« BackJournal of Dermatological Science
Article in Press

The effect of a low glycemic load diet on acne vulgaris and the fatty acid composition of skin surface triglycerides☆

  • Robyn N. Smith

      Affiliations

    • School of Applied Sciences, RMIT University, GPO Box 2476V, Melbourne, Victoria 3001, Australia
    • Corresponding Author InformationCorresponding author. Tel.: +61 3 9925 2720; fax: +61 3 9925 5241.
    • ,
  • Anna Braue

      Affiliations

    • Department of Dermatology, Royal Melbourne Hospital, Parkville, Australia
    • ,
  • George A. Varigos

      Affiliations

    • Department of Dermatology, Royal Melbourne Hospital, Parkville, Australia
    • Department of Dermatology, Royal Children's Hospital, Parkville, Australia
    • ,
  • Neil J. Mann

      Affiliations

    • School of Applied Sciences, RMIT University, GPO Box 2476V, Melbourne, Victoria 3001, Australia
    • Australian Technology Network, Centre for Metabolic Fitness, Australia

Received 7 May 2007; received in revised form 17 September 2007; accepted 7 November 2007. published online 12 July 2010.
Corrected Proof

Article Outline

Summary 

Background

Dietary factors have long been implicated in acne pathogenesis. It has recently been hypothesized that low glycemic load diets may influence sebum production based on the beneficial endocrine effects of these diets.

Objective

To determine the effect of a low glycemic load diet on acne and the fatty acid composition of skin surface triglycerides.

Methods

Thirty-one male acne patients (aged 15–25 years) completed sebum sampling tests as part of a larger 12-week, parallel design dietary intervention trial. The experimental treatment was a low glycemic load diet, comprised of 25% energy from protein and 45% from low glycemic index carbohydrates. In contrast, the control situation emphasized carbohydrate-dense foods without reference to the glycemic index. Acne lesion counts were assessed during monthly visits. At baseline and 12-weeks, the follicular sebum outflow and composition of skin surface triglycerides were assessed using lipid absorbent tapes.

Results

At 12 weeks, subjects on the experimental diet demonstrated increases in the ratio of saturated to monounsaturated fatty acids of skin surface triglycerides when compared to controls [5.3±2.0% (mean±S.E.M.) vs. −2.7±1.7%, P=0.007]. The increase in the saturated/monounsaturated ratio correlated with acne lesion counts(r=−0.39, P=0.03). Increased follicular sebum outflow was also associated with an increase in the proportion of monounsaturated fatty acids in sebum (r=0.49, P=0.006).

Conclusion

This suggests a possible role of desaturase enzymes in sebaceous lipogenesis and the clinical manifestation of acne. However, further work is needed to clarify the underlying role of diet in sebum gland physiology.

Keywords: Acne vulgaris, Dietary glycemic load, Sebum, Skin surface triglycerides

 

Back to Article Outline

1. Introduction 

Dietary factors have long been implicated in the pathogenesis of acne [1], [2]. It is well known that increased sebum production plays a fundamental role in acne [3] and evidence suggests that dietary manipulation alters sebaceous gland output. Extreme caloric restriction dramatically decreases the sebum excretion rate and these changes can be reversed when a normal diet is resumed [4], [5]. Other studies have demonstrated that increased consumption of dietary fat or carbohydrate increases sebum production [6], and modifications to the type of carbohydrate can also alter sebum composition [7], [8]. Altogether, these studies suggest that the quantity and composition of foods, when changed significantly, may affect underlying mechanisms involved in sebum production.

Evidence suggests that diet may be an important source of substrate for the synthesis of sebaceous lipids [1]. Human sebum is comprised mainly of triglycerides (40–60%), wax esters (19–26%) and squalene (11–15%), with some cholesterol and cholesterol esters [9], [10], [11]. These lipids can be synthesized from a variety of sources (e.g. glucose, acetate, and fatty acids) which serve to donate two carbon fragments [12], [13]. However, some dietary lipids (especially fatty acids) can also pass unchanged from the circulation to the sebaceous cells. It is presumed that undifferentiated cells of the sebaceous gland acquire the dietary lipids whilst in the basal layer exposed to the circulation [14]. This notion is supported by the observation that sebum contains linoleic acid, an essential fatty acid that cannot be synthesized in vivo and therefore must be obtained from the diet.

The triglyceride fraction of sebum is presumably responsible for acne development [15], [16]. Bacteria can hydrolyze sebaceous triglycerides [17], liberating the fatty acids which can penetrate the follicular wall and become incorporated into the metabolism of the surrounding epidermis. The application of free fatty acids on rabbit ears or hairless mice has been shown to induce hyperkeratinization and epidermal hyperplasia similar to that seen in comedo formation [15], [18]. However, the hyperkeratotic effect may not be a feature of all fatty acids, as recent evidence suggests that only monounsaturated fatty acids (MUFAs) stimulated the morphological changes whereas saturated fatty acids (SFAs) have little effect [15]. Human sebum is known to contain a high proportion of MUFAs, with a characteristic double bond at the Δ6 position rather than at the standard Δ9 position [16]. The most abundant of these is sapienic acid (16:1Δ6), which is formed by the Δ6 desaturation of palmitic acid (16:0) [19]. Sapienic acid is unique to human sebum and has not been identified in other human tissues or in sebaceous gland secretions of other animals [16]. It is presumed that this fatty acid may play a role in acne pathogenesis, however its role is not well defined [20].

In the present study, we examined the influence of diet on the fatty acid composition of skin surface triglycerides. Recent evidence suggests that low glycemic load diets may affect sebum production based on the beneficial hormonal effects of these diets [21]. The glycemic load may be interpreted as a measure of the blood glucose and insulin-raising potential of the diet, as it represents the rate of carbohydrate absorption (indicated by the glycemic index) and the quantity of carbohydrate consumed [22]. Previous studies indicate that the dietary manipulation of the quality and quantity of carbohydrates can affect the composition of fatty acids in sebum [7], [8]. A relative excess of dietary carbohydrate (500g/day) can increase the proportion of 16:1 in sebum, however the effect on the other fatty acids of sebum is varied depending on the type of carbohydrate used [7], [8]. Based on these observations, one can speculate that the composition of fatty acids in sebum may vary with alterations in the dietary glycemic load. Therefore, the objective of the present study was to determine the effect of a low glycemic load diet on acne and the fatty acid composition of skin surface triglycerides.

Back to Article Outline

2. Materials and methods 

2.1. Study population 

Male volunteers with acne were recruited for a dietary intervention study that was conducted at RMIT University (Melbourne, Australia). Informed consent was obtained from participants and guardians (if aged <18years) and the study had the approval of the RMIT Human Ethics committee. This study included only male participants, aged 15–25 with mild-moderate facial acne. Participants were required to have had acne for longer than 6 months prior to recruitment. Volunteers were excluded if they were taking medications known to affect acne or glucose metabolism. A wash-out period of 6-months was required for oral retinoids or 2-months for oral antibiotics or topical therapy.

2.2. Study design 

Eligible participants were recruited between June 2003 and June 2004. Participants were randomly assigned to either the low glycemic load (LGL) or control group (see Fig. 1) at the time of their baseline appointment. Randomization was carried out by computer generated random numbers and allocation to groups was performed by a third party.

The experimental protocol followed a parallel, dietary intervention with monthly visits at an academic research clinic (weeks 0, 4, 8, and 12). At baseline and 12-weeks, participants also sat a 1-h sebum sampling test. Due to scheduling interferences (e.g. school, work commitments), not all of the study participants were available to sit the 1-h sebum sampling test. Fig. 1 shows the proportion of participants who completed the sebum test as per protocol.

2.3. Dietary intervention 

Participants were informed that the study's intent was to compare the dietary carbohydrate to protein ratio and were not informed of the study's true hypothesis. The LGL diet was low in glycemic load, achieved through a reduction in carbohydrate intake and by reducing the dietary glycemic index. The LGL group was educated on how to substitute high glycemic index foods with foods higher in protein (e.g. lean meat, poultry or fish) and lower in glycemic index (e.g. whole grain bread, pasta, fruits). Some staple foods were supplied and participants were urged to consume these or similar foods on a daily basis. Each participant's dietary directions were isocalorically matched with their baseline diet as determined from 7-day weighed/measured food records. The recommended LGL diet consisted of 25% energy from protein, 45% from low GI carbohydrates and 30% energy from fats. In contrast, the control group received carbohydrate-dense staples and were instructed to eat these or similar foods daily. The foods provided had moderate to high glycemic index values and were typical of their normal diet as evidenced from 7-day weighed/measured food records. The control group were not informed about the glycemic index, but were urged to include carbohydrates as a regular part of their diet. Actual dietary intakes, as assessed from weighed/measured food records, have been reported elsewhere [24].

2.4. Standardized topical cleanser 

All subjects were provided with a topical cleanser (Cetaphil® gentle skin cleanser, Galderma, Forrests Hill, Australia) and were advised to use it in place of their normal wash, soap or cleanser. The cleanser provided was non-comedogenic and contained no active agents for acne. Subjects began using the topical cleanser 2 weeks prior to baseline and were asked to maintain a standard level of usage during the trial.

2.5. Dermatology assessment 

Scaling of the acne was performed by a dermatology registrar who was blinded to the group assignment of the participants. The registrar assessed facial acne occurrence and severity only, using a modified Cunliffe-Leeds lesion count technique [24].

2.6. Sebutape sample collection method 

Skin surface lipids were collected using Sebutape® test strips (CuDerm Corporation, Dallas, USA) at baseline and 12-weeks. Sebutape® is a lipid absorbent strip that has been shown to be a reproducible and non-invasive method for estimating the output of sebum from active follicles [25].

Prior to the application of the Sebutape® adhesive strips, residual surface lipid was removed from the forehead using an isopropyl alcohol swab. For sampling, two lipid absorbing strips were applied to the forehead for a 1h collection, using disposable gloves and alcohol-rinsed forceps to prevent lipid-contamination. The forehead was chosen as the site of collection as the contribution of epidermal lipids to the total composition of skin surface lipids has been shown to be minimal (less than 10μg/cm2 compared with average recoveries of 150–300μg/cm2 for sebum) [11]. During the sampling time, the room temperature was kept between 18 and 21°C.

Following collection, one test strip was applied to storage card (supplied by CuDerm) for photometric analysis and the second strip was placed in a chloroform–methanol rinsed, teflon-capped glass vial. The glass vial was flushed with nitrogen and then frozen at −80°C for later compositional analysis.

2.7. Photometric assessment of the follicular sebum outflow 

High resolution photographs were taken of the test strips against a black background using a digital camera. Photographs were converted to black and white binary images for quantitative analysis using Image Tool Software (Version 2 alpha, University of Texas Health Science Center, San Antonio, Texas). This software was used to quantify the total spot area (number of black pixels) as a percentage of the total test area (total pixelated area). Sebum outflow was determined as the percentage lipid area (percentage of black area of the whole test area) on Sebutape® strips following a 1h collection.

2.8. Patient's assessment of skin oiliness 

At baseline and 12-weeks, each participant was asked to rate their facial skin oiliness using a 7-point scale. Reponses were scored using a scale from 0 (‘not at all’) to 6 (‘extremely’). This question was a component of an acne-specific questionnaire relating to acne symptoms and quality of life [26].

2.9. Analysis of the fatty acid composition of skin surface triglycerides 

Lipids from the test strip were extracted as previously described [25]. Triglycerides were isolated by preparative thin layer chromatography using silica 60G plates that had been heated in an oven at 100°C and cooled to room temperature in a dessicator. A reference standard of cholesterol, cholesterol oleate, oleate, methyl oleate, triolein, palmityl oleate and squalene (Nu Chek Prep Inc. Elysian, Minnesota and Sigma chemical Co Missouri) was used to identify the triglycerides and wax ester fractions. After spotting the samples, the plates were developed according to the method of Doran et al. [27].

Lipids were visualized on the plates under ultraviolet light after spraying with a methanoic solution of 2′,7′-dichlorofluorescein. The bands containing wax esters and triglycerides were isolated and saponified according to the method of Sinclair et al. [28]. The fatty acid methyl esters (FAME) were analyzed by gas chromatography using a Shimadzu GC-17A chromatograph, equipped with a flame ionization detector and integrated software. Each sample was injected onto a 60m BPX-70 (0.32mm internal diameter and 0.25mm film thickness) bonded phase, fused silica capillary column (SGE, Ringwood, Victoria, Australia) and samples were run according to Sinclair et al. [28] using a splitless injection technique. A commercial standard of C14:0, C16:0, C16:1D9, C18:0, C18:1Δ9, C18:2Δ9,12, C18:3Δ9,12,15, C20:0 (Nu Chek Prep Inc. Elysian, Minnesota) was used for identification of sample FAMEs. Isolated sebaceous wax esters served as a source of identification of 16:1Δ6 [25].

2.10. Statistical analysis 

All statistical analyses were performed using SPSS 11.0 for Windows (SPSS Inc., Chicago, Illinois). Data conforming to a normal distribution were analyzed using an independent t test. We analyzed ordinal and non-normally distributed data using Mann–Whitney U test for independent groups and the Wilcoxon signed rank test for paired data. Changes from baseline are reported in percentages, with statistical analyses done for absolute values. We compared changes in lesion counts using analysis of covariance of log-transformed data with baseline lesion counts as the covariate. Bivariate linear regression analysis was also conducted, pooling data from both groups, to explore relationships between outcome variables. P-values less than 0.05 were considered significant. As the primary endpoint for this paper was the fatty acid composition of skin surface triglycerides, the data was analyzed per protocol.

Back to Article Outline

3. Results 

3.1. Subjects 

Fig. 1 shows the trial profile. Fifty-four subjects were recruited for the parallel, dietary intervention study. Seven participants did not complete the study (5 in CO and 2 in LGL), 4 were removed from data set (2 began acne medications and 2 were non-compliant) and 12 were unable to participate in the sebum test. Thirty-one subjects completed the trial and sebum test as per protocol.

3.2. Study outcomes 

Table 1 shows the clinical characteristics of the subjects who completed the sebum test as per protocol. Baseline characteristics according to group allocation were not significantly different for total (P=0.15), inflammatory (P=0.44) and non-inflammatory (P=0.45) lesion counts. At 12 weeks, a greater improvement in total lesion counts was observed in the LGL group compared to controls after adjusting for baseline differences. Total acne lesion counts decreased by 59% in the LGL group and by 38% in the control group (P=0.046). However, there were no significant group differences with regards to changes in inflammatory lesions. The LGL group also showed reductions in weight (P<0.001) and BMI (P<0.001) when compared to the control group (Table 1).

Table 1. Clinical outcomes according to dietary group for subjects who completed sebum test as per protocol
Baselinen12 weeksnChange from baselinea (%)P
Total acne lesions
LGL48.9±5.8b1618.9±2.816−59.4±4.60.046
Control38.7±5.11524.9±4.915−37.8±8.8

Inflammatory lesions
LGL38.1±4.71616.7±2.716−50.0±9.10.10
Control32.1±4.31522.6±4.315−30.3±10.2

Weight (kg)
LGL74.1±2.91670.5±2.416−4.4±1.0<0.001
Control72.6±4.01573.4±4.1151.0±0.5

BMI (kg/m2)
LGL23.0±0.51621.9±0.416−4.7±1.1<0.001
Control22.2±0.81522.3±0.8150.5±0.6

aChanges from baseline are reported in percentages, with statistical analyses done for absolute values. Lesion count data were analyzed using ANCOVA on log transformed data with baseline counts as the covariate. Body composition data was analyzed using Mann–Whitney.

bResults expressed as mean±SEM. Analysis of baseline data revealed no significant group differences for listed variables.

Fig. 2, Fig. 3 illustrate the changing pattern of increasing sebum output and the follicular sebum outflow as determined from the percentage of the lipid impregnated area of test strips. The follicular sebum outflow did not differ between groups at baseline (P=0.07) and did not change following 12-weeks of dietary intervention. However, there was a significant decline in the reported oiliness of the skin at 12 weeks in the LGL group (Fig. 4). The mean response score of skin oiliness decreased from baseline in the LGL group (P=0.013), whereas no significant change was observed in the control group.

  • View full-size image.
  • Fig. 4. 

    Mean response score of skin oiliness (95% CI) according to dietary group at baseline and follow-up. Note: a lower score reflects a decline in reported skin oiliness. *Follow-up score significantly different from baseline P<0.05.

The relative amounts of the major fatty acids in the triglyceride fraction of sebum are summarized in Table 2. No significant group differences were observed for the change in the proportion of individual fatty acids in sebum. However, the two treatment groups demonstrated opposing trends for changes in the SFAs/MUFAs ratio. The SFAs/MUFAs ratio significantly increased in the LGL group (P=0.019) and the change at 12 weeks was significantly different from that of controls (5.3±2.0% for the LGL group vs. −2.7±1.7% for the control group, P=0.007). Similarly, the 16:0/16:1Δ6+Δ9 ratio increased in the LGL group compared to the control group (4.5±2.3% for the LGL group vs. −5.5±2.2% for the control group, P=0.004).

Table 2. Weights percent of the major fatty acids in the triglyceride fraction of sebum according to dietary group
Fatty acidBaselinen12 weeksnChange from baselinea (%)Pb
14:0 (wt% of total)
LGL9.22±0.49c169.34±0.35162.9±2.90.75
Control9.63±0.62159.62±0.6415−0.07±2.4

14:1 (wt% of total)
LGL2.29±0.18162.24±0.1716−3.9±6.20.55
Control2.38±0.11152.46±0.12154.3±4.3

15:0 (wt% of total)
LGL6.20±0.32166.49±0.31165.4±2.60.71
Control6.75±0.28156.95±0.40152.3±2.8

16:0 (wt% of total)
LGL23.3±0.51623.9±0.5162.9±2.10.07
Control24.3±0.41523.6±0.615−2.4±1.7

16:1Δ6+Δ9 (wt% of total)
LGL19.3±0.61619.0±0.616−0.7±3.30.23
Control19.2±0.61519.8±0.5153.8±2.4

18:0 (wt% of total)
LGL2.85±0.21162.63±0.1316−3.4±3.80.34
Control2.65±0.15152.62±0.2115−0.5±6.7

18:2 Δ9,12 (wt% of total)
LGL0.52±0.06150.43±0.0315−8.9±6.90.65
Control0.43±0.04140.42±0.06143.7±15.2

20:0 (wt% of total)
LGL0.53±0.05160.51±0.0416−2.1±6.20.22
Control0.46±0.05150.47±0.05154.5±6.3

Total saturated (wt% of total)
LGL44.8±0.91645.6±0.6162.2±1.40.09
Control46.3±0.81545.7±0.915−1.4±1.1

Total monounsaturated (wt% of total)d
LGL32.8±1.21631.8±1.016−2.3±2.60.14
Control32.0±1.01532.4±0.9151.7±1.6

aChanges from baseline are reported in percentages, with statistical analyses done for absolute values.

bP-value corresponds with an independent-sample t test or Mann–Whitney for changes in absolute data.

cResults are expressed as mean±S.E.M. An independent-sample t test revealed no significant group differences for listed fatty acids at baseline.

dIncludes 14:1, 16:1 and 18:1 monounsaturated series.

3.2.1. Sebaceous fatty acids associated with acne improvement and sebum output 

We found that an increase in the SFAs/MUFAs ratio was predictive of the clinical improvement in acne (Fig. 5). Furthermore, an increase in the SFAs/MUFAs was associated with a decline in the follicular sebum outflow (Fig. 6). In light of this, we explored the improvement in acne in terms of changes in sebum output and found that the relationship was not significant (P=0.70). However, the sebum outflow was found to be a function of the increasing proportion of MUFAs in sebum (Fig. 6).

  • View full-size image.
  • Fig. 5. 

    (a) Change in acne lesion count as a function of the change in the ratio of SFAs/MUFAs in sebaceous triglycerides. (b) Change in acne lesion count as a function of the change in the ratio 16:0/16:1Δ6+Δ9 in sebaceous triglycerides.

  • View full-size image.
  • Fig. 6. 

    (a) Change in the quantity of skin surface lipid as a function of the change in the MUFAs in sebaceous triglycerides. (b) Change in the quantity of skin surface lipid as a function of the change in the ratio of SFAs/MUFAs.

Back to Article Outline

4. Discussion 

Increased sebum production is an obligatory condition for developing acne. Support for an association between acne and sebum production comes from three lines of evidence: (i) acne patients have higher rates of sebum secretion than controls [3], [29]; (ii) children have low sebum output and do not get acne; and (iii) sebum-suppressive agents alleviate acne symptoms [30], [31]. Although the role of sebum in acne remains to be defined, there is suggestive evidence that compositional changes, which occur with increasing sebum production, may influence events involved in comedo formation [14]. Therefore, profiling the fatty acids of sebum may contribute to our understanding of events involved in acne pathogenesis.

In the present study, we examined the influence of a LGL diet on clinical assessments of acne and the composition of skin surface triglycerides. After 12 weeks, we found that the subjects on a LGL diet demonstrated significantly greater reductions in total lesion counts when compared with controls. Although we were unable to detect an effect of dietary intervention on sebum output or the composition of individual fatty acids, we did observe opposing trends in the SFAs/MUFAs ratio of skin surface triglycerides. Subjects on the LGL diet demonstrated an increase in the SFAs/MUFAs ratio compared to a decrease seen in the control group. The LGL group also demonstrated an increase in the 16:0/16:1Δ6+Δ9 ratio, thereby suggesting a decrease in the enzymatic desaturation of 16:0 with a LGL diet. Interestingly, these changes were found to correlate with the clinical improvement in acne, thereby suggesting that the desaturation of sebaceous fatty acids may play a role in acne development.

The enzyme, Δ6-desaturase, is responsible for converting palmitic acid (16:0) into the monounsaturated fatty acid sapienic acid (16:1Δ6) in a unique type of reaction that is characteristic of human sebaceous glands. In sebaceous secretions of other species, the pattern of desaturation occurs at the Δ9 carbon rather than at the Δ6 position [32]. Human sebaceous glands do not express Δ9 desaturase [20], however accumulation of some Δ9 forms may occur in undifferentiated cells in the basal layer of the glands. As cells differentiate and move into the unvascularized interior of the gland, the Δ9 fatty acids initially acquired by cells may be diluted by the subsequent synthesis of Δ6 fatty acids. In adult sebum, sapienic acid accounts for around 25% of the total fatty acids and the contribution of 16:1Δ9 is less than 0.5% [16]. In the present study, we were unable to resolve the Δ6 and Δ9 isomers by gas chromatography, however we suspect that basal levels of 16:1Δ9 would be fairly constant and therefore any change in 16:1Δ6+Δ9 is most likely to reflect a change in the endogenous production of 16:1Δ6.

The expression of Δ6-desaturase and the resultant accumulation of sapienic acid in sebum may be an important factor in sebaceous lipogenesis. Support for this role comes from evidence of sebaceous gland hypoplasia in asebia mice which fail to express Δ9 desaturase [33]. Progenitor cells of the human sebaceous gland do not express Δ6-desaturase, however this enzyme is highly expressed in lipid-containing cells situated one cell layer away [20]. This pattern of expression supports a role of Δ6-desaturase in sebaceous lipogenesis and it has recently been proposed that sapienic acid production may provide a functional marker of sebaceous gland activity [20]. Our study found that a higher follicular sebum outflow was associated with an increase in the proportion of MUFAs in sebum, which was largely explained by variations in 16:1Δ6+Δ9 (data not shown). This observation supports previous reports of high sapienic acid levels in conditions of increased sebum production (i.e. puberty and acne) [34], [35]. We also observed that increased sebum outflow was associated with a decrease in the SFAs/MUFAs ratio, which supports the suggestion that desaturase activity plays a fundamental role in sebum production.

Sebum composition has also been implicated in the abnormal follicular keratinization which is associated with acne development. Katsuda et al. [15] recently demonstrated that fatty acid subtypes can have discrete effects on skin surface morphology and epidermal proliferation. The authors showed that the topical application of MUFAs (oleic acid and palmitoleic acid) induced scaly skin, abnormal keratinization and epidermal hyperplasia. In contrast, triglycerides (triolein) and SFAs (palmitic and stearic acid) had no affect on skin morphology. It is thought that the MUFAs of sebum interfere with the intracellular calcium dynamics of follicular keratinocytes and the intercellular lipid bilayer structure of the epidermal water barrier. Katsuda et al. have postulated that reducing sebum secretion and/or the production of free MUFAs may serve as potential targets for acne therapy. In support of this hypothesis, our study found that an increase in SFAs relative to MUFAs in sebum was predictive of the clinical improvement in acne. This relationship was largely explained by an increase in the 16:0/16:1Δ6+Δ9 ratio. Given that the LGL group demonstrated an increase in the 16:0/16:1Δ6+Δ9 ratio when compared to controls, this may serve to explain the greater acne improvement in the LGL group.

The precise mechanism by which dietary glycemic load influences the sebum composition is unknown. Given that sebaceous glands can synthesize lipids from a variety of precursors, one could assume that the supply of glucose is not a limiting factor. To synthesize lipids, sebaceous glands require energy, NADPH and acetyl-coenzyme A, which can be acquired through β-oxidation of fatty acids and/or the catabolism of glucose [36]. However, to sustain the normal pattern of sebaceous lipids, Downie and Kealey have hypothesized that endogenous glycogen may be an important provider of NADPH, as well as substrates (e.g. glycerophosphate and acetate), for the synthesis of triglycerides [36], [37]. Glycogen and glucose concentrations have been shown to decrease as cells accumulate lipid and move from the periphery to the centre of the glands [38], [39]. The glycogen content of undifferentiated sebaceous cells is also 6.3 times that of the epidermal keratinocytes [39]. Furthermore, glycogen stores and triglyceride synthesis are greatly reduced in isolated sebaceous glands, even in the presence of acetate, lactate and amino acid substrates, thereby suggesting that endogenous glycogen is essential for sebaceous triglyceride synthesis [38].

We have hypothesized that a LGL diet may affect sebum composition via metabolic effects (e.g. fuel storage) and/or secondary effects on hormone levels (e.g. free testosterone and adrenal androgens) (see Fig. 7). Accumulating evidence suggests that high glycemic load diets can increase glycogen storage within body tissues (i.e. muscle, liver) when compared to isoenergetic, low glycemic load diets [40], [41]. Therefore, it is possible that dietary manipulation of the glycemic load may affect glycogen stores in sebaceous glands, which may be a limiting and directing factor in sebaceous lipogenesis. Furthermore, we have shown that a LGL diet can reduce testosterone bioavailability and dehydroepiandrosterone sulfate concentrations [23], which may be explained by the insulin lowering effect of these diets [21]. As sebum production is largely under androgenic control, a reduction in circulating androgen levels may also account for the changes in sebum composition. This raises the possibility that nutritional intake may influence sebum production via the synchronized modulation of androgens and glycogen stores.

  • View full-size image.
  • Fig. 7. 

    Schematic illustrating the hypothesized metabolic-endocrine pathway linking a low glycemic load diet and the changes in the fatty acid profile of human sebum production. Abbreviations: SHBG, sex hormone binding globulin; DHEA-S, dehydroepiandrosterone sulphate.

There are some issues relating to this study that warrant further consideration. Firstly, it is possible that the relative changes in sebum composition could also be related to the change in body mass in the LGL group [42]. This weight loss occurred despite dietary instructions being given to maintain the subject's baseline kilojoule intake. There is increasing evidence to indicate that LGL diets enhance weight loss via increases in satiety and fat oxidation [40], [41], and studies suggest that obese adolescents lose weight on LGL diets without the need for an imposed energy restriction [43], [44]. This study found that the weight loss in the LGL group correlated with starting BMI, indicating that extent of obesity at baseline was a significant predictor of the overall weight loss (data not shown). Secondly, it is possible that use of the mild non-comedogenic skin cleanser may have assisted in the management of acne-associated symptoms [45]. Although the participants were stabilized on the mild cleanser two weeks prior to baseline, the acne improvement seen in controls suggests a possible therapeutic effect of the cleanser use. However, as the cleanser was standardized for both groups, this would not explain the differences observed between groups.

In conclusion, this study is the first to report the effect of a LGL diet on follicular sebum outflow and composition of skin surface triglycerides. We observed that after 12 weeks a LGL diet reduced weight and total lesion counts when compared to controls. The LGL group also demonstrated a significant increase in the ratio of SFAs/MUFAs, which was largely explained by alterations in the ratio of 16:0/16:1Δ6+Δ9. These changes suggest either a decrease in desaturase activity or the levels of the desaturase enzyme. As changes in the ratio of 16:0/16:1Δ6+Δ9 correlated with the clinical improvement in acne, the desaturation of 16:0 may have practical implications for studying the disease process. However, as the participants in the LGL group lost weight we are unable to isolate the independent effects of dietary composition from that of weight loss. Therefore, the role of diet in sebum composition is yet to be fully clarified and further studies are required to isolate the underlying mechanistic factors.

Back to Article Outline

Acknowledgements 

We would like to thank Henna Mäkeläinen, Leah Williamson and Nicole Fitzpatrick for their assistance with this project.

Back to Article Outline

Appendix A. Appendix: Examples of the types of foods recommended to subjects on the LGL and control diets and their respective glycemic index classificationa 

MealLGL dietControl diet
Food itemGIbFood itemGI
BreakfastNatural muesliLowCorn based cerealHigh
Bran flakesLowRice based cerealHigh
OatsLow-ModWheat biscuit cerealMod-High
Wholegrain breadLowWhite or wholemeal breadHigh
Milk, low fatLowMilk, low fatLow
Apple juiceLowOrange juiceLow-Mod
Peanut butter–HoneyLow-Mod

LunchWholegrain breadLowWhite or wholemeal breadHigh
Salad vegetables–Salad vegetables–
Tuna, lean meats and poultry–Cheese, low fat–

DinnerLean meat, poultry, fish–Calrose riceHigh
LegumesLowPotatoesHigh
Vegetables, excluding potatoesLowVegetables, excluding potatoesLow
Durum wheat pastaLow-ModLean meat, poultry, fish–
Basmati riceModSoft drinkMod
Apple juiceLow

SnacksFresh or tinned fruitLowMuesli barMod
YoghurtLowPotato crispsMod
Crispbread, wholegrainLowSweet biscuitsMod-High
NutsLowPuffed rice and corn cakesHigh
PopcornLowPretzelsHigh
Continental fruit breadLowEnglish muffinHigh

aGI of listed foods have been sourced from GI tables [46], [47], which used glucose as the reference food.

bGlycemic index classification: low <55; moderate 56–69; high >70.

Back to Article Outline

References 

  1. Rasmussen J. Diet and acne. Int J Dermatol. 1977;16:488–491
  2. Michaëlsson G. Diet and acne. Nutr Rev. 1981;39:104–106
  3. Harris H, Downing D, Stewart M, Strauss J. Sustainable rates of sebum secretion in acne patients and matched normal control subjects. J Am Acad Dermatol. 1983;8:200–203
  4. Downing D, Strauss J, Pochi P. Changes in skin surface lipid composition induced by severe caloric restriction in man. Am J Clin Nutr. 1972;25:365–367
  5. Pochi P, Downing D, Strauss J. Sebaceous gland response in man to prolonged total caloric deprivation. J Invest Dermatol. 1970;55:303–309
  6. Llewellyn A. Variations in the composition of the skin surface lipid associated with dietary carbohydrates. Proc Nutr Soc. 1967;26:11
  7. MacDonald I. Changes in the fatty acid composition of sebum associated with high carbohydrate diets. Nature. 1964;203:1067–1068
  8. MacDonald I. Dietary carbohydrates and skin lipids. Br J Dermatol. 1967;79:119–121
  9. Downing D, Strauss J, Pochi P. Variability in the chemical composition of human skin surface lipids. J Invest Dermatol. 1969;53:322–327
  10. Kellum R. Human sebaceous gland lipids. Arch Dermatol. 1967;95:218–220
  11. Greene R, Downing D, Pochi P, Strauss J. Anatomical variation in the amount and composition of human skin surface lipid. J Invest Dermatol. 1970;54:240–247
  12. Cassidy D, Lee C, Laker M, Kealey T. Lipogenesis in isolated human sebaceous glands. FEBS Lett. 1986;200:173–176
  13. Middleton B, Birdi I, Heffron M, Marsden J. The substrate determines the rate and pattern of neutral lipid synthesized by isolated human sebaceous glands. FEBS Lett. 1988;231:59–61
  14. Downing D, Stewart M, Wertz P, Strauss J. Essential fatty acids and acne. J Am Acad Dermatol. 1986;14:221–225
  15. Katsuda Y, Iida T, Inomata S, Denda M. Unsaturated fatty acids induce calcium influx into keratinocytes and cause abnormal differentiation of epidermis. J Invest Dermatol. 2005;124:1008–1013
  16. Nicolaides N. Skin lipids: their biochemical uniqueness. Science. 1974;186:19–26
  17. Shalita A. Genesis of free fatty acids. J Invest Dermatol. 1974;62:332–335
  18. Maeda T. Electrons microscopic study of experimentally-induced comedo effects of vitamin A acid on comedo formation. J Dermatol. 1991;18:397–407
  19. Stewart M, Grahek M, Cambier L, Wertz P, Downing D. Dilutional effect of increased sebaceous gland activity on the proportion of linoleic acid in sebaceous wax esters in epidermal acylceramides. J Invest Dermatol. 1986;87:733–736
  20. Ge L, Gordon J, Hsuan C, Stenn K, Prouty S. Identification of the D-6 desaturase of human sebaceous glands: expression and enzyme activity. J Invest Dermatol. 2003;120:707–714
  21. Cordain L, Lindeberg S, Hurtado M, Hill K, Eaton B, Brand-Miller B. Acne vulgaris – a disease of Western civilization. Arch Dermatol. 2002;138:1584–1590
  22. Brand-Miller J, Thomas M, Swan V, Ahmad Z, Petocz P, Colagiuri S. Physiological validation of the concept of glycemic load in lean young adults. J Nutr. 2003;133:2728–2732
  23. Smith R, Mann N, Braue A, Mäkeläinen H, Varigos G. The effect of a high protein, low glycemic load diet versus a conventional, high glycemic load diet on biochemical parameters associated with acne vulgaris. A randomized, investigator-masked, controlled trial. J Am Acad Dermatol. 2007;57:247–256
  24. Smith RN, Mann NJ, Braue A, Mäkeläinen H, Varigos GA. A low-glycemic-diet improves symptoms in acne vulgaris patients: a randomized controlled trial. Am J Clin Nutr. 2007;86:107–115
  25. Nordstrom K, Schmus H, McGinley K, Leyden J. Measurement of sebum output using a lipid absorbent tape. J Invest Dermatol. 1986;87:260–263
  26. Martin A, Lookingbill D, Botek A, Light J, Thiboutot D, Girman C. Health-related quality of life among patients with facial acne – assessment of a new acne-specific questionnaire. Clin Exp Dermatol. 2001;26:380–385
  27. Doran T, Baff R, Jacobs P, Pacia E. Characterization of human sebaceous cells in vitro. J Invest Dermatol. 1991;96:341–348
  28. Sinclair A, O’Dea K, Dunstan G, Ireland P, Niall M. Effects on plasma lipids and fatty acid composition of very low fat diets enriched with fish or kangaroo meat. Lipid. 1987;22:523–529
  29. Strauss J, Pochi P. The quantitative gravimetric determination of sebum production. J Invest Dermatol. 1961;36:293–298
  30. Strauss J, Kligman A, Pochi P. The effect of androgens and estrogens on human sebaceous glands. J Invest Dermatol. 1962;39:139–155
  31. Farrell L, Strauss J, Stranieri A. The treatment of severe cystic acne with 13-cis-retinoic acid. Evaluation of sebum production and the clinical response in a multiple-dose trial. J Am Acad Dermatol. 1980;3:602–611
  32. Stewart M. Sebaceous gland lipids. Sem. Dermatol. 1992;11:100–105
  33. Zheng Y, Eilertsen K, Ge L, Zhang L, Sundberg J, Prouty S, et al. Scd1 is expressed in sebaceous glands and is disrupted in the asebia mouse. Nat Genet. 1999;23:268–270
  34. Perisho K, Wetz PW, Madison KC, Stewart ME, Downing DT. Fatty acids of acylceramides from comedomes and from the skin surface of acne patients and control subjects. J Invest Dermatol. 1988;90:350–353
  35. Sansome-Bassano G, Cummings B, Seeler A, Reisner R. Differences in the lipid constituents of sebum from pre-pubertal and pubertal subjects. Br J Dermatol. 1980;103:131–137
  36. Downie M, Guy R, Kealey T. Advances in sebaceous gland research: potential new approaches to acne management. Int J Cos Sci. 2004;26:291–311
  37. Downie M, Kealey T. Human sebaceous glands engage in aerobic glycolysis and glutaminolysis. Br J Dermatol. 2004;151:320–327
  38. Downie M, Kealey T. Lipogenesis in the human sebaceous gland:glycogen and glycerophosphate are substrates for the synthesis of sebum lipids. J Invest Dermatol. 1998;111:199–205
  39. Im M, Hoopes J. Enzymes of carbohydrate metabolism in normal human sebaceous glands. J Invest Dermatol. 1974;62:153–160
  40. Brand-Miller J, Holt S, Pawlak D, McMillan J. Glycemic index and obesity. Am J Clin Nutr. 2002;76(suppl 2):281S–285S
  41. Kiens B, Richter E. Types of carbohydrate in an ordinary diet affect insulin action and muscle substrates in humans. Am J Clin Nutr. 1996;63:47–53
  42. Tsai M, Chen W, Cheng Y, Wang C, Chen G, Hsu T. Higher body mass index is a significant risk factor for acne formation in schoolchildren. Eur J Dermatol. 2006;16:251–253
  43. Spieth L, Harnish J, Lenders C, Raezer L, Pereira M, Hangen J, et al. A low glycemic index diet in the treatment of pediatric obesity. Arch Pediatr Adolesc Med. 2000;154:947–951
  44. Ebbeling C, Leidig M, Sinclair K, Hangen J, Ludwig D. A reduced-glycemic load diet in the treatment of adolescent obesity. Arch Pediatr Adolesc Med. 2003;157:773–779
  45. Draelos Z. The effect of a daily facial cleanser for normal to oily skin on the skin barrier of subjects with acne. Cutis. 2006;78(Suppl. 1):34–40
  46. Foster-Powell K, Holt S, Brand-Miller J. International table of glycemic index and glycemic load values. Am J Clin Nutr. 2002;76:5–56
  47. The Glycemic Index and GI database is produced by Sydney University and is available at http://www.glycemicindex.com (accessed 2003).

☆ This study was funded by a research grant from Meat and Livestock Australia. The corresponding author receives a postgraduate scholarship from MINTRAC (National Meat Industry Training Council of Australia). The remaining authors have no potential conflict of interest. This study was the responsibility of the investigators. MLA had no role in data collection, data analysis, data interpretation or submission of this paper for publication.

PII: S0923-1811(07)00363-5

doi:10.1016/j.jdermsci.2007.11.005

« BackJournal of Dermatological Science

Article Tools

* Image Usage & Resolution