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

F. Syed; E. Ahmadi; S.A. Iqbal; S. Singh; D.A. McGrouther; A. Bayat

Disclosures

The British Journal of Dermatology. 2011;164(1):83-96. 

In This Article

Results

Differential Transcription of Type I and III Collagen mRNA in Different Lesional Sites of Keloid Scars

Keloid fibroblasts from intralesional and perilesional sites exhibited raised type I and III collagen mRNA levels (Fig. 2). However, type I collagen mRNA was significantly different at the various lesional sites of the keloid scar (Fig. 2a). Collagen I and III mRNA levels were significantly higher (P < 0·05) in the perilesional site of the keloids compared with extralesional and intralesional sites. Furthermore, in the top (p0) and middle (p0 and p1) of the keloid scars there was a significantly higher expression (P < 0·05) of collagen I mRNA compared with the extralesional site (p0–p3), and collagen III mRNA was significantly higher (P < 0·05) in the middle (p0, p1 and p2) compared with extralesional and intralesional (top) sites (p0–p3) (Fig. 2b). Interestingly, as the passage number increased, the expression of collagen I mRNA decreased in the following order (p3 < p2 < p1 < p0). Consistent with the in vitro results, similar expression patterns of collagen I and III were also observed in different lesional sites in the in vivo keloid tissue samples (Fig. 2c).

Figure 2.

Expression of type I and type III collagen mRNA in different sites of the keloid lesion by quantitative reverse transcription–polymerase chain reaction (RT-PCR) and ethidium bromide-stained agarose gel electrophoresis. Fibroblasts from different sites of the keloid were cultured from passage 0 (p0) to passage 3 (p3). Total RNA was extracted from fibroblasts and real-time RT-PCR was performed as described in the Materials and methods section. The expression of collagen I and collagen III mRNA was normalized to the average of two internal reference genes (RPL32 and SDHA) and the results were expressed as fold change. (a) Collagen I mRNA expression pattern. **P < 0·05 for the comparison between the mRNA expression in cultured fibroblasts from the perilesional site of the keloid (p0–p3) vs. the extralesional site (p0–p3). *P < 0·05 for the comparison between expression of mRNA from the keloid top (p1–p3) vs. the perilesional site (p0–p3). ##P < 0·05 for the comparison between the mRNA expression in keloid middle (p0 and p1) vs. extralesional site. #P < 0·05 for the comparison between the mRNA expression in top (p0–p3) vs. middle (p0 and p1). §P < 0·05 for the comparison between the keloid middle (p2 and p3) vs. the perilesional site (p0, p1 and p3). (b) Collagen III mRNA expression pattern. **P < 0·05 for the comparison between the mRNA expressed in cultured fibroblasts from the perilesional site of the keloid (p0–p3) vs. extralesional site (p0–p3). *P < 0·05 for the comparison between the expression of mRNA from the keloid top (p1–p3) vs. the perilesional site (p0–p3). ##P < 0·05 for the comparison between the mRNA expression in keloid middle (p0–p3) vs. extralesional site (p0–p3). #P < 0·05 for the comparison between the mRNA expression in the top (p0–p3) vs. the middle (p0, p2 and p3). §P < 0·05 for the comparison between the keloid middle (p0–p3) vs. the perilesional site (p0, p2 and p3) (n = 8). (c) Collagen I and III mRNA expression pattern in keloid tissue samples from different lesional sites. Total RNA was extracted from tissue samples collected in RNAlater and real-time RT-PCR was performed as described in the Materials and methods section. *P < 0·05 for the comparison between the expression of collagen I mRNA from extralesional and intralesional (top) lesional sites vs. perilesional site. #P < 0·03 for the comparison between the expression of collagen III mRNA from extralesional and intralesional sites vs. perilesional site. Results are presented as mean ± SD of triplicates (n = 5).

Establishment of Capture Sandwich Enzyme-linked Immunosorbent Assay for Collagen I

An optimal antibody pair was identified as the capture antibody [first primary antibody ab59435 and second primary antibody (COL-1) ab6308] and recombinant human collagen I was used for the establishment of the ELISA. Percentage positive values were calculated based on the ELISA results as described previously.[54,55] The sensitivity and specificity of the collagen I ELISA were 100% and 99%, respectively (data not shown). The cross-reactivity was compared with bovine collagen (0·2%), with human type II collagen (0·5%) and with human type III collagen (0·0%); the sensitivity and limitation of the assay were 5 ng mL−1 and 50 μg mL−1, respectively. The intra- and interassay variations were 2·1–8·2% and 3·3–10·2%, respectively, from three samples with different concentrations. This demonstrated excellent reproducibility in the assay. Furthermore, our ELISA technique appears to be superior, in terms of sensitivity and specificity, and is less costly compared with the commercially available diagnostic kit for human collagen type I ELISA (Table S3; see Supporting Information) (Cosmo Bio Ltd, Tokyo, Japan).[56]

Quantitative Measurement of Synthesis of Collagen I by Capture Sandwich Enzyme-linked Immunosorbent Assay

Collagen I was measured in the media, as well as in keloid cultured fibroblasts from different passages (p0–p3). The secretion of collagen I was significantly higher (P < 0·05) in the media collected from the perilesional site (p0) of the keloid cultured fibroblasts, compared with extralesional and intralesional sites. The secretion of collagen I varied between the passages (P < 0·05) from different sites of the keloid lesions. As the passage number increased, the secretion of collagen I decreased in the medium (Figs 3a and S2a; see Supporting Information). However, there was no significant difference observed in the secretion of collagen I between the passages of fibroblasts cultured from the extralesional site. High levels of collagen I synthesis were observed in the fibroblasts cultured from the perilesional site (P = 0·02) of the keloid compared with extralesional and intralesional sites (Figs 3b and S2b; see Supporting Information). In the cultured fibroblasts, there was no consistency observed in the synthesis of collagen I in the intralesional site of the keloid, compared with extralesional (p1–p3) sites of the keloid. There was, however, a significant increase (P < 0·05) in collagen type I synthesis levels seen in the intralesional site (middle) passage (p0 and p1) compared with the extralesional site (p0–p3). However, there was a significant (P = 0·03) reduction observed in the synthesis of collagen I in the extralesional site of the keloid fibroblasts from p1 to p3, compared with the perilesional site cells from p0 to p2 (Figs 3b and S2b; see Supporting Information).

Figure 3.

In vitro quantitative measurement of collagen I by capture sandwich enzyme-linked immunosorbent assay (ELISA). (a) Secretion of collagen I in the fibroblast growth medium cultured from different sites of the keloid lesion. The cultured fibroblast medium was collected at different passages (p0–p3). ELISA was performed in triplicate using 50 μL of culture medium and 50 μL of ELISA coating buffer (2×) as described in the Materials and methods section. ###P < 0·05 for the comparison between extralesional p0–p3 vs. perilesional p0. ##P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (middle) p0 and p1. #P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (top) p1. *P < 0·05 for the comparison between perilesional p0 vs. intralesional p0–p3. (b) Synthesis of collagen I in cultured fibroblasts extracted from different sites of the keloid lesion. The cells were subcultured at every 90–100% confluence from p0 to p3. The collected cells (104 cells mL−1) were lysed by freeze/thaw cycles and 50 μL of cell lysate and 50 μL of ELISA coating buffer (2×) were used as described in the Materials and methods section. ***P < 0·05 for the comparison between extralesional p0–p3 vs. perilesional p0 and p1. **P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (middle) p0 and p1. *P < 0·05 for the comparison between perilesional p0 vs. intralesional p0–p3. Results are presented as mean ± SD of triplicates (n = 8).

Quantitative Measurement of Collagen III Production by Indirect Enzyme-linked Immunosorbent Assay

Collagen III expression also showed a similar pattern to that of collagen I in keloid fibroblasts and cultured medium from the perilesional site. Synthesis of collagen III in the keloid fibroblasts was correlated with the equivalent secretion in the culture medium. Expression of collagen III was significantly higher (P < 0·05) in the medium taken from the intralesional and perilesional sites of the keloid, compared with the extralesional site. There was a significant (P ≤ 0·03) difference observed in the secretion of collagen III between the passages of all lesional sites of keloid (Figs 4a and S3a; see Supporting Information). Perilesional keloid fibroblasts showed a significantly higher level (P < 0·05) of collagen III compared with cells from the extralesional site. There were some variations in the synthesis of collagen III in the different passages, for instance, in extralesional p0–p3 and in intralesional p0–p3, which were not significant (P > 0·05) (Figs 4b and S3b; see Supporting Information). However, there was a statistically significant difference (P = 0·02) observed in the synthesis of collagen III in p0 and p3 of perilesional keloid fibroblasts (Figs 4 and S3; see Supporting Information). The collagen I/III protein ratio decreased from p0 to p3 in all keloid fibroblasts cultured from all lesional sites. In comparison with other lesional sites, perilesional fibroblasts expressed the highest collagen I/III ratio in p0, then the ratio decreased, which reflects the increase of collagen III (Table 2).

Figure 4.

In vitro quantitative measurement of collagen III by indirect enzyme-linked immunosorbent assay (ELISA). (a) Secretion of collagen III in the cultured fibroblast growth medium, extracted from different sites of the keloid lesion. The cultured fibroblast medium was collected at different passages (p0–p3). ELISA was performed in triplicate using 50 μL of culture medium and 50 μL of ELISA coating buffer (2×) as described in the Materials and methods section. ###P < 0·05 for the comparison between extralesional p0–p3 vs. perilesional p0–p3. ##P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (middle) p0–p3. #P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (top) p0–p3. (b) Synthesis of collagen III in cultured fibroblasts, extracted from different sites of the keloid lesion. The cells were subcultured at every 90–100% confluence from p0 to p3. The collected cells (104 cells mL−1) were lysed by freeze/thaw cycles and 50 μL of cell lysate and 50 μL of coating buffer (2×) was used in the ELISA as described in the Materials and methods section. ***P < 0·05 for the comparison between extralesional p0–p3 vs. perilesional p0–p3. **P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (middle) p0–p3. *P < 0·05 for the comparison between extralesional p0–p3 vs. intralesional (top) p0–p3. Results are presented as mean ± SD of triplicates (n = 8).

Quantitative In-cell Western Blotting Assay for Analysis of Collagen Expression

In-cell Western blotting was used to corroborate our results of qRT-PCR and ELISA for collagen I and III expression. In-cell Western blotting assay showed a similar pattern of collagen expression as seen in the qRT-PCR and ELISA. Collagen I expression in keloid fibroblasts decreased as the passage number increased and collagen III expression increased as the passage number increased. Collagen I and III protein expression was high in the perilesional site (margin) of keloid fibroblasts compared with the intralesional (top and middle) and extralesional sites (Fig. 5).

Figure 5.

Collagen I and III expression kinetics in different passages of keloid fibroblasts. Serum-starved keloid fibroblasts were seeded in 96-well plates (10 000 cells per well) and grown up to 24 h. The cells were fixed with 4% formaldehyde/phosphate-buffered saline. After blocking, the total collagen I and III were revealed by incubation with anticollagen I and III, followed by incubation with IRDye 800-labelled secondary antibodies. A representative output infrared image of keloid fibroblasts for collagen I and III (visible in green) expression in different sites of different passages from 96-well plates is shown. Bar graphs represent the quantification of mean collagen I and III protein expression in different passages (p1–p4) from four different patients. Results are presented as mean ± SD of triplicates.

Collagen I and III Detection by Histological Staining

H&E staining showed normal epidermis in the extralesional site, whereas intralesional (top) and perilesional sites showed a thicker, irregular epidermis (Fig. 6). There was an increased number of thickened dense collagen bundles observed in the reticular dermis in the perilesional site of the keloid, compared with the intralesional and extralesional sites. In the perilesional site of the keloid, the thick collagen bundle zones were highly cellular at the papillary dermis and less so at the reticular dermis (Fig. 6).

Figure 6.

Morphological analysis of different keloid lesional sites using haematoxylin and eosin (H&E). Keloid tissue was formalin fixed for 24 h and embedded in paraffin. Five-micrometre sections were cut and, using standard protocol, H&E staining was performed as described in the Materials and methods section. Areas of inflammation were also observed in the intralesional top section of keloid (blue arrows). K, keratin layer; EP, epidermis; PD, papillary dermis; RD, reticular dermis. Original magnification × 400.

Herovici staining showed that in all the keloid lesional sites examined, the reticular dermis was composed of a complex mixture of red and blue staining fibres. The red fibres (collagen I) were always thicker and of a coarser arrangement than those which stained blue (collagen III) (Fig. 7). The staining pattern differed in histological sections obtained from all the lesional sites, in terms of the amount of blue and red staining, and the collagen type I to III distribution. In the extralesional site, there was a reduced amount of collagen III present only in the papillary dermis, and the remainder stained for collagen type I. However, this pattern was not observed in the intralesional and perilesional sites of the keloid. In the intralesional and perilesional sites, collagen I and III were present across the tissue sections, with more collagen III in the papillary dermis in relation to collagen I. The opposite applied to the reticular dermis. However, in the perilesional site of reticular dermis, highly dense bundles of collagen I fibres were observed while collagen III was almost completely absent (Fig. 7). Taken together, our results clearly demonstrate that collagen I and III are highly expressed in perilesional sites, compared with intralesional and extralesional sites of the keloid scar.

Figure 7.

Morphological analysis of different keloid lesional sites using Herovici staining. Formaldehyde-fixed tissue was used for the Herovici staining. The tissues were embedded in paraffin and cut into 5-μm sections and stained with picro methyl blue and picro acid fuchsin as described previously.51 Young collagen and reticulin shown in blue, mature collagen in red, cytoplasm in yellow/green and nuclei in black. K, keratin layer; EP, epidermis; PD, papillary dermis; RD, reticular dermis. Original magnification × 400.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....