Report

Oxford, UK
International
IJD
Blackwell
0011-9059
45 Publishing,
Publishing
Journal Ltd,
of
Ltd.
Dermatology
2007

Physiological effects of ultrasound mist on fibroblasts

Physiological
Lai
Report
and Pittelkow
effects of ultrasound mist on fibroblasts

Jengyu Lai, DPM, and Mark R. Pittelkow, MD

From the Department of Dermatology, Abstract

Mayo Clinic College of Medicine, Rochester, Background Chronic wounds present an increasing challenge in healthcare and consume
Minnesota a substantial portion of healthcare cost. Although new treatments have been developed,
treatment success has not been improved greatly. Ultrasound has long been employed in
Correspondence
Mark R. Pittelkow, MD medicine. Its unique ability to deliver energy makes it an ideal candidate as a wound care
Department of Dermatology modality. We proposed that ultrasound would differentially affect intracellular signaling pathways
Mayo Clinic and, with the ability to assess this effect using a noncontact form of ultrasound, were provided
200 First Street South-west with a means to test this proposal.
Rochester, MN 55905
Methods The cellular morphology, mitogenic activities, expression of keratinocyte growth
E-mail: [email protected]
factor (KGF) and transforming growth factor β-1 (TGF-β1), and activation of extracellular
regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) signaling pathways of dermal
fibroblasts were studied after ultrasound treatment. Untreated and scrape-wounded fibroblasts
were utilized as controls.
Results There was no difference in morphology observed, except for vacuolization in
ultrasound-treated fibroblasts. Mitogenic activities were similar between ultrasound-treated and
scrape-wounded fibroblasts. Ultrasound-treated fibroblasts exhibited a much earlier increase in
KGF expression, ERK activation, and JNK activation. The ERK/JNK ratio was increased
markedly in ultrasound-treated fibroblasts.
Conclusion We conclude that ultrasound induces cellular responses that may be beneficial to
wound healing.

Ultrasound delivers energy through a pressure field

Introduction
generated by the transducer and causes the molecules of its
An estimated 2% of the American population suffers from transmission medium to oscillate or vibrate. Ultrasound is
chronic wounds, resulting from diabetes mellitus, venous in- used both therapeutically and diagnostically in medicine.
sufficiency, and excessive pressure. According to the American Ultrasound has been shown to accelerate wound healing, and
Diabetes Association and other studies, diabetes mellitus several mechanisms have been suggested.12,13 Ultrasound
is the sixth leading cause of death, and the medical cost for activates inflammatory cells, resulting in wound debridement
diabetic care ranges from $98 to $113 billion per year.1–7 It is and the production of chemical mediators, which activate
estimated that 15% of patients with diabetes have foot ulcers, fibroblasts.14,15 This activation leads to the earlier accumula-
and that one in 150 patients with diabetes has an amputation. tion of endothelial cells in tissues and the promotion of colla-
In addition, more than 50% of nontraumatic lower limb gen synthesis via the stimulation of calcium influx, alteration
amputations, nearly 82,000 new amputations, are in the dia- of membrane permeability, and enhancement of fibroblast
betic population each year. Approximately 84% of amputa- proliferation.14–18 In addition, collagen deposited after ultra-
tions begin as a simple ulcer. More significantly, 30 – 50% of sound treatment is stronger and better organized.15,19–24 Not
amputation patients undergo further amputation of the surprisingly, tissue treated with ultrasound progresses
remaining lower extremity within 5 years, with a 50% mor- through the phases of healing more rapidly.15
tality rate.6,8 It is estimated that 1–1.2% of the population in Chemical mediators produced after ultrasound treatment
the USA suffers from venous stasis ulcers.9 Both forms of include tissue growth factors, which, in turn, mediate the
chronic ulcer compromise a patient’s ability to ambulate, activation of intracellular signaling pathways.25–29 In addition,
cause pain and mental depression, and cost billions of dollars mechanical stimulation of cellular membranes occurs during
in healthcare. New treatment modalities are constantly being the process of energy delivery. At least part of the signal
developed and refined.10,11 transduction pathway that mediates responses to forces is 587

© 2007 The International Society of Dermatology International Journal of Dermatology 2007, 46, 587– 593
588 Report Physiological effects of ultrasound mist on fibroblasts Lai and Pittelkow

attributable to the adapter protein, Shc. Transduction of the fixed and examined by the Electron Microscopy Core Facility
Shc-mediated signal is coupled to the activation of extracellu- (Mayo Clinic). Untreated cells served as controls.
lar signal-regulated kinases (ERKs) and c-Jun N-terminal For SEM, tissue was fixed in Trump’s fixative (1% glutaraldehyde
kinase (JNK), which, in turn, regulate downstream transcrip- and 4% formaldehyde in 0.1 m phosphate buffer, pH 7.2).38 Tissue
tion factors and nuclear events linked to specific cell was then rinsed for 30 min in two changes of 0.1 m phosphate
responses.30,31 We have shown that the epidermal growth buffer, pH 7.2. Some of the samples were postfixed for 1 h in
factor (EGF) ligand–receptor system is the primary receptor phosphate-buffered 1% OsO4. This step was omitted in a third
tyrosine kinase signaling pathway in keratinocytes that regu- sample. After rinsing in two changes of distilled water for 30 min,
lates epidermal keratinocyte proliferation, differentiation, tissue was dehydrated in progressive concentrations of ethanol to
and cell survival following oxidative stress.32–36 Moreover, we 100%, and either critical point dried or placed into Peldri™. All
have recently demonstrated that the Shc adapter protein com- samples were mounted on aluminum stubs and sputter coated
plex is specifically phosphorylated in response to cell survival with gold/palladium. Images were captured on a Hitachi S4700
signals triggered following EGF receptor activation by scanning electron microscope operating at 5 kV.
ligand.32–34 In addition to the ERK and JNK signaling pathways, For TEM, tissue was fixed in Trump’s fixative, and then rinsed for
which have been shown to be functional cascades in keratino- 30 min in three changes of 0.1 m phosphate buffer, pH 7.2,
cytes, the other main family of mitogen-activated protein followed by a 1-h postfix in phosphate-buffered 1% OsO4. After
kinases (MAPKs) present in keratinocytes is the p38 kinase rinsing in three changes of distilled water for 30 min, the tissue was
members. We have demonstrated previously that the ERK, en bloc stained with 2% uranyl acetate for 30 min at 60 °C. After
JNK, and p38 kinase pathways are expressed, and that the en bloc staining, the tissue was rinsed in three changes of distilled
activation of these pathways is coordinately regulated by water, dehydrated in progressive concentrations of ethanol and
keratinocytes in response to growth factor-mediated or 100% propylene oxide, and embedded in Spurr’s resin.39
stress-induced stimuli.32–35,37 We hypothesized that ultra- Thin (90-nm) sections were cut on a Reichert Ultracut E
sound treatment would induce the activation of both ERK ultramicrotome, placed on 200-mesh copper grids, and stained
and JNK. As the balance of ERK and JNK may predict the with lead citrate. Micrographs were taken on a JEOL 1200 EXII
fate of the cell, we examined the ERK/JNK ratios following operating at 60 kV.
ultrasound treatment of cell cultures. Two hours and 1, 2, 3, and 4 days after treatment, tritiated
thymidine was added to the cells to assess DNA synthesis.
Nonultrasound-treated and nonscrape-wounded fibroblast
Materials and Methods
cultures were used as controls. Two hours before the end of the
Normal dermal fibroblasts were isolated from neonatal foreskin time period, 1 µCi/10 µL of tritiated thymidine per milliliter of
and maintained in Dulbecco’s modified Eagle’s medium (DMEM) medium was added to the cell culture plate. Cells were incubated
containing 10% fetal bovine serum. Cells were seeded at 70 – 80% for 2 h at 37 °C. The radioactive medium was aspirated and the
confluence on six-well plates for protein studies, 24-well plates for cells were washed with cold 10% trichloroacetic acid and rinsed
mitogenic studies, 100-mm2 plates for RNA studies, glass slides with water. All liquid drops were carefully aspirated. This was
for scanning electron microscopy (SEM), and Aclar for followed by the addition of 0.2 mL of 0.2 m NaOH containing
transmission electron microscopy (TEM). 40 µg/mL of herring sperm DNA to each culture plate. Cells were
The ultrasound mist was applied 1.0 cm from the cell culture returned to the incubator for several hours, and 100 µL of liquid
and delivered an energy of 0.002 W/cm2, according to the was taken in 3 mL of scintillation cocktail.40
manufacturer’s specifications. The duration of application was Two hours and 1, 2, and 3 days after treatment, cells were
determined as the longest time during which the ultrasound mist extracted with TriZol (Life Technologies, Inc., Carlsbad, CA) for
did not dislodge cells from the plate surface. After the cell cultures total RNA; 20 µg of total RNA from each sample was used for
had become confluent, a sterile saline mist was applied (30 s to Northern blotting. Probes for keratinocyte growth factor (KGF),
100-mm2 plates, 15 s to six-well plates, and 7 s to 24-well plates) transforming growth factor β-1 (TGF-β1), and glyceraldehyde-3-
to the cells in a circular motion using a form of ultrasound at 40 kHz phosphate dehydrogenase (GAPDH) were generated from
that is noncontact in nature, except for a fine mist that is created plasmids. GAPDH was assessed for equal loading.
by the system evaluated (Celleration™, Eden Prairie, MN). As Two hours and 1, 2, 3, and 4 days after treatment, cells were
ultrasound mists may detach cells from the supporting surface, an extracted with MAPK lysis buffer for total proteins.41 Extracts were
in vitro cell scrape-wounded model and untreated fibroblasts were heated at 100 °C for 5 min, loaded on to a 12% polyacrylamide gel
employed as controls. Wounding was created by making two for electrophoresis, and transferred to a membrane. The
intersecting scrapes on the cell cultures. membrane was blotted against activated ERK 1/2 (Cell), total ERK
Fibroblasts were seeded on to glass slides for SEM and Aclar (Santa Cruz Biotechnology, Santa Cruz, CA), and JNK (Cell
for TEM. Both types of slide were trimmed to fit six-well plates. Signaling, Beverly, MA). Total ERK was assessed for equal
Immediately, 2 h, and 1 day after ultrasound treatment, cells were loading.

International Journal of Dermatology 2007, 46, 587– 593 © 2007 The International Society of Dermatology
Lai and Pittelkow Physiological effects of ultrasound mist on fibroblasts Report 589

Figure 1 Transmission electron

microscopy. Fibroblasts were fixed after
ultrasound treatment and observed under
a transmission electron microscope:
(a) untreated control; (b) immediately
after treatment; (c) 1 h after treatment;
(d) 1 day after treatment. Magnification:
×50,000. Examples of vacuoles are
indicated by arrows. The large dark
organelles in the cytoplasm are lysosomes

The amounts of RNA in Northern blotting and protein in Western

blotting were measured by the digital imaging software TotalLab
(Phoretix, Durham, NC), and equal loading was calibrated on the
basis of the amounts of GAPDH in Northern blotting and total ERK
in Western blotting.

Results

Samples were examined under SEM and TEM at various

magnifications. No difference in the cell membrane was
Figure 2 Cellular DNA synthesis. Quiescent fibroblasts treated
observed between the control and any treatment group (TEM with tritiated thymidine incorporation assay. No difference was
data shown in Fig. 1; SEM data not shown); however, observed between ultrasound-treated and scrape-wounded
ultrasound-treated cells exhibited large vacuoles in the cytoplasm fibroblasts
(Fig. 1b–d). As vacuoles are associated with the degradation
of cellular proteins and macromolecules, we suspected that
ultrasound waves mechanically stimulated the intracellular
cytoskeleton and induced cellular remodeling. showed a decrease in KGF expression on the first day and a
DNA synthesis decreased 2 h after treatment, possibly related steady increase afterwards. TGF-β1 expression was elevated
to limited cell detachment from the plate, and then increased in ultrasound-treated fibroblasts in the first 2 days, and was
significantly at days 1 and 2 and declined at day 3. No significant decreased in wounded fibroblasts in the first 2 days (Fig. 3).
difference was observed between ultrasound-treated and scrape-
wounded fibroblasts (Fig. 2), that is, ultrasound induced ERK activation
similar reparative activities to scrape-wounding. Phospho-ERK expression in ultrasound-treated fibroblasts
Ultrasound-treated fibroblasts exhibited a marked increase was markedly increased 2 h after treatment and returned to
in KGF expression shortly after treatment, but this decreased baseline on day 2; however, no significant changes in ERK acti-
gradually. In contrast, fibroblasts in the wounding group vation over time were observed in wounded fibroblasts (Fig. 4).

© 2007 The International Society of Dermatology International Journal of Dermatology 2007, 46, 587– 593
590 Report Physiological effects of ultrasound mist on fibroblasts Lai and Pittelkow

Figure 3 Northern blotting. Total RNA

was extracted from fibroblasts after
ultrasound treatment (US) or scrape-
wounding (Wound). The intensities of
keratinocyte growth factor (KGF) and
transforming growth factor β-1 (TGF-β1)
were calibrated on the basis of
glyceraldehyde-3-phosphate
dehydrogenase (GAPDH)

JNK activation converting a chronic wound to an acute state. Therefore, it

Phospho-JNK in ultrasound-treated fibroblasts was markedly would be expected that scrape-wounding would induce cell
increased at 2 h, but returned to baseline by day 2. Similarly, replication and repair. KGF specifically promotes the prolif-
phospho-JNK was increased in wounded fibroblasts and eration and migration of keratinocytes, rather than fibro-
returned to baseline on day 2 (Fig. 4). The substantial increase blasts.42 TGF-β stimulates the regeneration of soft tissue and
in phospho-JNK immediately after ultrasound treatment the production and deposition of collagen by fibroblasts.43,44
may result from the physical interactions of the ultrasound Therefore, even though ultrasound treatment increased the
mist with cell membranes, as JNK is a member of the stress- production of KGF and TGF-β, it produced a similar stimu-
activated kinase family. lation of fibroblast DNA synthesis to wounding.
Ultrasound treatment also increased the activation of both
ERK/JNK ratio ERK and JNK, as had been proposed. Interestingly, the ERK/
The ERK/JNK ratio was increased on day 1 in ultrasound- JNK ratios in ultrasound-treated and wounded fibroblasts
treated fibroblasts; however, it was markedly decreased at 2 h exhibited opposite patterns; however, various downstream
and on day 1, and increased soon after, in wounded fibro- events, such as protein expression, that may be beneficial to
blasts (Fig. 5). Ultrasound treatment enhanced the ERK- wound healing were not measured. The cellular repair
associated repair mechanism at day 1, whereas scrape-wounding induced by ultrasound mist treatment is short term and
only showed similar effects at days 2 – 4. The balance between distinct from cell wounding. This observation implies that
ERK and JNK activation may regulate cell proliferation, ultrasound treatment may need to be applied at a frequency
growth arrest, and other cell repair and survival mechanisms. to optimize ERK activation, such as every 1– 2 days. In addi-
tion, under these treatment conditions in vitro, JNK activa-
tion may indicate cellular stress that is not necessarily linked
Discussion
to apoptosis and cell death, but rather to compensatory
It is common practice to debride chronic wounds and responses mediating cellular repair. Further investigations on
enhance healing by the removal of necrotic tissue, thus dose-dependent energy delivered to cells and the influx of calcium

International Journal of Dermatology 2007, 46, 587– 593 © 2007 The International Society of Dermatology
Lai and Pittelkow Physiological effects of ultrasound mist on fibroblasts Report 591

Figure 4 Western blotting. Top:

(a) phospho-extracellular regulated kinase
(phospho-ERK); (b) phospho-c-Jun
N-terminal kinase (phospho-JNK);
(c) total ERK as reference. Bottom:
the amount of phospho-ERK was
significantly higher at 2 h and on day 1 in
ultrasound-treated fibroblasts. The levels
of phospho-JNK were increased at 2 h in
ultrasound-treated fibroblasts (U) and
at 2 h and on day 1 in wounded
fibroblasts (W)

cavitation in the fluid, breaking it apart into small particles of

well-controlled size. Once the particles of fluid are released
from the tip, a second phenomenon drives them away and
towards the wound. This second phenomenon is called
acoustic streaming, or acoustic radiation force. At higher
pressure levels, the traveling acoustic wave actually imparts
a net force on the propagation medium, in this case the air.
The force is sufficient to drive the particles of fluid along
with it.

Figure 5 Extracellular regulated kinase/c-Jun N-terminal kinase Conclusion

(ERK/JNK) ratio. The ERK/JNK ratio was increased early in On the basis of these and other observations, ultrasound
ultrasound-treated fibroblasts and late in wounded fibroblasts
seems to be a promising modality for wound healing. Never-
theless, further investigation of the expression of cytokine
genes, cellular membrane effects, and cellular regenerative
channels may provide more detailed information on the outcomes may better demonstrate the effects of ultrasound
mechanisms by which ultrasound mist induces cellular and predict the clinical efficacy of ultrasound on chronic
responses. wounds. We theorize that ultrasound waves from atomi-
This ultrasound mist system uses two ultrasonic effects to zation and acoustic streaming vibrate the impacted cell
generate and propel the therapeutic mist towards the wound. membrane and create mechanical stimulation that initiates
The first effect is ultrasonic atomization, sometimes referred intracellular signaling networks.30,31 Ultrasound mist therapy
to as nebulization. This is similar to ultrasonic humidifiers. could also be applied to wound beds, where sharp debride-
The saline solution which is applied to the metal face is atom- ment is not applicable, to promote stromal and epithelial
ized through vibration of the surface. This vibration causes regeneration.

© 2007 The International Society of Dermatology International Journal of Dermatology 2007, 46, 587– 593
592 Report Physiological effects of ultrasound mist on fibroblasts Lai and Pittelkow

Acknowledgments 19 Turner S, Powell E, Ng C. The effect of ultrasound on the

healing of repaired cockerel tendon: is collagen cross-linkage
This project was supported by Celleration, Inc., Eden Prairie, a factor? J Hand Surg 1989; 14B: 428 – 433.
MN, and the National Institutes of Health (NIH) grant T32 20 El-Batouty M, El-Gindy M, El-Shawat X. Comparative
HD 07447. The authors wish to thank Karen Squillace for evaluation of the effects of ultrasound and ultraviolet
technical support. irradiation on tissue regeneration. Scand J Rheumatol 1986;
15: 381– 386.
21 Dyson M. The effect of ultrasound on the rate of wound
References
healing and the quality of scar tissue. In: International
1 Association AD. Economic costs of diabetes in the U.S. in Symposium on Therapeutic Ultrasound, Winnipeg, 1981.
2002. Diabetes Care 2003; 26: 917 – 932. 22 Jackson BA, Schane JA, Starcher BC. Effect of ultrasound
2 Bloomgarden ZT. Neuropathy, information technology, therapy on the repair of Achilles tendon injuries in rats. Med
cost of diabetes care, and epidemiology. Diabetes Care Sci Sports Exercise 1989; 23: 171–176.
1998; 21: 1198 –1202. 23 Webster DF, Harvey W, Dyson M, et al. The in vitro
3 Bloomgarden ZT. Epidemiology and neuropathy. Diabetes stimulation of collagen synthesis in human fibroblasts by
Care 1998; 21: 2023 – 2027. ultrasound induced cavitation. Ultrasonics 1980; 16: 33.
4 Currie CJ. The epidemiology and cost of inpatient care 24 Byl NN, McKenzie AL, Wong T, et al. Incisional wound
for peripheral vascular disease, infection, neuropathy, healing: a controlled study of low and high dose ultrasound.
and ulceration in diabetes. Diabetes Care 1998; 21: J Orthop Sports Phys Ther 1993; 18: 619 – 628.
42 – 48. 25 Baker KG, Robertson VJ, Duck FA. A review of therapeutic
5 Mehta SS. Determining an episode of care using claims data. ultrasound: biophysical effects. Phys Ther 2001; 81:
Diabetic foot ulcer. Diabetes Care 1999; 22: 1110 –1115. 1351–1358.
6 O’Brien JA. Direct medical costs of complications resulting 26 Flemming K, Cullum N. Therapeutic ultrasound for pressure
from type 2 diabetes in the US. Diabetes Care 1998; 21: sores. Cochrane Database Syst Rev 2000; 4.
1122 –1128. 27 Flemming K, Cullum N. Therapeutic ultrasound for venous
7 Ollendorf DA. Potential economic benefits of leg ulcers. Cochrane Database Syst Rev 2000; 4.
lower-extremity amputation prevention strategies in 28 ter Haar G. Therapeutic ultrasound. Eur J Ultrasound 1999;
diabetes. Diabetes Care 1998; 21: 1240 –1245. 9: 3 – 9.
8 Mancini L. The diabetic foot: epidemiology. Rays 1997; 22: 29 Maxwell L, Collecutt T, Gledhill M, et al. The
511– 523. augmentation of leucocyte adhesion to endothelium by
9 Angle N, Bergan JJ. Chronic venous ulcer. Br Med J 1997; therapeutic ultrasound. Ultrasound Med Biol 1994; 20:
314: 1019 –1023. 383 – 390.
10 Lindholm C, Bergsten A, Berglund E. Chronic wounds and 30 Chen K-D, Li Y-S, Kim M, et al. Mechanotransduction in
nursing care. J Wound Care 1999; 8: 5 –10. response to shear stress. J Biol Chem 1999; 274: 18 393–18
11 Eaglstein WH, Falanga V. Chronic wounds. Surg Clin North 400.
Am 1997; 77: 689 – 700. 31 Bogoyevitch MA, Ketterman AJ, Sugden PH. Cellular
12 Johannsen F, Nyholm A, Karlsmark T. Ultrasound therapy stresses differentially activate c-Jun N-terminal protein
in chronic leg ulceration: a meta-analysis. Wound Repair kinases and extracellular signal-regulated protein kinases in
Regeneration 1998; 6: 121–126. cultured ventricular myocytes. J Biol Chem 1995; 270: 29
13 Houghton PE, Campbell KE. Choosing an adjunctive 710 – 29 717.
therapy for the treatment of chronic wounds. Ostomy/ 32 Peus D, Vasa RA, Beyerle A, et al. UVB activates ERK1/2
Wound Manage 1999; 45: 43 – 52. and p38 signaling pathways via reactive oxygen species in
14 DeDeyne P, K-Volders M. In vitro effects of therapeutic cultured keratinocytes. J Invest Dermatol 1999; 112:
ultrasound on the nucleus of human fibroblasts. Phys Ther 751– 756.
1995; 75: 629 – 634. 33 Peus D, Meves A, Vasa RA, et al. H2O2 is required for
15 Young S, Dyson M. Macrophage responsiveness to UVB-induced EGF receptor and downstream signaling
therapeutic ultrasound. Ultrasound Med Biol 1990; 16: pathway activation. Free Radic Biol Med 1999; 27:
809 – 816. 1197 –1202.
16 Al-Karmi IS, Dinno MA, Stoltz DA, et al. Calcium and the 34 Peus D, Vasa RA, Meves A, et al. H2O2 is an important
effects of ultrasound on frog skin. Ultrasound Med Biol mediator of UVB-induced EGF-receptor phosphorylation in
1994; 20: 73 – 81. cultured keratinocytes. J Invest Dermatol 1998; 110:
17 Dinno MA, Dyson M, Young SR, et al. The significance of 966 – 971.
membrane changes in the safe and effective use of 35 Pittelkow MR. Growth factors in cutaneous biology and
therapeutic and diagnostic ultrasound. Phys Med Biol 1989; disease. Adv Dermatol 1992; 7: 55 – 81.
34: 1543 –1552. 36 Coquette A, Berna N, Poumay Y, et al. The keratinocyte in
18 Mortimer AJ, Dyson M. The effects of therapeutic cutaneous irritation and sensitization. In: Kydonieus AF,
ultrasound on the nucleus of human fibroblasts. Ultrasound Wille JJ, eds. Biochemical Modulation of Skin Reactions.
Med Biol 1988; 14: 499 – 506. New York: CRC, 2000: 125 –143.

International Journal of Dermatology 2007, 46, 587– 593 © 2007 The International Society of Dermatology
Lai and Pittelkow Physiological effects of ultrasound mist on fibroblasts Report 593

37 Choquet D, Felsenfeld DP, Sheetz MP. Extracellular matrix to constitutively associate with insulin receptor substrate-1
rigidity causes strengthening of integrin–cytoskeleton and -2 and cause severe cell resistance to both insulin and
linkages. Cell 1997; 88: 39 – 48. insulin-like growth factor I. J Clin Endocrinol Metab 1999;
38 McDowell EM, Trump BF. Histologic fixatives suitable for 84: 3197 – 3206.
diagnostic light and electron microscopy. Arch Pathol Lab 42 auf demKeller U, Krampert M, Kumin A, et al. Keratinocyte
Med 1976; 100: 405 – 414. growth factor: effects on keratinocytes and mechanisms of
39 Spurr AR. A low-viscosity epoxy resin embedding action. Eur J Cell Biol 2004; 83: 607 – 612.
medium for electron microscopy. J Ultrastruct Res 1969; 43 Brunner G, Blakytny R. Extracellular regulation of
26: 31– 43. TGF-beta activity in wound repair: growth factor latency as
40 Peus D, Hamacher L, Pittelkow MR. EGF-receptor tyrosine a sensor mechanism for injury. Thromb Haemost 2004; 92:
kinase inhibition induces keratinocyte growth arrest and 253 – 261.
terminal differentiation. J Invest Dermatol 1997; 109: 44 Wang J, Zheng H, Sung CC, et al. Cellular sources of
751– 756. transforming growth factor-beta isoforms in early and
41 Auclair M, Vigouroux C, Desbois-Mouthon C, et al. chronic radiation enteropathy. Am J Pathol 1998; 153:
Antiinsulin receptor autoantibodies induce insulin receptors 1531–1540.

© 2007 The International Society of Dermatology International Journal of Dermatology 2007, 46, 587–593