Temporal differences in the influence of ischemic factors and deformation on the metabolism of engineered skeletal muscle

Debby Gawlitta, Cees W. J. Oomens, Dan L. Bader, Frank P. T. Baaijens, Carlijn V. C. Bouten


Prolonged periods of tissue compression may lead to the development of pressure ulcers, some of which may originate in, for example, skeletal muscle tissue and progress underneath intact skin, representing deep tissue injury. Their etiology is multifactorial and the interaction between individual causal factors and their relative importance remain unknown. The present study addressed the relative contributions of deformation and ischemic factors to altered metabolism and viability. Engineered muscle tissue was prepared as previously detailed (14) and subjected to a combination of factors including 0% oxygen, lactic acid concentrations resulting in pH from 5.3 to 7.4, 34% compression, and low glucose levels. Deformation had an immediate effect on tissue viability {[3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (MTT) assay}, which increased with time. By contrast, hypoxia evoked metabolic responses (glucose and lactate levels) within 24 h, but viability was only reduced after 48 h. In addition, lactic acidification downregulated tissue metabolism up to an acid concentration (∼23 mM) where metabolism was arrested and cell death enhanced. A similar tissue response was observed during glucose deprivation, which, at negligible concentration, resulted in both a cessation of metabolic activity and a reduction in cell viability. The combination of results suggests that in a short-term (<24 h) deformation, extreme acidification and glucose deprivation increased the level of cell death. By contrast, nonextreme acidification and hypoxia influenced tissue metabolism, but not the development of cell death. These data provide more insight into how compression-induced factors can lead to the onset of deep tissue injury.

  • C2C12
  • hypoxia
  • acidification
  • glucose deprivation
  • pressure ulcers

pressure ulcers may develop after prolonged periods of tissue deformation (3, 4, 8). The deformation may lead to reduced blood flow due to the collapse of vessels with the associated ischemia of the supplied tissues. The ulcers may affect several tissue layers, starting at the skin and progressing toward the bone or in reversed order, starting in the deep tissues and growing toward the surface. When the ulcer is located in the subcutaneous layers underneath an intact skin, it is called deep tissue injury (DTI; 11). When DTI proceeds to the skin, the result may be a stage III or IV open pressure ulcer (1). These pressure ulcers may develop in specific patient groups, often involving skeletal muscle tissues. For example, spinal cord-injured individuals exhibit such inherent risk factors as immobility, desensitivity, and altered (muscle) tissue properties (7, 26). As it is known that skeletal muscle is particularly sensitive to compressive loading (8, 27), it will be the focus of the present study.

Ischemia, reperfusion, lymphatic blockage, and cellular deformation are each considered to play some role in pressure ulcer development (8, 10, 19, 23, 24). Most probably, the onset of damage development is multifactorial in nature and requires the contribution of several of these factors (4, 14, 27). The various factors may also contribute to a different degree to the development of damage associated with different tissue layers, such as skin, fat, or muscle.

Despite its potential multifactorial origin, the most established hypothesis for tissue damage development so far is mechanically induced capillary occlusion, resulting in tissue ischemia and thus hypoxia (oxygen deprivation). However, previous studies on skeletal muscle might suggest a lesser role for oxygen deprivation in this tissue than has been previously assumed. Indeed, Stekelenburg (27) showed that highly deformed regions of rat muscle correlated with areas of tissue damage as opposed to the ischemic regions resulting from tissue deformation. The ischemic period in this study was 2 h, which is below the threshold of 4–5 h for the onset of skeletal muscle necrosis (21). A similar result was reported in vitro, in engineered skeletal muscle constructs that developed tissue damage due to compression rather than oxygen deprivation (14). During a 22-h compression period, the tissue viability was affected by imposed global deformations of 20 and 40%, but not by hypoxia. This may be explained by the oxygen conformance behavior of the cells, in that they consume less oxygen under oxygen deprivation, which may extend to 22 h. In addition, they may adjust by switching to anaerobic metabolism. However, at some point, the cells must respond to oxygen deprivation. Therefore, experiments for longer periods of time are required. Also, the results may imply that ischemia with its associated hypoxia played a less important role in damage development than tissue deformation. However, hypoxia cannot be considered to be the only adverse environmental event occurring during ischemia. In addition, other factors, such as acidosis, hyperkalemia, hypercapnia, lactate build-up, and glucose depletion will take place in ischemic tissues (13, 22). Apart from these phenomena, calcium overload and accumulation of free radicals play important roles in damage development in ischemic skeletal muscle (15).

The objective of the present study was to provide an improved understanding of the relative contributions of the events occurring during tissue compression, i.e., deformation and ischemia, to the development of muscle damage. Effects of deformation, tissue hypoxia, glucose depletion, and acidification due to lactate presence were determined in an in vitro muscle model. It was hypothesized that hypoxia would stimulate the tissue to switch to anaerobic metabolism. This implies that glucose is degraded into lactate, generating energy without the necessity for oxygen consumption. Thus the employment of anaerobic metabolism will create more lactate and thus a more acidic environment, which may compromise cell activity and tissue viability. Furthermore, compression of the cells within the tissue is thought to affect the cells on a shorter term than ischemic factors.

The separate and combined effects of the compression-induced factors on metabolism and viability will be studied in an engineered skeletal muscle model that was previously applied in pressure ulcer-related research (14). The engineered muscle tissues were assembled according to protocols modified from those of Dennis and coworkers (9) and Vandenburgh et al. (32, 33) to produce tissue with great ease of fabrication and high reproducibility.


Cell and Tissue Culture

C2C12 murine myoblasts (ECACC, Salisbury, UK) were kept below 80% confluency in growth medium, containing high glucose DMEM (GIBCO, Breda, The Netherlands), 15% FBS, 2% HEPES, 1% nonessential amino acids and, 0.5% gentamicin (all purchased from Biochrom, Berlin, Germany). Cells were harvested between passages 11 and 18 and resuspended in a gel mixture as described elsewhere (14). Briefly, ∼4 million cells were suspended in a mixture (total 1 ml) of 0.39 ml GM, 0.5 ml collagen I (3.2 mg/ml, BD Biosciences, Alphen a/d Rijn, The Netherlands), 27 μl NaOH (0.25 M, Sigma, Zwijndrecht, The Netherlands), and 83 μl Matrigel (BD Biosciences) and molded into shape between two Velcro anchoring points in six-well plates. The six-well plates were precoated with 1.5 ml SYLGARD 184 elastomere (Mavom b.v., Alphen a/d Rijn, The Netherlands). To this surface, house-shaped Velcro anchoring points (5×7 mm) were glued ∼10 mm apart with their “rooftops” facing each other. The prepared dishes were filled with PBS and sterilized by exposure to ultraviolet light for ∼90 min before the cells were seeded.

After molding the gel/cell mixture between the Velcro anchoring points, the dishes were placed into an incubator. Subsequently, after 1–2 h, the constructs were covered with growth medium. After 24 h, the growth medium was replaced by differentiation medium, which had the same formulation as growth medium, except that 0.4% Ultroser G (BioSepra, Cergy-Saint-Christophe, France) was used instead of FBS. Differentiation medium was refreshed every other day. The engineered muscle constructs were allowed to mature for an additional 7–9 days before experiments were performed.

Experimental Conditions

Several experimental conditions were imposed on the constructs for a period of 5 days, as indicated in Table 1. It can be seen that in some cases two or more factors were imposed simultaneously, for example, hypoxia and deformation. Under all experimental conditions, temperature was maintained at 37°C and CO2 level at 5%.

View this table:
Table 1.

Overview of experimental conditions imposed on engineered muscle for 5 days

Sampling of the medium was performed at time point 0 and repeated every 24 h for 5 consecutive days without refreshing the medium in all groups. Every medium sampling of 100 μl was withdrawn from a total of 3 ml. The concentrations of glucose, lactate, and LDH that were determined from this sampling were corrected for the loss of solutes and volume, according to formula 1: Math Math(1) with cp indicating concentrations of glucose, lactate, or LDH at dayp” (representing 1–5, respectively). The corrected concentrations are represented by c′p. The formulas are based on c = m/V, with mass, m, and volume, V.

Tissue deformation.

For experimental conditions involving unconfined tissue compression, the mean undeformed height of the samples was determined as 530 ± 69 μm (n = 16). A tissue compression of 34 ± 8% was achieved by placing rectangular stainless steel weights on top of the constructs on day 0, with spacers of 350 μm at their four corners (Fig. 1 A). These weights remained in place until day 5.

Fig. 1.

Photographs of the experimental set-up. A: one of the stainless steel loads is shown with the arrows indicating 350-μm spacers on one side of the load. B: top view of the air-tight box is shown with four 6-well plates inside. One of the rubber sheets for medium sampling of 2 wells is pointed out by the arrow.


Samples were deprived of oxygen by incubation in an air-tight box (Fig. 1B), which was continuously flushed with a 95% N2 and 5% CO2 gas mixture over the 5-day period. Evaporation was prevented by overlaying the medium with mineral oil (Sigma, Zwijndrecht, The Netherlands), which was previously shown not to hinder oxygen diffusion (13). This layer of oil was also placed on medium of the normoxic (20% oxygen) samples that were kept in the incubator. In contrast, the air-tight box was heated in a warm water bath to 37°C. Sampling of the medium was performed with negligible disturbance of the hypoxic environment inside the box. Rubber sheets on top of the box were pierced with a needle but retained their air-tight property after repetitive puncturing (Fig. 1B).

Effects of the air-tight box in a warm water bath on tissue metabolism and viability were determined in a separate experiment in which the box was continuously flushed with an atmosphere of 95% air and 5% CO2. There were no differences in either the tissue viability or the tissue metabolism over a 5-day period between culture samples in the box compared with those in an incubator (D. Gawlitta, unpublished observations).


The influence of acidic pH values on sample viability was established by the addition of l-(+)-lactic acid (L1875, Sigma) to the culture medium to achieve final pH values of 5.3, 5.8, 6.1, 6.4, 6.7, 7.4 (control) at day 0. A broad range of pH values was chosen, as the values reported in literature to be the lower threshold compatible with living cells also greatly varied [6 (pH 7.1); 29 (pH 6.5); 37; 38 (pH 7.0)]. The effect of the acidic media was evaluated in both the tissue constructs and in confluent undifferentiated monolayers, cultured in 12-well dishes. Based on effects of these pH values in 24 h (assessed by an MTT assay, as described in the next section), a pH value was chosen for the 5-day experiment. The medium pH was not adjusted after time point 0.

Glucose concentrations.

To assess the influence of glucose deprivation on the engineered muscle samples, two different concentrations of glucose were chosen. Tissue survival was assessed in high glucose (4.5 g/l) and low glucose (1 g/l) differentiation media without refreshment.

Analyses of Metabolism and Viability

To assess the metabolic activity of the cells, glucose, lactate, and pH were measured every 24 h from the extracted medium samples. In addition, an end point [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] reduction (MTT) assay was performed. The assay is based on the ability of a mitochondrial dehydrogenase enzyme present in viable cells to cleave the yellow MTT to form purple formazan crystals, which accumulate within healthy cells. Permeabilization of the cells and subsequent solubilization of the crystals results in a purple solution, from which absorption can be determined. It should be recognized that the amount of formazan produced could indicate either differences in cell number, caused by growth or death, differences in the metabolic activity of the cells per se, or a combination of both. Therefore, the viability of cultures was additionally determined from the release of LDH (lactate dehydrogenase) in the extracted medium.

Glucose assay.

Medium samples were diluted 40× in water. Assay reagent was prepared according to the manufacturer's protocol. Briefly, 2% vol/vol o-dianisidine (D2679, Sigma) was mixed with 98% vol/vol glucose oxidase-peroxidase reagent (G3660, Sigma). One part of diluted sample was mixed with two parts of assay reagent in 96-well plates. After 30 min of incubation, the reaction was stopped by the addition of 6 M sulfuric acid. Absorbance was measured at 540 nm against a background signal recorded at 650 nm on a plate reader. Calibration with a glucose standard was performed for each plate.

Lactate assay.

Medium samples were diluted 20× and one part was mixed with 10 parts of lactate reagent (TB 735–10, Kordia Life Sciences, Leiden, The Netherlands), prepared according to the manufacturer's recommendations. After 10 min, absorbance was read on a plate reader at 540 nm against background at 650 nm. Calibration curves were obtained using a range of l-(+)-lactic acid (L1875, Sigma) dilutions for each measurement.


Measurement of pH was performed by the application of pH indicator paper (Merck, Haarlem, The Netherlands) ranging pH 6.4–8.0 and pH 5.4–7.0 with 0.2–0.3 unit intervals. Paper readings were confirmed with a glass tip electrode (SP10T, Consort, Turnhout, Belgium) and pH meter (P501, Consort).

MTT assay.

Metabolic activity of cells was detected by measuring the reduction of MTT to the purple formazan crystal. After incubation in MTT-medium mixture for 30 min, the samples were rinsed with PBS and the wet weight of 5-mm pieces that were cut from the tissue between the Velcro was determined. The 5-mm width of the excised tissue was based on the dimensions of the loaded tissue. Formazan crystal was extracted and dissolved in a 90% DMSO-10% Triton-X mixture. Absorbance was read at 570 nm.

LDH assay.

LDH release of the cells was assessed by an in vitro toxicology assay kit, which was LDH-based (TOX7, Sigma). Samples were diluted 5× in water to a total volume of 30 μl prior to the addition of the reagent. The reaction was terminated at 20 min with 10 μl 1 N hydrochloric acid. Absorbance was read on a plate reader at 490 nm and corrected for background at 650 nm.

Statistical Analysis

All data are presented as means ± SE. The engineered samples were considered to be very reproducible in that they all contained an equal amount of cells. Therefore, no normalization was carried out, except for the MTT data, which were normalized to the samples wet weight. Differences over time and between groups were assessed using an ANOVA performed with SPSS software, version 12.0.1. A Bonferroni post hoc test was then applied and statistical significance was prescribed at the 5% level (*P < 0.05; **P < 0.01; ***P < 0.001).



The effects of medium pH on the viability of both myoblast monolayers and differentiated muscle tissue were examined after 24 h. Changes in the metabolic activity of the monolayer culture, as assessed by MTT absorption, are illustrated in Fig. 2A. There was a clear decrease in metabolic activity with decreasing values down to pH 5.8, although no subsequent decrease at pH 5.3 was observed. By contrast, the metabolism of muscle tissues revealed no differences between the pH range 5.8–7.4, but a significant decrease at pH 5.3 (Fig. 2B). On the basis of this finding, a lowest pH of 5.3 was applied in additional experiments.

Fig. 2.

Effect of pH on cell viability assessed by MTT absorption for monolayer (A) and muscle tissue cultures (B) after 24 h. Significant differences are indicated by the bars in A, whereas in B, the difference against all other groups is shown (***P < 0.001; **P < 0.01; *P < 0.05).

Effects of Glucose Deprivation

The effect of glucose deprivation during a 5-day period in the absence of medium refreshment was significant from 24 h onward (Fig. 3). With reference to glucose availability (Fig. 3A), it was evident that by day 1 there was no additional glucose available to the cells in the low glucose medium. Accordingly, metabolism was limited in the low glucose medium, with lactate accumulation reaching a maximum at day 1 and thereafter remaining constant (Fig. 3B). The control group, associated with a culture medium containing 4.5 g/l glucose, revealed an increase in lactate concentrations up to day 3, thereafter remaining constant. As a result of limited glucose-derived energy production, loss of cellular integrity was consistently higher in the low glucose group compared with the control group (Fig. 3C). It should be noted, however, that the latter group still revealed a monotonic increase in LDH release over the 5-day culture period. The differences between the control and low glucose group were also evident from pH profiles (Fig. 3D). The pH of the control medium decreased with time until day 3, thereafter remaining fairly constant. By contrast, the pH of the low glucose medium did not change over the culture period.

Fig. 3.

Effects of glucose deprivation on temporal profiles for glucose utilization (A), lactate production (B), LDH release (C), and medium pH (D), for several experimental conditions control (a) and low glucose (b; Table 1). Significant differences of group b compared with control are indicated above the graphs (***P < 0.001; *P < 0.05). Other intergroup differences were not statistically significant.

Effects of Lactic Acidification

Three separate experimental conditions with initial pH values of 6.5 (condition f), 6.0 (j), and 5.3 (k) were compared with the control group at pH 7.4 (a). The results for glucose concentration, as presented in Fig. 4A, reveal that the groups with an initial pH of 6.5 and 6.0 used increasing amounts of glucose until day 3, although utilization was always less than in the corresponding control groups. By contrast, the group starting with the most acidic value (pH 5.3) used very little glucose throughout the culture period. The corresponding lactate values were clearly higher in the three experimental conditions at the start of the culture period (Fig. 4B). However, by day 2 there was little difference in the concentration for two experiment groups f and j and the control group (a). In a similar manner to its glucose utilization behavior, the experimental group (k) did not produce a change in the lactate concentration over the 5-day culture period. The LDH release profiles for all four groups increased monotonically with culture period (Fig. 4C). However, the experimental group (k) with the lowest initial pH value showed significantly increased LDH release compared with the control group from the first day onward.

Fig. 4.

Effects of lactic acidification on temporal profiles for glucose utilization (A), lactate production (B), LDH release (C), and medium pH (D), for several experimental conditions [control (a), pH 6.5 (f), pH 6.0 (j), pH 5.3 (k)]. Significant differences between the groups f, j, k and the control group are indicated above the graphs (***P < 0.001; *P < 0.05).

Effects of Deformation and Hypoxia

The effects of deformation (d) or hypoxia (c) per se or both simultaneously (e) were assessed for the 5-day culture period in a separate set of experiments (Fig. 5). The deformed tissue samples did not significantly change their glucose utilization (the rate of decline of glucose concentrations with time; Fig. 5A), lactate production (the rate of increase in lactate concentrations with time; Fig. 5B), or LDH release (Fig. 5C) compared with control tissues. By contrast, groups involving hypoxia in the absence or presence of deformation had consumed significantly more glucose (derived from the steeper slopes of the curves) than the control group on days 1 and 2. There was a general increase in lactate concentrations with culture time for the three experimental groups. However, the hypoxic group (c) was the only one that demonstrated a statistically significant increase in lactate concentrations compared with control values at corresponding time points. The LDH release for both hypoxic groups was significantly increased from day 2 until day 5 compared with the other two groups as evident from Fig. 5C. Changes in pH were similar in the deformed and the control groups, whereas the pH of the media from both hypoxic groups was more acidic in nature, although the differences were not statistically significant. The hypoxic group exhibited the lowest pH from day 3 onward (Fig. 5D). It is interesting to note that the results from the control group in this set of experiments differed from those in other experiments.

Fig. 5.

Effects of deformation and hypoxia on temporal profiles for glucose utilization (A), lactate production (B), LDH release (C), and medium pH (D), for several experimental conditions [control (a), deformed (d), hypoxic (c), “hypoxic, deformed” (e)]. Significant differences between groups d, c, e, and the control are indicated above the graphs (***P < 0.001; **P < 0.01; *P < 0.05).

Effects of Deformation and/or Hypoxia Combined with Acidification

The effects of lactic acid, deformation, and hypoxia imposed on tissue either separately or in different combinations were analyzed. Figure 6A reveals that glucose utilization for all four experimental groups (f, g, h, i) followed a similar trend. Indeed, glucose utilization for each group was significantly impeded compared with the control samples (P < 0.01 for days 1–5). These data suggest that the acidification to a pH value of 6.5 was the dominant influence on glucose utilization. With respect to lactate production, the levels for both the experimental groups and the control group were similar between days 2 and 4 (Fig. 6B). However, after 5 days in culture, significantly elevated levels of lactate were detected in the presence of both hypoxia and tissue deformation (g and i, Fig. 6B).

Fig. 6.

Effects of deformation and hypoxia during acidification on temporal profiles for glucose utilization (A), lactate production (B), LDH release (C), and medium pH (D) for several experimental conditions [control (a), pH 6.5 (f), pH 6.5, deformed (h), pH 6.5, hypoxic (g), pH 6.5, hypoxic, deformed (i)]. Significant differences between groups f, h, g, i, and the control are indicated above the graphs (***P < 0.001; **P < 0.01).

Despite the metabolic effects observed, LDH release was unaffected by all experimental conditions compared with the control group. The temporal changes in pH largely reflected the levels of lactate concentrations. Thus a minimum pH value of 6.2 (Fig. 6D) is associated with lactate values of ∼25 mM (Fig. 6B).

Tissue Viability After 5 Days (MTT)

Tissue viability of all experimental groups was assessed after the 5-day culturing period by an MTT assay (Fig. 7, A–D). Surprisingly, the low glucose group showed the highest overall metabolic activity, although this group (b) exhibited the maximum LDH release as well (Fig. 7A). At day 5, significant effects of the addition of lactic acid were observed (Fig. 7C) that were not found after 1 day of culturing (as shown in a previous section in Fig. 2B). Tissue viability was not only affected by time, but also gradually decreased with increasing acidity, such that at a pH of 5.3 (k) there was a zero value for the MTT absorption.

Fig. 7.

A–D: MTT absorptions per milligram tissue are shown as assessed at day 5. A: effects on MTT conversion for “low glucose” (b); B: for pH 6.5 (f), pH 6.5, deformed (h), pH 6.5, hypoxic (g), and pH 6.5, hypoxic, deformed (i); C: for pH 6.5 (f), pH 6.0 (j), and pH 5.3 (k); D: for deformed (d), hypoxic (c), and hypoxic, deformed (e) are depicted compared with the control group (a). Significant differences from control values are indicated by an asterisk; or between groups by bars and asterisks (*P < 0.05; **P < 0.01; ***P < 0.001). Images above the graphs show tissue MTT staining; in the lower right image a rectangle shows the tissue area from which MTT absorption values were determined. E: the images show the MTT staining pattern in tissue after 24 h for a control (a) and a deformed (d) sample.

From Fig. 7D it was clear that hypoxia in the absence (c) or the presence of deformation (e) resulted in a dramatic reduction in tissue viability, whereas the effect of deformation alone (d) was significant but not as dramatic as that involving hypoxia. The effect of deformation on tissue viability on day 5 as assessed by MTT was comparable to the one found after 24 h of compression in pilot experiments (Fig. 7E).

The effect of acidification per se (f) or combined with hypoxia (g) and deformation (h) on MTT conversion revealed that by day 5, only activity in the acidic group (f) was present, although ∼50% of that observed in the control group (Fig. 7B). If, however, hypoxia (g) or deformation (h) or both (i) were imposed on top of lactic acid, metabolic (MTT) activity was reduced to 0. Comparing the effects by the separate factors of deformation (d), hypoxia (c), or acidification (f) on MTT absorption the reduction in metabolic activity was most dominant in the hypoxic group. However, combining the other two factors, deformation and acidification (h), resulted in a stronger effect than one of them per se.


Deep tissue injury (DTI) has gained increasing interest in recent years (1, 11). DTI will develop under intact skin in the subcutaneous layers as a result of compression. It may progressively deteriorate toward the skin surface, resulting in a type III or IV ulcer. However, little is known about its etiology and currently these ulcers under intact skin are classified as “unstageable” pressure ulcers (1). Understanding development of muscle tissue damage may contribute in understanding the etiology of the onset of DTI and its rapid progression.

The present study addressed some factors that are thought to be involved in the onset of muscle tissue damage in pressure ulcer development. The influence of ischemic factors and deformation were assessed in an engineered muscle model. The advantage of the ex vivo situation is the ability to vary several etiological factors independently of each other and to monitor the tissue response in real time.

Comparison of engineered muscle culturing in high vs. low glucose medium for 5 days showed that within 1 day after glucose depletion, lactate production and acidification of the medium were arrested and a strong increase in LDH release followed (Fig. 3). In apparent contrast to these findings, which suggest a severe cell death response after glucose depletion, metabolic activity as assessed by MTT conversion exceeded the response of all other groups (Fig. 7). Nonetheless, the metabolic activities of the control and low glucose groups, assessed on day 5, had decreased significantly compared with that observed at day 0. The low metabolic response in the control group may be attributed to the acidification of the medium with time. The still viable cells after prolonged glucose starvation may in the mean time have used other fuel resources such as glycogen stores, fatty acids, and possibly amino acids. Still, in conclusion, glucose deprivation resulted in a rapid increase of cell death.

Several concentrations of lactic acid were added to produce a range of pH values measured at room temperature and atmospheric CO2 levels. It should be noted, however, that the initial pH values do not always correspond to the actually preset pH (Fig. 4D). This is due to the buffering capacity of the medium under 5% CO2 and the temperature of 37°C.

After addition of acidified medium to monolayer or tissue cultures, the influence on tissue viability was more pronounced in the 2D cultures than in 3D. This may be caused by the differential status of the cells, i.e., myoblasts in 2D as opposed to myotubes in 3D. It may additionally be explained by a protective effect of the extracellular gel environment associated with the 3D cultures.

From the metabolic data obtained during the 5-day culturing period of tissue in acidic media it was clear that increasing amounts of initially introduced lactic acid resulted in a decrease of glucose utilization of the cells and thus the equilibrium levels were at a higher glucose concentration (Fig. 4A). A pH of 5.3 appeared to have impeded tissue metabolism because no glucose was consumed, nor was additional lactate produced. This may suggest that a threshold value for lactic acidity between pH 6.0 (17 mM) and pH 5.3 (25 mM) exists for a transition from diminishing glucose metabolism to a metabolic arrest within a 5-day period. The presence of different initial concentrations of lactic acid, however, did not result in different lactate or plateau levels of pH, but rather caused convergence of the levels to ∼23 mM and pH 6.2, respectively. These values may be considered threshold values for tissue death. The MTT data obtained at day 5 (Fig. 7C) showed that pH values of 6.5, 6.0, and 5.3 had affected viability, whereas after 1 day, only a pH of 5.3 had decreased viability (Fig. 2B). This may be caused by either minimally increased cell death as assessed by LDH release, which accumulates to a measurable effect after 5 days, or by its effect on cellular metabolism as assessed by different amounts of glucose utilization. The latter has previously been described as an “energy-saving” mechanism (25, 34). Nevertheless, presence of elevated levels of lactic acid decreased glucose metabolism with time, but only had effect on cell death if the acidic threshold was exceeded, which was the case for the most acidic “pH 5.3” group.

The plateau level of extracellular lactate concentration is within the range of 20–30 mM that was applied to simulate lactic acidosis (17) and reported to build up during anaerobic exercise (31). The apparent threshold value for pH of 6.2 was comparable to the values shown to promote death in various cell types (29, 36). The effect of pH lowering in their monolayer cultures was observed within 24 h in concert with our observations in the 2D cultures.

The effects of deformation and hypoxia on cell death have previously been described (14). To review, deformations of 20 and 40% led to increased cell death, whereas no effects of hypoxia on cell death were found within 22 h. In the present study, the experimental period was extended and the resulting changes in metabolism were determined. Hypoxia and combined hypoxia and deformation led to an increased glucose utilization, increased LDH release, and decreased pH values (Fig. 5). However, only the hypoxic tissues in the absence of deformation had increased their lactate formation, which is indicative of a transition to anaerobic metabolism. In the present study, viability (MTT) was measured on the 5th day, when no significant activity remained in both hypoxic groups to possibly explain this difference in metabolism. Nonetheless, although no changes in glucose, lactate, or LDH concentrations were observed for the deformed vs. the control tissue, a significant decrease in metabolic activity (MTT) of the deformed tissue was found. The effect of compression on metabolic activity of the tissue (MTT) was also apparent within 24 h (Fig. 7E). The compression was expected to significantly affect LDH release compared with the control. This was not the case, which may be explained by the relatively large size of the LDH molecule that, in addition, may have been hindered by the compressed collagen network in the deformed portion of the tissue.

The following may explain the effects on tissue viability observed in the deformed tissues. Glucose, lactate, and LDH levels were assessed from medium samples rendering information on the complete construct. MTT data, however, were obtained more locally from the excised central (and thus deformed) portion of the sample (Fig. 7D, d). The local effects of deformation may therefore be “lost” compared with the metabolism of the remaining tissue if determined from the medium. On the contrary, hypoxia affected all of the tissue and resulted in significant changes of medium composition. Combined hypoxia and deformation did lead to a more acidic pH, but not more lactic acid. This may be caused by formation of other acids than lactic acid by these cells or by the death of the cells.

Concisely, deformation affected tissue viability at the short term and to a lesser extent, whereas hypoxia initiated an anaerobic adaptive response before killing the cells. This response of skeletal muscle is its oxygen conformance behavior (2, 5). Due to this adaptive mechanism skeletal muscle tissue is considered one of the most hypoxia tolerant tissues (5, 18, 34). The effect of deformation and hypoxia combined were dominated by the presence of hypoxia as established by the applied methods.

Finally, the factors acidification, deformation, and hypoxia were combined to obtain information on possible interaction. Glucose use appeared to be predominantly influenced by the presence of lactic acid compared with the control. In contrast, in lactate production, a significant trend for elevated lactate concentration (and more acidic pH) in both hypoxic acidic groups was found compared with the nonhypoxic acidic groups. Although effects on metabolism were noticed, no difference in LDH release among the groups or compared with control was detected. The MTT metabolic assay, however, did reveal a significant difference in metabolic activity of the acidic group compared with the acidic, deformed, and hypoxic groups. Taken together, these data suggest that medium acidity influenced tissue metabolism, but hypoxia and deformation aggravated these effects.

The effect of ischemic hypoxia has been studied extensively in cardiac myocytes (16, 20, 35, 36). The basic finding was that hypoxia alone did not cause apoptosis but that acidosis, reoxygenation, and reperfusion did. Their hypoxic cultures (without medium refreshment) became apoptotic after 2–3 days, at which time LDH release was observed in our cultures. Moreover, they confirmed that addition of lactic acid to pH 6.8 caused apoptosis independently of hypoxia in their myocytes.

Stekelenburg and colleagues (27, 28) performed indentation experiments on rat hindlimb muscle. Results showed that tissue damage areas correlated with regions of maximum shear strain as opposed to tissue areas that were ischemic alone. This was based on observations during and after load application of 2 h. The chosen time period was well below the 4–5 h of ischemia known to cause ischemic tissue damage (21). In the present study, simulated ischemia and deformation were imposed for extended periods of time. Acidosis, hypoxia, and deformation all contributed to the development of damage in engineered muscle tissue. Deformation decreased viability (14) after the onset of compression within 24 h. The effect of hypoxia was apparent after 24 h as the cellular metabolism was affected, but only resulted in increased cell death after 48 h. Although the effect of imposed lactic acidosis appeared to be instantaneous, the time to produce damaging amounts by hypoxic cells was considerably more extended. These findings support the hypothesis presented by Stekelenburg (27) that initially cell damage is dominated by deformation, but it is aggravated by ischemia at later time points.

In the future, both hyperkalemia and hypercapnia should be incorporated in in vitro simulations of ischemic conditions. In addition, the effects of gradual reperfusion and hindered lymphatic transport on tissue status will need to be addressed (12, 30). Nonetheless, even in the absence of these factors, the multi-factorial base of pressure ulcer development in deep tissues from compression and ischemic factors has been confirmed.

In conclusion, within muscle tissues, both deformation and ischemia can play a role in damaging tissue. However, deformation affected tissue viability over a short time period, whereas the tissue model was better equipped to withstand ischemic factors and thus could survive ischemia alone for longer periods. It should be noted that the applied deformation was 34% and that higher deformations may occur in vivo, resulting in increased amounts of cell death. The ischemic factors tended to initiate cellular survival processes before cell death. During hypoxia, anaerobic metabolism was adopted, resulting in lactic acid accumulation and acidification of the medium. To maintain anaerobic metabolism, the tissue consumed more glucose. Thus this process may continue as long as sufficient glucose is available. In addition, lactic acidification should not exceed threshold values. Moreover, metabolism was downregulated by oxygen conformance behavior together with an energy-saving mechanism induced by acidification. When these factors were all combined and imposed for 5 days, the deformation had damaged the tissue initially. Subsequently, the hypoxia-induced elevated lactic acid production eventually exceeded the acid threshold, as there was ample glucose present to continue metabolism. Thus, as long as no deformation or ischemic thresholds are exceeded, the tissue may survive compression.


This research was supported by the Dutch Technology Foundation, applied science division of Nederlandse Organisatie voor Wetenschappelyle Onderzoeh, and the Technology Program of the Ministry of Economic Affairs.


The authors thank Rob van den Berg for his contribution to the design and manufacturing of the experimental system.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract