations of solutes or by the formation of intracellular ice. Methods to protect cells from the damaging effects of freezing have focused on the addition of cryoprotective chemicals and the determination of optimal cooling rates. Based on other studies of innocuous intracellular ice formation, this study investigates the potential for this ice to protect cells from injury during subsequent slow cooling. V-79W Chinese hamster fibroblasts and Madin-Darby Canine Kidney (MDCK) cells were cultured as single attached cells or confluent monolayers. The incidence of intracellular ice formation (IIF) in the cultures at the start of cooling was pre-determined using one of two different extracellular ice nucleation temperatures (-5 or -10 degrees C). Samples were then cooled at 1 degrees C/min to the experimental temperature (-5 to -40 degrees C) where samples were warmed rapidly and cell survival assessed using membrane integrity and metabolic activity. For single attached cells, the lower ice nucleation temperature, corresponding to increased incidence of IIF, resulted in decreased post-thaw cell recovery. In contrast, confluent monolayers in which IIF has been shown to be innocuous, show higher survival after cooling to temperatures as low as -40 degrees C, supporting the concept that intracellular ice confers cryoprotection by preventing cell dehydration during subsequent slow cooling.     

New bioengineering insights into human neural precursor cell expansion in culture

Understanding initial cell growth, interactions associated with the process of expansion of human neural precursor cells (hNPCs), and cellular events pre- and postdifferentiation are important for developing bioprocessing protocols to reproducibly generate multipotent cells that can be used in basic research or the treatment of neurodegenerative disorders. Herein, we report the in vitro responses of telencephalon hNPCs grown in a serum-free growth medium using time-lapse live imaging as well as cell-surface marker, aggregate size, and immunocytochemical analyses. Time-lapse analysis of hNPC initial expansion indicated that cell-surface attachment in stationary culture and the frequency of cell-cell interaction in suspension conditions are important for subsequent aggregate formation and hNPC growth. In the absence of cell-surface attachment in low-attachment stationary culture, large aggregates of cells were formed and expansion was adversely affected. The majority of the telencephalon hNPCs expressed CD29, CD90, and CD44 (cell surface markers involved in cell-ECM and cell-cell interactions to regulate biological functions such as proliferation), suggesting that cell-surface attachment and cell-cell interactions play a significant role in the subsequent formation of cell aggregates and the expansion of hNPCs. Before differentiation, about 90% of the cells stained positive for nestin and expressed two neural precursor cells surface markers (CD133 and CD24). Upon withdrawal of growth cytokines, hNPCs first underwent cell division and then differentiated preferentially towards a neuronal rather than a glial phenotype. This study provides key information regarding human NPC behavior under different culture conditions and favorable culture conditions that are important in establishing reproducible hNPC expansion protocols.     

Experimental study of intracellular ice growth in human umbilical vein endothelial cells

Study of the intracellular ice formation (IIF) and growth is essential to the mechanistic understanding of cellular damage through freezing. In the aid of high speed and high resolution cryo-imaging technology, the transient intracellular ice formation and growth processes of the attached human umbilical vein endothelial cells (HUVEC) were successfully captured during freezing. It was found that the intracellular ice nucleation site was on the cell membrane closer to the nucleus. The ice growth was directional and toward the nucleus, which covered the whole nucleus before growing into the cytoplasm. The crystal growth rate in the nucleus was much larger than that in the cytoplasm, and its morphology was influenced by the cooling rate. During the thawing process, small crystals fused into larger ones inside the nucleus. Moreover, the cumulative fraction of the HUVEC with IIF was mainly dependent on the cooling rate not the confluence of the cells attached.     

Mechanical signaling through the discoidin domain receptor 1 plays a central role in tissue fibrosis

ABSTRACT

The preservation of tissue and organ architecture and function depends on tightly regulated interactions of cells with the extracellular matrix (ECM). These interactions are maintained in a dynamic equilibrium that balances intracellular, myosin-generated tension with extracellular resistance conferred by the mechanical properties of the extracellular matrix. Disturbances of this equilibrium can lead to the development of fibrotic lesions that are associated with a wide repertoire of high prevalence diseases including obstructive cardiovascular diseases, muscular dystrophy and cancer. Mechanotransduction is the process by which mechanical cues are converted into biochemical signals. At the core of mechanotransduction are sensory systems, which are frequently located at sites of cell-ECM and cell-cell contacts. As integrins (cell-ECM junctions) and cadherins (cell-cell contacts) have been extensively studied, we focus here on the properties of the discoidin domain receptor 1 (DDR1), a tyrosine kinase that mediates cell adhesion to collagen. DDR1 expression is positively associated with fibrotic lesions of heart, kidney, liver, lung and perivascular tissues. As the most common end-point of all fibrotic disorders is dysregulated collagen remodeling, we consider here the mechanical signaling functions of DDR1 in processing of fibrillar collagen that lead to tissue fibrosis.     

Modeling positive regulatory feedbacks in cell-cell interactions

Our current understanding of molecular mechanisms of cellular regulation still does not support quantitative predictions of the overall growth kinetics of normal or malignant tissues. However, discernment of the role of growth-factor mediated cell-cell communication in tissue kinetics is possible by the use of simple mathematical models. Here we discuss the design and use of mathematical models in quantifying the contribution of autocrine and paracrine (i.e., humoral) interactions to the kinetics of tissue growth. We present models that include a humorally mediated regulatory feedback among cells built into phenomenological mathematical models of growth. Application of these models to data exemplifies the finite contributions of positive feedback in cell-cell interactions to the overall tissue growth. In addition, we propose a perturbation approach to allow separation of cell-cell interactions dependent on the perturbing agent (such as hormone antagonists in hormone-dependent tissues) from cell-cell interactions independent of it.     

Cooling rate dependent biophysical and viability response shift with attachment state in human dermal fibroblast cells

While studies on the freezing of cells in suspension have been carried out extensively, corresponding studies with cells in the attached state and in tissue or tissue-equivalents are less developed. As attachment is a hallmark of the tissue state it is important to understand its impact on biophysics and viability to better apply freezing towards tissue preservation. The current study reports on observed biophysical response changes observed during freezing human dermal fibroblasts in suspension, attached cell, and fibrin tissue-equivalent models. Specifically, intracellular ice formation is shown to increase and dehydration is inferred to increase from suspension to attached systems. Biophysical model parameters fit to these experimental observations reflect the higher kinetics in the attached state. Post-thaw viability values from fast cooling rates were higher for suspension systems, and correlated well with the amount of IIF observed. On the other hand, viability values from slow cooling rates were higher for attached systems, although the degree of dehydration was predicted to be comparable to suspension cells. This disconnect between biophysics and viability predictions at slow rates clearly requires further investigation as it runs counter to our current understanding of dehydration injury in cells. This may suggest a possible protective effect of the attachment state on cell systems.     

Cooling rate dependent biophysical and viability response shift with attachment state in human dermal fibroblast cells

While studies on the freezing of cells in suspension have been carried out extensively, corresponding studies with cells in the attached state and in tissue or tissue-equivalents are less developed. As attachment is a hallmark of the tissue state it is important to understand its impact on biophysics and viability to better apply freezing towards tissue preservation. The current study reports on observed biophysical response changes observed during freezing human dermal fibroblasts in suspension, attached cell, and fibrin tissue-equivalent models. Specifically, intracellular ice formation is shown to increase and dehydration is inferred to increase from suspension to attached systems. Biophysical model parameters fit to these experimental observations reflect the higher kinetics in the attached state. Post-thaw viability values from fast cooling rates were higher for suspension systems, and correlated well with the amount of IIF observed. On the other hand, viability values from slow cooling rates were higher for attached systems, although the degree of dehydration was predicted to be comparable to suspension cells. This disconnect between biophysics and viability predictions at slow rates clearly requires further investigation as it runs counter to our current understanding of dehydration injury in cells. This may suggest a possible protective effect of the attachment state on cell systems.     

Intracellular freezing and survival in the freeze tolerant alpine cockroach Celatoblatta quinquemaculata

The alpine cockroach Celatoblatta quinquemaculata is common at altitudes of around 1500 m on the Rock and Pillar range of Central Otago, New Zealand where it experiences freezing conditions in the winter. The cockroach is freeze tolerant, but only to c. -9 degrees C. The cause of death at temperatures below this is unknown but likely to be due to osmotic damage to cells (shrinkage). This study compared the effect of different ice nucleation temperatures (-2 and -4 degrees C) on the viability of three types of cockroach tissue (midgut, Malpighian tubules and fat body cells) and cooling to three different temperatures (-5, -8, -12 degrees C). Two types of observations were made (i) cryomicroscope observations of ice formation and cell shrinkage (ii) cell integrity (viability) using vital stains. Cell viability decreased with lower treatment temperatures but ice nucleation temperature had no significant effect. Cryomicroscope observations showed that ice spread through tissue faster at -4 than -2 degrees C and that intracellular freezing only occurred when nucleated at -4 degrees C. From temperature records during cooling, it was observed that when freezing occurred, latent heat immediately increased the insect's body temperature close to its melting point (c. -0.3 degrees C). This "rebound" temperature was independent of nucleation temperature. Some tissues were more vulnerable to damage than others. As the gut is thought to be the site of freezing, it is significant that this tissue was the most robust. The ecological importance of the effect of nucleation temperature on survival of whole animals under field conditions is discussed.     

Active control of the nucleation temperature enhances freezing survival of multipotent mesenchymal stromal cells

Cryopreservation is a technique that has been extensively used for storage of multipotent mesenchymal stromal cells (MSCs) in regenerative medicine. Therefore, improving current cryopreservation procedures in terms of increasing cell viability and functionality is important. In this study, we optimized the cryopreservation protocol of MSCs derived from the common marmoset Callithrix jacchus (cj), which can be used as a non-human primate model in various pathological and transplantation studies and have a great potential for regenerative medicine. We have investigated the effect of the active control of the nucleation temperature using induced nucleation at a broad range of temperatures and two different dimethylsulfoxide concentrations (Me₂SO, 5% (v/v) and 10%, (v/v)) to evaluate the overall effect on the viability, metabolic activity and recovery of cells after thawing. Survival rate and metabolic activity displayed an optimum when ice formation was induced at −10 °C. Cryomicroscopy studies indicated differences in ice crystal morphologies as well as differences in intracellular ice formation with different nucleation temperatures. High subzero nucleation temperatures resulted in larger extracellular ice crystals and cellular dehydration, whereas low subzero nucleation temperatures resulted in smaller ice crystals and intracellular ice formation.     

Effects of freezing on membranes and proteins in LNCaP prostate tumor cells

Fourier transform infrared spectroscopy (FTIR) and cryomicroscopy were used to define the process of cellular injury during freezing in LNCaP prostate tumor cells, at the molecular level. Cell pellets were monitored during cooling at 2 °C/min while the ice nucleation temperature was varied between − 3 and − 10 °C. We show that the cells tend to dehydrate precipitously after nucleation unless intracellular ice formation occurs. The predicted incidence of intracellular ice formation rapidly increases at ice nucleation temperatures below − 4 °C and cell survival exhibits an optimum at a nucleation temperature of − 6 °C. The ice nucleation temperature was found to have a great effect on the membrane phase behavior of the cells. The onset of the liquid crystalline to gel phase transition coincided with the ice nucleation temperature. In addition, nucleation at − 3 °C resulted in a much more co-operative phase transition and a concomitantly lower residual conformational disorder of the membranes in the frozen state compared to samples that nucleated at − 10 °C. These observations were explained by the effect of the nucleation temperature on the extent of cellular dehydration and intracellular ice formation. Amide-III band analysis revealed that proteins are relatively stable during freezing and that heat-induced protein denaturation coincides with an abrupt decrease in α-helical structures and a concomitant increase in β-sheet structures starting at an onset temperature of approximately 48 °C.     

Effects of freezing on membranes and proteins in LNCaP prostate tumor cells

Fourier transform infrared spectroscopy (FTIR) and cryomicroscopy were used to define the process of cellular injury during freezing in LNCaP prostate tumor cells, at the molecular level. Cell pellets were monitored during cooling at 2 degrees C/min while the ice nucleation temperature was varied between -3 and -10 degrees C. We show that the cells tend to dehydrate precipitously after nucleation unless intracellular ice formation occurs. The predicted incidence of intracellular ice formation rapidly increases at ice nucleation temperatures below -4 degrees C and cell survival exhibits an optimum at a nucleation temperature of -6 degrees C. The ice nucleation temperature was found to have a great effect on the membrane phase behavior of the cells. The onset of the liquid crystalline to gel phase transition coincided with the ice nucleation temperature. In addition, nucleation at -3 degrees C resulted in a much more co-operative phase transition and a concomitantly lower residual conformational disorder of the membranes in the frozen state compared to samples that nucleated at -10 degrees C. These observations were explained by the effect of the nucleation temperature on the extent of cellular dehydration and intracellular ice formation. Amide-III band analysis revealed that proteins are relatively stable during freezing and that heat-induced protein denaturation coincides with an abrupt decrease in alpha-helical structures and a concomitant increase in beta-sheet structures starting at an onset temperature of approximately 48 degrees C.     

Intercellular ice propagation: experimental evidence for ice growth through membrane pores

Propagation of intracellular ice between cells significantly increases the prevalence of intracellular ice in confluent monolayers and tissues. It has been proposed that gap junctions facilitate ice propagation between cells. This study develops an equation for capillary freezing-point depression to determine the effect of temperature on the equilibrium radius of an ice crystal sufficiently small to grow through gap junctions. Convection cryomicroscopy and video image analysis were used to examine the incidence and pattern of intracellular ice formation (IIF) in the confluent monolayers of cell lines that do (MDCK) and do not (V-79W) form gap junctions. The effect of gap junctions on intracellular ice propagation was strongly temperature-dependent. For cells with gap junctions, IIF occurred in a directed wave-like pattern in 100% of the cells below -3 degrees C. At temperatures above -3 degrees C, there was a marked drop in the incidence of IIF, with isolated individual cells initially freezing randomly throughout the sample. This random pattern of IIF was also observed in the V-79W monolayers and in MDCK monolayers treated to prevent gap junction formation. The significant change in the low temperature behavior of confluent MDCK monolayers at -3 degrees C is likely the result of the inhibition of gap junction-facilitated ice propagation, and supports the theory that gap junctions facilitate ice nucleation between cells.