Collagen Study

Abstract

Background

Oral collagen peptides supplementation was reported to improve skin integrity and counteract skin aging.

Aims

A randomized, double-blinded, placebo-controlled study was conducted to clinically evaluate the impact of low-molecular-weight collagen peptides on the human skin.

Patients/Methods

Healthy adult participants (n = 100) were randomly assigned to receive a test product containing low-molecular-weight collagen peptides or a placebo. Parameters of skin wrinkles, elasticity, hydration, and whitening (melanin and erythema indexes) were measured at baseline and after 4, 8, and 12 weeks.

Results

Compared with the placebo group, the average skin roughness, maximum of all peak-to valley values, maximum peak height of the wrinkle, and average maximum height of the wrinkle were significantly improved in the test group. Parameters of skin elasticity, including overall elasticity, net elasticity, and biological elasticity, were also significantly improved in the test group at Week 12 as compared with the placebo group. Moreover to the test product.

Conclusions

Taken together, these findings suggest that low-molecular-weight collagen peptides supplementation can safely enhance human skin wrinkling, hydration, elasticity, and whitening properties.

1 INTRODUCTION

As humans long for youth and beauty without aging, cosmetic supplements to improve biophysical properties of the skin have gained popularity. 1 In particular, collagen supplements became a research hotspot owing to their potential to improve skin integrity and counteract skin aging. During intrinsic and extrinsic aging, skin changes involve both epidermal and dermal thickness, which are mainly attributed to the loss of collagen and elasticity, as new extracellular matrix component synthesis by dermal fibroblasts diminishes and matrix metalloproteinase (MMP)-mediated collagen degradation is accelerated. 2 Therefore, the dermal collagen network undergoes the accumulation of shorter and fragmented collagen fibers, resulting in wrinkled and loosen skin. 2, 3 Further, the capacity to contain moisture and the amount of hyaluronic acid in the skin decrease with age, therefore leading to dry skin of the elderly. 2

Regarding antiaging strategies, collagen peptides are promising for enhancing skin moisture, elasticity, and density without significant adverse effects. 4 Collagen hydrolysates generated by enzymatic modulation of collagen molecules are composed of peptides with varying lengths. These peptides can be further degraded into small dipeptides and tripeptides resistant to peptidase activity, which can be found in the human bloodstream 1–2 h after oral intake. 5-8 Moreover, in vivo studies demonstrated that collagen-derived bioactive peptides can reach the skin and are retained in the tissue for up to 2 weeks, 9 where they can attenuate ultraviolet B-induced aging. 10

Low-molecular-weight collagen peptides are a form of collagen hydrolysates rich in hydroxyproline, glycine, and proline amino acids. It was suggested that these peptides can stimulate human fibroblasts to synthesize extracellular matrix molecules and strengthen the skin tissue. 7 In recent years, several clinical trials reported consistent data on oral collagen supplements with beneficial effects on the skin. 11-13 The aim of this study is to evaluate the clinical efficacy of oral low-molecular-weight collagen peptides supplementation for preventing wrinkles as well as promoting skin whitening via biophysical and skin imaging techniques.

 
2 MATERIALS AND METHODS 

Test product is a food supplement that contains low-molecular-weight collagen peptides (GPVGPS Collagen®, Daehan Chemtech Co., Ltd.) obtained from fish scales of Nile tilapia (Oreochromis niloticus), along with maltodextrin, silicon dioxide, and sucralose. The placebo contained all these ingredients but no nutrients. The test product and placebo were provided as a 2.5 g white powder blend, which was administered with water daily, in the morning before a meal. Detailed percentage of each ingredient is shown in Table 1.

TABLE 1. Ingredient composition of the test product and placebo product.

Ingredient	Test		Placebo
	mg	%	mg	%
Low molecular collagen peptide (GPVGPS Collagen®)	2000.00	80.00	0.00	0.00
Maltodextrin	348.75	13.95	2348.75	93.95
Flavor powder	125.00	5.00	125.00	5.00
Silicon dioxide	25.00	1.00	25.00	1.00
Sucralose	1.25	0.05	1.25	0.05
Total	2500.00	100.00	2500.00	100.00

2.2 Study design

In our randomized, double-blinded, placebo-controlled clinical trial, all participants were randomly distributed to either the test product or placebo at a 1:1 ratio. Each participant took a packet of powder, which was either a study product or a placebo, once a day for 12 weeks. The number of assessments included four visits: a baseline visit at Week 0 and three more visits determined at Weeks 4, 8, and 12.

2.3 Study participants

A total of 100 adults with their age in 35–60 years, who had dry skin and periorbital wrinkles, were enrolled in the study. Written informed consent of all participants was obtained. The eligibility criteria of the healthy volunteers were as follows: a score of three or more on the 10-grade crow's feet photo scale 14 and skin hydration less than 50 arbitrary units measured using a Corneometer (Courage & Khazaka Electronic) on both cheeks. The study had several exclusion criteria, which were carefully selected to ensure the safety and integrity of the results. These criteria included current pregnancy or lactation, major visceral organ disease, topical immunomodulators within the past 3 months, and intake of oral bioactive substances within 14 days before the participation.

2.4 Measurements

Under a controlled environment with 40%–60% humidity and 20°C–24°C room temperature, the clinical efficacy of each parameter was measured. All participants were instructed to maintain rested conditions and restrict water intake 1 h before the assessment. Skin wrinkles were evaluated on the periorbital area using PRIMOS CR (GFMesstechnik, Teltow, Germany), an optical three-dimensional non-contact measuring device. Among the skin surface descriptors in PRIMOS, Ra, Rmax, Rp, Rv, and Rz were assessed to represent skin wrinkles and roughness. In addition, the eye wrinkle volume was assessed to determine three-dimensional volumetric changes during the period of this study. Skin surface hydration was measured using Corneometer CM 825 (Courage & Khazaka Electronic), which detects the hydration level below the stratum corneum by electrical capacitance. Skin surface hydration was assessed on the forehead (2 cm above the glabella), forearm (10 cm apart from the inner side of the wrist), cheeks (right angle intersection between lateral canthus and the tip of the nose), and 1 cm below the eyes. The average value of three stabilized measurements was used.

Skin elasticity was evaluated using Cutometer MPA 580 (Courage & Khazaka Electronic). The study evaluated the skin's extension when subjected to a suction vacuum applied above the cheeks using a 450-mbar vacuum. This involved a pattern of 2-s exposure followed by a 2-s non-exposure period, repeated three times. Skin overall (R2), net (R5), and biological (R7) elasticities were determined. Skin whitening was assessed using Mexameter MX 18 (Courage & Khazaka Electronic) on the cheek, representing melanin and erythema indexes. The average value of three stabilized measurements was used.

2.5 Safety assessments

After being randomly assigned to the study, the participants were included for safety analysis if the test product or placebo was consumed at least once. A comprehensive physical examination at each visit was conducted to assess the safety. Furthermore, blood samples from all participants were collected at baseline and after the end of study to measure the complete blood cell count, liver and renal function, and glucose and cholesterol levels. Urine samples were also collected and analyzed for abnormalities in the same manner.

Statistical analyses applied descriptive statistics for all data and included comparison before and after treatment using paired t or Wilcoxon signed-rank tests, and between groups using two-sample t or Wilcoxon rank sum tests. For either normal or abnormal result in urinalysis, McNemar's test was used. The incidence rate of adverse effects in each group was compared using chi-square or Fisher's exact tests. SAS software (version 9.4 for Windows; SAS Institute Inc.) was used for all statistical analyses. p-values < 0.05 were considered statistically significant.

3 RESULTS 

3.1 Baseline characteristics of the participants

The randomized participants at baseline were allocated to the test (n = 51) or placebo (n = 49) groups. Six participants of the test group were removed from the analysis: four with consent to withdrawal and two due to protocol violation. In the placebo group, seven participants withdrew from the study. Hence, the per-protocol population included a total of 87 participants, including 45 and 42 participants in the test and placebo groups, respectively.

The safety population included 100 patients who consumed the test product or placebo at least once. The baseline clinical characteristics of the participants are listed in Table 2. No statistically significant differences were found between the groups regarding age, sex, height, and weight. Moreover, no statistically significant differences in lifestyles (e.g., smoking, alcohol, outdoor activity time, daily sleeping hours, and frequency of using sunblock) were observed between the groups.

TABLE 2. Baseline clinical characteristics of the per-protocol study population.

Characteristics	Test group (n = 45) Mean ± SD or n (%)	Placebo group (n = 42) Mean ± SD or n (%)	p-value
Sex
Male	12 (26.67)	12 (28.57)	0.84a
Female	33 (73.33)	30 (71.43)
Age (years)	45.27 ± 6.36	43.71 ± 6.0	0.25b
Height (cm)	163.03 ± 6.87	163.57 ± 7.14	0.72b
Weight (kg)	59.72 ± 10.51	61.91 ± 11.63	0.47b
Smoking status
Never smoker	42 (93.33)	41 (97.62)	0.62a
Quit (≥1 year)	3 (6.67)	1 (2.38)
Quit (<1 year)	0 (0.0)	0 (0.0)
Current smoker	0 (0.0)	0 (0.0)

Abbreviation: SD, standard deviation.
a p-values were determined by chi-square or Fisher's exact tests.
b p-values were determined by two-sample t or Wilcoxon rank sum tests.

3.2 Effect on skin wrinkling

The skin wrinkling parameters—average skin roughness (Ra), maximum of all peak-to-valley values (Rmax), the maximum peak height of the wrinkle (Rp), and average maximum height of the wrinkle (Rz)—measured using the PRIMOS optical system were found to be considerably improved in the test group compared with the placebo group at 12 weeks (p < 0.0001 for all parameters; Figure 1, Table 3). The starting level of skin wrinkling was similar between the two groups. In the test group, Ra, Rmax, Rp, and Rz considerably improved from baseline to 12 weeks (p < 0.0001 for all parameters), whereas they remained unchanged or tended to worsen in the placebo group during the same period. In the test group, the maximum valley depth of the wrinkle (Rv) also tended to be reduced at 12 weeks compared with the baseline value (p = 0.6527), whereas in the placebo group tended to increase (p = 0.4625; Table 3).

 

FIGURE 1 

Normalized changes in skin wrinkle parameters of participants in the per-protocol study population. (A) Average skin roughness (Ra), (B) maximum of all peak-to-valley values (Rmax), (C) maximum peak height of the wrinkle (Rp), and (D) average maximum height of the wrinkle (Rz). Changes in parameter values from baseline are shown in micrometers (μm). (E) Eye wrinkle volume shown in cubic millimeters (mm 3 ). Data are expressed as mean ± standard deviation. *p < 0.05, **p < 0.001, and ***p < 0.0001 between groups as determined by two-sample t or Wilcoxon rank sum tests. 

TABLE 3. Skin wrinkling parameters of the per-protocol study population.

Parameter	Time of visit	Test group		Placebo group		Intergroup comparison p- value a
		Mean (SD)	p- value a	Mean (SD)	p- value a
Ra (μm)	Baseline	24.55 (4.49)		24.22 (5.67)		0.4497
	4 weeks	24.06 (4.18)	<0.0001	24.22 (5.51)	0.9828	0.0008
	8 weeks	23.61 (4.18)	<0.0001	24.47 (6.03)	0.0563	<0.0001
	12 weeks	23.31 (4.20)	<0.0001	24.63 (6.01)	0.0116	<0.0001
Rmax (μm)	Baseline	317.67 (82.23)		296.93 (101.04)		0.0794
	4 weeks	308.55 (74.34)	0.0019	298.28 (103.89)	0.6021	0.0455

Note: Skin wrinkle parameters: Ra, average skin roughness; Rmax, maximum of all peak-to-valley values; Rp, the maximum peak height of the wrinkle; Rv, maximum valley depth of the wrinkle; Rz, the average maximum height of the wrinkle.
a Intra-group comparisons; p-values were determined by paired t or Wilcoxon signed-rank tests.
b Intergroup comparisons; p-values were determined by two-sample t or Wilcoxon rank sum tests.

After 12 weeks, the eye wrinkle volume in the test group was reduced remarkably compared to the baseline (from 19.01 ± 7.23 to 16.40 ± 6.66 mm 3 , p < 0.0001), whereas the parameter in the placebo group showed an increment (from 17.19 ± 6.58 to 17.67 ± 6.94 mm 3 , p = 0.3449). Furthermore, comparison of the changes from baseline in the eye wrinkle volume between the two groups showed statistically significant differences (p < 0.0001; Figure 1, Table 3).

3.3 Effect on skin hydration

In the test group, skin hydration was greatly enhanced at 12 weeks compared with baseline values regarding the skin of the forehead (from 56.15 ± 8.47 to 60.18 ± 8.49), forearms (from 34.15 ± 7.12 to 36.74 ± 6.96), cheeks (from 42.57 ± 6.43 to 46.47 ± 6.95), and below the eyes (from 59.67 ± 8.05 to 63.71 ± 8.21) (p < 0.0001 for all assessed area). No significant differences in skin hydration levels were observed between the test and placebo groups at baseline. Notably, skin hydration changes from baseline were much greater in the test group than in the placebo group at 4, 8, and 12 weeks (p < 0.0001; Figure 2).

FIGURE 2

Normalized changes in skin hydration in the per-protocol study population. Hydration parameters were determined in the (A) forehead, (B) both forearms (average values are shown), (C) both cheeks (average values are shown), and (D) 1 cm below both eyes (average values are shown). Changes in parameter values from baseline are shown in arbitrary units. Data are expressed as the mean ± standard deviation. *p < 0.05, **p < 0.001, and ***p < 0.0001 between groups as determined by two-sample t or Wilcoxon rank sum tests.

3.4 Effect on skin elasticity

The skin overall (R2), net (R5), and biological (R7) elasticities showed greater improvements in the test group compared with the placebo group at 12 weeks (p < 0.0001 for all parameters; Figure 3, Table 4). However, the change from the baseline of the R5 value was statistically significant only at 12 weeks (p = 0.0032), whereas that of R2 and R7 was statistically significant from Weeks 4 (p < 0.0001) and 8 (p = 0.018), respectively.

FIGURE 3

Normalized changes in skin elasticity parameters of the per-protocol study population. (A) Overall elasticity (R2), (B) net elasticity (R5), and (C) biological elasticity (R7). Changes in parameter values from baseline are shown in arbitrary units. Data are expressed as mean ± standard deviation. *p < 0.05, **p < 0.001, and ***p < 0.0001 between groups as determined by two-sample t or Wilcoxon rank sum tests.

TABLE 4. Skin elasticity parameters of the per-protocol study population.

Parameter (μm)	Time of visit	Test group		Placebo group		Intergroup comparison p- value b
		Mean (SD)	p- value a	Mean (SD)	p- value a
R2	Baseline	0.69 (0.07)		0.69 (0.07)		0.9677
	4 weeks	0.71 (0.07)	<0.0001	0.69 (0.07)	0.0136	0.0003
	8 weeks	0.72 (0.07)	<0.0001	0.69 (0.07)	0.1211	<0.0001
	12 weeks	0.74 (0.06)	<0.0001	0.70 (0.07)	0.0182	<0.0001
R5	Baseline	0.54 (0.10)		0.53 (0.09)		0.8551
	4 weeks	0.54 (0.10)	0.3768	0.53 (0.09)	0.8024	0.4448
	8 weeks	0.53 (0.10)	0.7292	0.51 (0.10)	0.0009	0.0124

representative of net elasticity; and R7, ratio of elastic recovery to total deformation, representative of biological elasticity.
a. Intra-group comparisons; p-values were determined by paired t or Wilcoxon signed-rank tests.
b. Intergroup comparisons; p-values were determined by two-sample t or Wilcoxon rank sum tests.

3.5 Effect on skin whitening

Skin-whitening parameters in the test group showed much greater changes during the treatment compared with those of the placebo group (Figure 4). No significant differences in skin-whitening levels were observed between the groups at baseline. Notably, the melanin index in the test group was lower at 12 weeks compared with baseline (from 100.81 ± 37.51 to 96.92 ± 38.55, p = 0.0004), whereas that of the placebo group increased (from 97.98 ± 32.18 to 101.85 ± 31.59, p = 0.0282). Moreover, the erythema index in the test group was considerably lower at 12 weeks compared with the baseline levels (from 340.50 ± 64.34 to 301.37 ± 68.54, p < 0.0001) and the placebo group also showed reduced erythema index at 12 weeks compared to baseline (from 327.77 ± 65.71 to 308.29 ± 61.06, p < 0.0001). Comparison of the changes from baseline in the erythema index between the two groups revealed statistically significant differences (p = 0.0002).

FIGURE 4

Normalized changes in skin-whitening parameters in the per-protocol study population. (A) Melanin index and (B) erythema index. Changes in parameter values from baseline are shown in arbitrary units. Data are expressed as mean ± standard deviation. *p < 0.05, **p < 0.001, and ***p < 0.0001 between groups as determined by two-sample t or Wilcoxon rank sum tests.

statistically analyzed, no significant variances were found between the two groups. Additionally, none of the participants experienced any adverse effects during the treatment period.

4 DISCUSSION

Low-molecular-weight collagen peptides are highly absorbable and favorably affect the quality and quantity of dermal component of the skin. 12, 13 Several in vivo studies revealed the underlying mechanisms of oral collagen supplementation. In particular, this therapeutic approach reduces the expression and activation of MMPs (specifically MMP-1, MMP-2, MMP- 3, MMP-9, and MMP-12) following ultraviolet B radiation exposure, which subsequently leads to increased synthesis of collagen and elastic fibers, as well as reduced collagen fragmentation. 15-18 In addition, reduced MMP-1 expression and higher levels of the tissue inhibitor of MMP-1 are believed to be involved in the antiaging effects of collagen peptides. 15-18 Further, bioactive collagen peptides, mainly consisting of prolyl- hydroxyproline, stimulate chemotaxis and proliferation of dermal fibroblasts, thereby enabling the production of dermal components: hyaluronic acids, elastic fibers, and collagen. 7, 19-21 The findings of the present study agree with these previous reports. Overall, oral consumption with low-molecular-weight collagen peptides for 12 weeks improved skin elasticity and reduced skin wrinkles compared with a placebo. Nonetheless, improvements in gross elasticity (R2), biological elasticity (R7), and from 8 weeks, net elasticity (R5) were noticeable at different time points (from Weeks 4, 8, and 12, respectively). The R5 parameter is commonly preferred for quantifying skin aging since it reflects the ability of the skin to recover after deformation and skin roughness without being affected by skin thickness, whereas R2 and R7 represent the elasticity of the skin and its ability to recover its original position, which is highly correlated with skin aging. 22 Therefore, the present findings suggest that improved R2 and R7 may reflect the early increase of dermal thickness, whereas enhancement of the net elasticity (R5) requires more time to be established.

As well as enhancing the skin antiaging properties, administration of low-molecular-weight collagen peptides was also found to impact skin hydration. As collagen peptides increase the production of hyaluronic acid, this water-binding glycosaminoglycan supports the hydrating effect. 18, 23 To date, the impact of collagen peptides on TEWL changes remains controversial. 11, 13, 16, 18 The exact mechanism and effect of collagen peptides on the epidermal barrier via TEWL remain to be elucidated.

Analysis of skin-whitening parameters (melanin and erythema indexes) revealed that they had higher values in the test group than in the placebo group, which indicated brighter and healthier skin after low-molecular-weight collagen peptides consumption. Limited information is available on the skin-whitening effect of collagen peptides; nonetheless, peptides can prevent ultraviolet B-induced photoaging via anti-inflammatory, antioxidant, and anti-melanogenesis activities in melanocytes and skin fibroblasts in hairless mice, 25 which correlated with our findings. It is inspiring that this clinical trial presented a significant decrement in both melanin and erythema index. These findings suggest that collagen peptides may be a sustainable and safe adjuvant supplement for other dermatologic diseases, such as Riehl's melanosis, melasma, and rosacea.

The present study has some limitations. First, a follow-up assessment was not conducted after the test product was discontinued. Second, although each skin parameter was evaluated via objective measurements, this is still a qualitative study and is not translatable to clinical appearance or other measuring tools. Third, this study was conducted using a product instead of a pure ingredient. Moreover, the establishment of a dose-dependent efficacy remains elusive, given our sole provision of 2000 mg of low-molecular-weight collagen within a 2.5 g product. Previous clinical studies have employed doses of 0.5–10 g of collagen peptides, all demonstrating favorable effects on skin elasticity without notable adverse outcomes. 4, 12, 13 The translated values from in vivo studies ranged from 480 to 1620 mg/day, which enhanced skin dryness and wrinkle reduction. 4 However, the optimal dosage of collagen supplementation warrants further investigation. Lastly, the study only enrolled participants from specific regions in Korea, and of specific sex and age groups.

Taken together, this randomized, double-blinded, placebo-controlled study demonstrates that oral supplementation of low-molecular-weight collagen peptides can be used to sustainably improve the physiologic properties of the skin, including wrinkles, elasticity, hydration, melanin index, and erythema index, without any significant adverse effects. Further studies are needed to clarify the exact therapeutic mechanisms by which this nutraceutical can act on a wide range of dermatologic diseases.

AUTHOR CONTRIBUTIONS

Conceptualization, Jangmi Suk, Young In Lee, and Ju Hee Lee; methodology, In Ah Kim, Inhee Jung, and Chaemin Baeg; formal analysis, Seol Hwa Seong and Ju Hee Lee; investigation, Jinhak Kim, Dongchan Oh, and Jangmi Suk; writing—original draft preparation, Seol Hwa Seong, Jinhak Kim and Dongchan Oh; writing—review and editing, Joohee Lee and Sooyeon Choi and supervision, Jinhak Kim, Dongchan Oh and Ju Hee Lee. All authors have read and agreed to the published version of the manuscript.

ACKNOWLEDGMENTS

The low-molecular-weight collagen peptides were prepared by Daehan Chemtech Co., Ltd., to which two authors are affiliated.

CONFLICT OF INTEREST STATEMENT

All authors declare no competing interests.

ETHICS STATEMENTS

The Institutional Review Board (IRB) of the Global Medical Research Center approved this study (IRB no.: GIRB-22318-HA). The study was conducted in full compliance with the principles of Good Clinical Practices and Declaration of Helsinki. Informed consent was obtained from all subjects involved in the study.