This study has demonstrated that biopsies taken from the growing margin (perilesional) of keloid scars show a greater collagen production than those taken from intralesional and extralesional sites from the same keloid lesion. In addition, as the passage number increases, the collagen I/III ratio decreases. To corroborate the qRT-PCR results of collagen mRNA expression in the present study, we have developed and evaluated a novel sensitive sandwich ELISA technique which measures the amount of secreted collagen in the media and in the cells, as well as utilizing in-cell Western blotting. Furthermore, to distinguish histologically collagen type I from type III in the same tissue, we utilized the Herovici staining technique in all the lesional biopsy sites.
Our results showed that there was a significant increase in collagen I and III mRNA in the perilesional site of the keloid compared with extralesional and intralesional sites in the in vitro cell culture and in vivo tissue samples. Concurrently, there was a significant decrease in the synthesis of collagen I and a marginal increase in collagen III in keloid fibroblasts as the passage number increased, as shown by qRT-PCR, ELISA and in-cell Western blotting. Furthermore, immunohistological investigation showed a higher expression of collagen I and III in the perilesional site of the keloid. This may indicate a possible mechanism of keloid fibroblast differentiation in vitro and this observation may also suggest that the expression of the genetically distinct collagen genes can be regulated in an independent manner. Furthermore, type I collagen mRNA and protein levels were found to increase proportionally in keloid cell culture. It appears that the production of type I collagen is largely determined by the level of the corresponding mRNA. Our results strongly correlate with previous studies which showed steady-state alterations in the collagen I/III ratio.[36,43] In contrast to our results, Ala-Kokko et al. observed no significant increase in the collagen I/III ratio in their study. However, this difference could be due to possible changes in the culturing and in the passage number of the keloid fibroblasts, as their study used passages 3–10, whereas we used passages 0–4 in our study.
Type III collagen mRNA has been reported to remain unaltered in both keloid tissue and cultured fibroblasts obtained from keloid lesions.[37,44] However, in our study type III collagen mRNA and protein levels were upregulated in keloid, and the collagen I/III ratio was higher in the perilesional site of the keloid. This ratio decreased as the passage number increased. These results are in agreement with a study performed by Naitoh et al. Peltonen et al. examined keloid tissue from six patients and reported that collagen III gene expression was not elevated in five cases. However, in one case, which had the most recent onset of keloid scarring, the fibroblasts showed higher collagen III expression. These authors subsequently suggested that heterogeneity between keloids may exist, and that type III collagen gene expression is transiently activated in the early stages of development in keloid lesions. Indeed, our findings are in agreement with others[36,43] and these negative results may be further explained by the heterogeneity within the keloid lesion itself. At any time point, it is likely that there are actively developing (perilesional) and relatively quiescent portions (intralesional) present within the keloid lesion. Clinically, we have also noted that the centre of the keloid is often less symptomatic, and that symptoms (in terms of pain and constant pruritus reported by the patient) are mostly confined to the margin or periphery of the lesion. In this study, we have further explored in detail the heterogeneous nature of keloid tissue in terms of expression of collagen, and collagen I/III ratios in extralesional and intralesional sites compared with perilesional sites. In support of our findings and conclusions, Zhang et al.[38,39] have demonstrated the upregulation of collagen III in hypertrophic scars and Naitoh et al., in keloid scars. Recently, Verhaegen et al. showed that collagen bundles were significantly thicker in keloid scars, and collagen bundles were more randomly organized in the deep dermis compared with normal skin using fast Fourier transform technique. This adds further support to our findings of thick hyalinized collagen bundles seen in the growing margins of keloid lesional biopsies compared with extralesional and intralesional biopsies by use of H&E and Herovici staining.
It is reported that keloid fibroblasts selectively increase the biosynthesis of type I collagen, and that this abnormal metabolism results in the deposition of collagen in keloid, and alters the steady-state type I/III collagen ratio in the process of wound healing.[36,37] Thus, we demonstrated in our study that perilesional site fibroblasts have a higher activity in collagen production in comparison with intralesional and extralesional keloid scar sites. Our results show that the type I/III collagen ratio may be changed towards a normal phenotype, with relatively simple modification of the fibroblast environment by higher passages. However, we have only investigated up to passage 4. The mechanism behind the decrease of collagen I gene and protein expression with the increase in passage number is unclear at this stage. Decreased collagen degradation could also be a possible cause for increased collagen levels seen in keloids, either influenced by increased levels of collagenase inhibitor or by decreased levels of collagenase itself.
Recent research has shown that epithelial–mesenchymal interactions play an important role in skin homeostasis, growth and differentiation.[60–62] The secretory role of keratinocytes is also important in terms of its effect upon keloid fibroblast behaviour and their growth kinetics, and may be modulated by the overlying keratinocytes in the epidermis.[59,61] Indeed, normal fibroblasts cocultured with keloid-derived keratinocytes produce increased levels of collagen I in a keloid-like manner. This may provide an alternative hypothesis to explain the decreased collagen I and III synthesis present as the passage number increases. Dalkowski et al. showed that the collagen I/III ratio decreased after freezing keloid fibroblasts, which is similar to the phenotype of a fresh scar. In our study, the collagen I/III ratio decreased as the passage number increased, which may be due to changes observed in the keloid fibroblast phenotypes utilized in our study. However, the exact effect of keloid keratinocytes upon keloid fibroblasts, in terms of collagen production in different passages of fibroblasts, remains unknown and should be the subject of further investigation.
In conclusion, this study has shown for the first time that perilesional keloid fibroblasts have a higher production of collagen I and III compared with intralesional and extralesional keloid cells. In addition, we showed a difference in cellular activity in very early compared with later passages in terms of collagen production. These findings may influence interpretation of results and possible selection of specific passages in cell-culture experiments. In addition, our results indicate that utilization of a higher passage number of keloid fibroblasts in in vitro cell culture experimentation may not provide an ideal model for studying keloids. Importantly, in terms of clinical relevance, our novel findings in differential collagen expression may provide direction for lesional site-specific therapy in the treatment of keloid scars and prevention of recurrence of keloid scars.
Funding was provided by NIHR (UK), the GAT Family Foundation, Steve and Kathy Fitzpatrick and the North American Foundation for the University of Manchester.
Conflicts of interest
We thank Mr Dean M. Barton (LI-COR, Cambridge, U.K.) for assistance with in-cell Western blotting experimental work.
The British Journal of Dermatology. 2011;164(1):83-96. © 2011
Cite this: Fibroblasts from the Growing Margin of Keloid Scars Produce Higher Levels of Collagen I and III Compared With Intralesional and Extralesional Sites: Clinical Implications for Lesional Site-directed Therapy - Medscape - Jan 01, 2011.