To receive more information about up-to-date research on micronutrients, sign up for the free, semi-annual LPI Research Newsletter here.
The term "retinoids" refers to vitamin A and the various molecules derived from vitamin A, which itself is also known as retinol (ROL). In the skin, ROL is converted to retinaldehyde (RAL) and then to retinoic acid (RA). RA modulates gene expression and influences cellular processes in both the epidermis and dermis, thereby exerting potent effects on skin health.
|Commonly Used Names and Abbreviations for Vitamin A
and its Derivatives
|RETINOID||ALSO KNOWN AS||ABBREVIATION|
Dietary retinyl esters (RE) and beta-carotene are converted to ROL by the liver. In the liver, ROL is either reesterified to RE for storage or released into the circulation (1). Bound to retinol-binding protein, ROL reaches the skin via capillaries in the dermis (2). Cellular uptake of ROL from plasma is thought to occur via receptor-mediated uptake or endocytosis (1, 3). Cells in both the epidermis and dermis are targets for retinoid action. Keratinocytes and fibroblasts convert ROL first to retinaldehyde (RAL) and then to all-trans-retinoic acid (at-RA) (4).
The physiological actions of retinoids in the skin are mediated primarily through their interactions with retinoic acid receptor (RAR) and retinoid X receptor (RXR), members of a superfamily of ligand-activated transcription factors (2). The primary isoforms present in human epidermis are RAR-alpha and RXR-gamma. More specifically, mRNA and protein levels of total and isoform-specific RXR and RAR were measured in human epidermal biopsies (5). There is approximately 5-fold more total RXR than total RAR protein. Of these totals, the isoforms present are: 90% RXR-alpha and no detectable RXR-beta or RXR–gamma and 87% RAR-gamma, 12-14% RAR-alpha, and no detectable RAR-beta. RAR-beta is present only in the dermis and not in keratinocytes (6).
Human epidermal cells preferentially bind at-RA (7). In skin cells, other retinoids (9-cis-RA, 13-cis-RA, and at-ROL) isomerize to at-RA, the primary ligand that mediates RAR-dependent responses in human skin (7-10). Skin cells also have the capacity to convert beta-carotene into vitamin A metabolites. Incubation with radiolabelled beta-carotene increased ROL concentration in cultures of human keratinocytes and melanocytes (11). Topically applied beta-carotene increased the retinyl ester content in excised human skin samples in hairless mouse skin, demonstrating that beta-carotene can serve as a precursor to epidermal vitamin A (12).
A very common side effect of topical retinoid therapy is "retinoid dermatitis," also referred to as retinoid irritation or retinoid reaction. Retinoid dermatitis is characterized by erythema, scaling, dryness, and pruritis (13-15). Topical retinoids induce changes in the epidermis that lead to increased proliferation and altered differentiation of keratinocytes (see Photoaging); this in turn disrupts the barrier of the skin and contributes to the features of retinoid dermatitis (personal communication, Dr. Gary Fisher, December 2012). The level of irritation correlates with the potency of the applied vitamin A metabolite. Thus, adjusting the timing, dose, and form of topical retinoid can influence the retinoid reaction. For example, 0.025% topical tretinoin has the same effect as 0.05% tretinoin on fine wrinkles and roughness, but with less irritation and after a longer exposure time (14). Notably, tretinoin concentrations less than 0.01% are largely ineffective in the treatment of photodamaged skin (16-18). Topical application of other retinoids, such as isotretinoin and retinol also lead to clinical improvements, again with less skin irritation and after a much longer exposure time than tretinoin (19).
Percutaneous absorption of topical retinoids is minimal, thus very little of the active ingredient reaches the systemic circulation. Repeated application of 2 g of 0.025% tretinoin gel (equivalent to 0.05 mg at-RA) to the face, neck, and upper part of the chest nightly for 14 days did not influence the concentration of plasma at-RA and its metabolites in a small sampling of human volunteers (N=4) (20). In another study, radiolabeled tretinoin (0.05%) in cream or emollient was applied to facial skin, and the amount of tretinoin reaching the circulation was measured after single or repeated (once daily for 28 days) application (21). In both cases, no significant changes in plasma tretinoin concentration were observed. Additionally, only 1-2% of the radiolabeled-tretinoin dose was excreted in the urine and feces, indicating that very low levels of topical tretinoin reach the systemic circulation.
Feeding rats diets devoid of vitamin A affects epithelial tissues in various parts of the body, including the respiratory, alimentary, and genitourinary tracts, and the eyes and paraocular glands (22). Specifically, there is widespread keratinization, meaning that normal epithelium is replaced by stratified keratinizing epithelium, which has different functional properties (22). There is also atrophy of many glands, including sweat and sebaceous glands (22, 23). Based on experimentation in animals, vitamin A deficiency is also associated with delayed wound repair (see Wound Healing) (24).
Ultraviolet radiation (UVR) damages skin via many mechanisms (see Challenges to Skin in the separate article on skin health). One way UVR contributes to photodamage is by modulating signaling pathways that influence collagen homeostasis in skin cells (4, 25). For example, UVR induces the transcription factor AP-1, which increases the expression of several matrix metalloproteases (MMPs) (collagenase, gelatinase, stromelysin), proteolytic enzymes that degrade dermal collagen and fibrillin (4).
In a small sampling of adult subjects (N=6), biopsies of irradiated and non-irradiated buttock skin were compared before and after pretreatment with at-RA (26). UVB (2X MED) increased DNA binding of the transcription factors AP-1 and NF-kappa-B followed by increased mRNA, protein, and activity levels of collagenase, stomelysin, and gelatinase—matrix-degrading metalloproteinases (MMPs) regulated by these transcription factors. Pretreatment with at-RA inhibited UVB-induced AP-1 DNA-binding activity (a process call transrepression (27)), as well as MMP expression and activity by 50-80%. In a second human study, skin samples were obtained in irradiated and non-irradiated buttock skin from 59 white subjects (mean age, 35 years) who were pretreated with tretinoin or vehicle 48 hours prior to UV exposure (28). Notably, the UVR dose used in this study did not cause erythema and was equivalent to five to 15 minutes of exposure to noonday sun. The authors demonstrated that UVR induced three MMPs, collagenase, gelatinase, and stromelysin, in the epidermis and dermis. Pretreatment with tretinoin inhibited UV-induced MMP expression and activity by 70-80% without affecting the expression of their inhibitor (tissue inhibitor of MMP type I, (TIMP)). In a separate study with the same study design, pretreatment with retinoic acid prevented UV-induced reductions in RXR-alpha and RAR-gamma protein levels and transcriptional activity in buttock skin compared to non-irradiated sites on the same individual (N=70) (29). Overall, retinoids appear to prevent photodamage by interfering with UVR-mediated activation of signaling pathways that damage dermal collagen.
Clinical signs of photoaging include fine and coarse wrinkles, mottled hyperpigmentation, actinic lentigines, freckles, roughness, telangiectasia, and sallowness (25, 30). These traits are reflective of histological changes in both the epidermis and dermis in sun-exposed skin.
Topical retinoids lead to visible improvement in fine wrinkling, smoothness, and hyperpigmentation of photodamaged skin (13-15, 25, 31). Topical retinoids induce a number of histological changes in both the epidermis and dermis, and it is thought that changes in dermal collagen underlie the observed clinical improvements (32-36). Both clinical and histological skin parameters return to baseline upon discontinuation of topical application of retinoids; thus, a long-term maintenance regimen is necessary to sustain retinoid-induced improvements (37).
Tretinoin. Among the retinoids, tretinoin (all-trans-retinoic acid, at-RA) is the most extensively studied topical agent for skin health (2, 13-15). Topical tretinoin is most effective for treating fine facial wrinkles, rough texture associated with photodamage, and mottled epidermal pigmentation.
In one of the earliest evaluations of the efficacy of topical tretinoin on photodamaged skin, punch biopsies were obtained before and after treatment with 0.05% tretinoin cream or vehicle to photodamaged forearm and facial skin in a small sampling of elderly volunteers (aged 66-77 years) (38). After three months (forearms, N=16) or six to 12 months (face; N=8), histological abnormalities characteristic of end-stage epidermal photoaging (fewer cell layers, smaller cells, flattening of the dermatoepithelial junction, dense perinuclear clustering of large melanocytes, and hyperkeratosis) were corrected in tretinoin-treated skin. Dermal changes were not obvious, and the permanence of the tretinoin-induced changes was not evaluated (38).
These observations set the stage for subsequent randomized controlled trials (RCTs) evaluating the safety and efficacy of topical retinoids in the treatment of photodamaged skin. In an early randomized, double-blind, vehicle-controlled trial, 30 patients (mean age, 50 years) with photodamaged skin applied 0.1% tretinoin cream or vehicle once nightly for four months to photodamaged facial and forearm skin (39). Significant improvement in fine facial wrinkling, coarse wrinkles, sallowness, roughness, and actinic lentigines was observed following topical tretinoin treatment. Ninety-two % of patients experienced "retinoid dermatitis" that lasted from two weeks to three months. Histologic data showed that tretinoin-treated forearm skin exhibited increased epidermal thickness, compaction of the stratum corneum (SC), and increased vascularity in the papillary dermis compared to vehicle-treated skin. A subset of subjects continued in an open-label format for a total of 10 months (N=21) or 22 months (N=16) of topical tretinoin treatment (40). Clinical changes in fine wrinkling and texture of facial skin were sustained despite reduced dose (0.05%) or frequency (every other day) of tretinoin application.
Biopsies obtained from photodamaged forearm skin and sun-protected buttock skin were compared in 26 healthy, white subjects (mean age, 56 years) (41). Collagen I formation (detected immunohistochemically) was 56% lower in the papillary dermis in photodamaged skin. In a separate analysis, 29 healthy subjects (mean age, 63 years) were randomly assigned to receive daily application of 0.1% tretinoin cream or vehicle for 10 to 12 months on photodamaged skin (41). Tretinoin treatment resulted in an 80% increase in collagen I formation compared to baseline. Thus, topical tretinoin partially restored collagen formation in the dermis of photodamaged skin.
These trials provided abundant information regarding the effect of topical tretinoin on photodamaged skin, but they are somewhat limited by small sample sizes. As a result, a number of larger, long-term, RCTs have been conducted that corroborate the efficacy of topical tretinoin in the treatment of clinical and histological features of photoaging.
Different doses of topical tretinoin were evaluated in a large, multicenter, double-blind, vehicle-controlled RCT involving 533 subjects (mean age 42 years) with mild to moderately photodamaged facial skin (16). Three concentrations of tretinoin were tested: 0.05%, 0.01%, and 0.001%. After six months of treatment, 0.05% and 0.01% tretinoin produced four significant epidermal differences compared to vehicle-treated skin: (1) increased epidermal thickness, (2) increased number of granular cell layers, (3) decreased melanin content, and (4) SC compaction. The 0.001% tretinoin was no more active than vehicle. A second very similar multicenter RCT compared these same doses of topical tretinoin in the treatment of photodamaged facial skin (17). In this trial, 296 subjects (mean age, 42.5 years) applied topical tretinoin emollient cream (TEC) or vehicle to the entire face nightly for 24 weeks. Significant clinical improvement in mottled hyperpigmentation, fine wrinkling, and roughness, and the same histological changes noted above occurred with 0.05% and 0.01% TEC compared to vehicle, though 0.05% TEC was the most effective dose. A third multicenter RCT compared 0.05% or 0.01% topical tretinoin applied daily for 24 weeks to photodamaged facial skin in 299 subjects (mean age, 42 years) (18). As above, 0.05% tretinoin improved the same clinical and histological features of photoaging after six months of treatment. 0.01% tretinoin was less effective, only improving fine wrinkling scores.
To assess longer-term clinical and histological effects, a subset of subjects from two of the multicenter trials (17, 18) continued in a 24-week double-blind extension trial (42). All subjects received active treatment of either 0.05% TEC (N=126) or 0.01% TEC (N=133), although subjects and investigators were blinded to dose. Clinical improvements in fine wrinkling, roughness, and mottled hyperpigmentation were maintained or enhanced over the extension period. Continued clinical improvement was more evident in the 0.01% TEC group, with clinical scores approaching those of the 0.05% dose. Histological parameters were more variable. In both groups, reversal of the increase in SC compaction, epidermal thickness, and number of granular cell layers was observed at the end of the extension period. Melanin content, on the other hand, continued to decrease at both doses of TEC. After 48 weeks of exposure, new histological changes were observed at both concentrations of TEC, including increased epidermal mucin content and reduced dermal elastin content.
Twenty-five subjects from two of the previously conducted controlled trials (16, 18) provided punch biopsies from the periorbital region of the face in order to assess ultrastructural changes associated with topical tretinoin treatment (36). While no changes were evident after six months, electron microscopy revealed clear morphological changes in the papillary dermis (replacement of disorganized collagen fibers with well-organized, packed fibers) after 12 months of topical 0.05% tretinoin treatment. A small number of subjects [(N=27) from abovementioned trial (17) were followed even further, up to four years, after varying exposures of 0.001–0.05% TEC, applied once weekly to once daily (43). As was observed after 48 weeks of TEC exposure (37), the epidermal changes observed during the initial phases of topical tretinoin treatment were no longer evident, and there was a significant increase in epidermal mucin content and a significant decrease in dermal elastin content. No signs of abnormal cell or tissue morphology were observed. More recently, 204 subjects (mean age, 63 years) with moderate to severe facial photodamage were randomized to apply 0.05% tretinoin emollient cream or vehicle once daily for two years (44). Topical tretinoin resulted in a significant improvement in clinical signs of photodamage (fine and coarse wrinkling, mottled hyperpigmentation, lentigines, and sallowness) compared to placebo after 24 months of treatment. Mild cutaneous irritation was higher in the tretinoin group and peaked in the first two months, but the overall incidence of adverse events was similar in the tretinoin and placebo groups at the end of the study.
Consistent with these changes in human skin, studies in hairless mice demonstrate that topical at-RA induces a "zone of repair" in the subepidermal dermis in UV-damaged skin. After inducing connective tissue damage by exposing hairless mice to UVR for 10 weeks, treatment with topical at-RA (0.05%) for 10 weeks significantly increased dermal vascularity and the size of the subepidermal repair zone (i.e., newly formed and organized collagen bundles) compared to untreated control mice (32). Topical at-RA also leads to stimulation of collagen synthesis and the effacement of UV-induced surface wrinkles in hairless mice (32-35).
Overall, long-term, continuous exposure to topical tretinoin results in significant improvement in clinical parameters, though histological changes vary over time (31). With continued application, some early epidermal changes (SC compaction, increased epidermal and granular layer thickness) return to baseline while other histological changes (increased epidermal mucin content, increased dermal collagen synthesis) become evident. Due to technical limitations of the employed methodologies (i.e., H&E staining) in the human trials, retinoid-induced epidermal changes are well described while dermal changes often go undetected. Nonetheless, due to the weight of evidence from studies in hairless mice and electron microscopic analysis of human skin samples, it is thought that clinical improvement to photodamaged skin is a consequence of increased collagen synthesis induced by at-RA.
Isotretinoin. In a randomized, double-blind, vehicle-controlled trial, 800 adults (mean age 53.5 years) with moderate to severe photodamage applied topical 0.1% isotretinoin (13-cis-retinoic acid) or vehicle to the face, forearms, and hands nightly for 36 weeks (45). According to physician and patient assessment, isotretinoin treatment resulted in significant improvements in overall appearance, fine and coarse wrinkling, texture and hyperpigmentation compared to baseline. Computer image analysis revealed a 20% reduction in facial wrinkle length following isotretinoin treatment compared to baseline; no change was observed in vehicle-treated skin. Most patients reported adverse events (erythema, peeling/scaling, burning, pruritis) as mild to moderate, though 5-10% of patients graded them as severe. Punch biopsies (N=120) from forearm skin showed that isotretinoin increased epidermal thickness but did not alter any other histological parameters. Consistent with other reports, there was no change in plasma retinoid levels throughout the study period.
Retinol. In a vehicle-controlled trial, the activity between all-trans-retinol (at-ROL) and all-trans-retinoic acid (at-RA) was compared in human skin (19). Vehicle, at-ROL (1.6%), and at-RA (0.025%) were applied to different sites on the buttock skin of seven subjects and compared after four days of continuous occlusion. Although a much higher concentration of ROL was needed to achieve similar results (ROL was ~20-fold less potent than RA), ROL induced the same histological changes (hyperplasia and spongiosis) as at-RA without causing erythema.
ROL has also been evaluated in the treatment of naturally aged human skin. Although the features of naturally aged and photodamaged skin differ, disrupted collagen homeostasis is thought to contribute to wrinkling in both situations (30). Skin biopsies from 72 individuals from four age groups (18-29 years, 30-59 years, 60-79 years, and ≥80 years) were compared (46). Naturally aged skin exhibits reduced fibroblast proliferation, increased MMP expression, and reduced collagen synthesis. In a separate analysis, 53 individuals (80 years and older) were treated with 1% topical retinol or vehicle on buttock skin for seven days (46). Topical ROL increased dermal fibroblast number, reduced MMP expression, and increased collagen synthesis compared to vehicle-treated sites. Thus, even after only seven days of application, topical retinol partially reversed some of the cellular abnormalities present in naturally aged skin. In a randomized, double-blind, vehicle-controlled study, 36 elderly subjects (mean age, 87 years) received topical 0.4% retinol (ROL) lotion or vehicle to their left or right upper, inner arm 3 times/week for 24 weeks (47). Topical retinol improved clinical appearance (fine wrinkling) and increased the expression of two matrix molecules, glycosaminoglycan and procollagen I, compared to vehicle-treated skin. Most subjects reported mild cutaneous irritation on the ROL-treated arm, including erythema, peeling, pruritis, dryness, and burning/stinging.
Vitamin A deficiency is associated with impaired immune function (see Vitamin A) and delayed wound healing. The effect of cod liver oil delivered locally and orally on the rate of wound healing was investigated in vitamin A-deficient and normal rats (48). In vitamin A-deficient rats, wounds treated with cod liver oil healed more quickly than untreated wounds; oral administration also accelerated wound healing in deficient animals, though to a lesser extent than topically applied cod liver oil. In vitamin A-replete rats, cod liver oil had no effect on the rate of wound healing. While it was not specifically determined that the vitamin A in cod liver oil was responsible for the accelerated healing in deficient animals, the authors demonstrated that linseed oil, rich in essential fatty acids, had no effect on the rate of wound healing in either scenario (48).
Vitamin A also modulates wound healing in the context of steroid therapy. When given in large to moderate doses within the first two to three days after injury, anti-inflammatory steroids reduce the rate of healing of surgical wounds (49). The interaction between vitamin A and the anti-inflammatory steroid, cortisone, on the rate of wound healing was investigated in rats (50). Wound strength was measured over time to reflect the rate of healing of experimentally induced surgical wounds. Vitamin A was delivered as intraperitoneal injections (i.p.) of peanut oil supplemented with 1,500 IU vitamin A in the presence and absence of cortisone. Vitamin A alone did not increase wound strength over control levels, yet vitamin A prevented the cortisone-induced reduction in wound strength. The mechanisms by which vitamin A counteracts cortisone-induced defects are not known, however, and further research in humans is necessary before the therapeutic application of vitamin A in wound healing can be realized.
Acne. Topical tretinoin (all-trans-retinoic acid, at-RA) is considered a very safe and effective treatment for mild to moderate acne (51, 52). Oral isotretinoin (13-cis-retinoic acid) is used to treat severe cases of acne that are resistant to topical therapies (14, 53).
Acne vulgaris affects the pilosebaceous unit, hair follicles in the skin that are associated with a sebaceous gland (54). The primary pathophysiological factors leading to acne include increased sebum production, altered growth and differentiation of follicular keratinocytes, bacterial colonization of the follicle by P.acnes, and inflammatory and immune reactions (52, 54). These factors contribute to the formation of the primary acne lesion, the microcomedone. As sebum and infiltrate accumulate, the microcomedone transforms into a visible, mature comedone, Both open and closed comedones represent noninflammatory acne lesions. When the follicular wall surrounding a mature comedone ruptures, inflammatory acne lesions known as pustules and papules occur (54).
Topical retinoids. Topical tretinoin influences the proliferation and differentiation of keratinocytes, thereby increasing follicular epithelial turnover, accelerated shedding of corneocytes, and the expulsion of mature comedones (52). Topical isotretinoin inhibits the accumulation of polymorphonuclear leukocytes (PMN) (55), a key feature of inflammatory dermatoses like acne, and reduces the number inflammatory lesions.
A multicenter, double-blind, vehicle-controlled trial assessed the efficacy of topical isotretinoin (13-cis-RA) in the treatment of mild to moderate facial acne (56). Two hundred sixty-eight subjects (both genders, aged, 13-30 years) applied 0.05% isotretinoin gel or vehicle twice daily for 14 weeks. Isotretinoin treatment significantly reduced both inflammatory and noninflammatory lesion counts compared to vehicle. More specifically, inflammatory lesion count was reduced by 55% with isotretinoin compared to 25% with vehicle; noninflammatory lesion count was reduced by 46% with isotretinoin compared to 14% with vehicle.
Percutaneous absorption of topical tretinoin is minimal (see Topical Application). Typically, side effects associated with topical tretinoin are local skin reactions, such as redness, peeling, dryness, itching, and burning (14, 52). Although use of topical retinoids is not associated with an increased incidence of birth defects in retrospective studies (57, 58), it is still advised to abstain from their use during pregnancy. As is the case in the treatment of visible signs of photoaging (see Photoaging), there is a need for a long-term maintenance regimen with topical retinoids in order to sustain remission of acne symptoms (52).
Oral isotretinoin. Oral isotretinoin is very effective in the treatment of severe cases of acne, however, safety issues are of major clinical importance (14, 53). Isotretinoin crosses the placenta and is teratogenic, meaning it causes developmental abnormalities. Therefore oral isotretinoin is strictly contra-indicated prior to and during pregnancy. Other side effects of oral retinoids resemble those associated with hypervitaminosis A and include mucocutaneous adverse effects, hyperostosis, and extraskeletal calcification (53). Oral isotretinoin reaches a steady-state concentration in plasma after one week of ingestion, yet there is no accumulation in the epidermis (59). After discontinuation of oral therapy, isotretinoin disappears from both the skin and serum within two to four weeks (59).
Oral isotretinoin has unique inhibitory activity on sebaceous glands. It decreases the proliferation of basal sebocytes, suppresses sebum production, and inhibits sebocyte differentiation (60, 61). In vitro experiments indicate that isotretinoin also induces sebocyte death, further contributing to its sebosuppressive effect (9, 62).
Skin is a major retinoid-responsive tissue and vitamin A metabolites have been in use for decades for the treatment of various skin conditions, including acne vulgaris and photoaging. Topical tretinoin penetrates the skin layers without reaching the systemic circulation. A very common side effect of topical tretinoin is a skin irritation known as "retinoid dermatitis." Treatment with lower tretinoin doses or other retinoid metabolites can achieve similar skin changes with less irritation but require a longer exposure time.
Long-term (at least six months), continuous exposure to topical 0.05% tretinoin (all-trans-retinoic acid, at-RA) results in significant improvement in clinical signs of photoaging, namely fine wrinkling, roughness, and hyperpigmentation. With continued application, clinical improvements are maintained while histological changes in the epidermis and dermis are more variable. In general, it is thought that clinical improvement to photodamaged skin is a consequence of increased dermal collagen synthesis induced by at-RA. Topical tretinoin is considered a very safe and effective treatment for mild to moderate acne, while oral isotretinoin is used to treat severe cases of acne that are resistant to topical therapies; both therapies require physician oversight.
Written in November 2012 by:
Giana Angelo, Ph.D.
Linus Pauling Institute
Oregon State University
Reviewed in December 2012 by:
Gary Fisher, Ph.D.
Harry Helfman Professor of Molecular Dermatology and
Director of the Photoaging and Aging Research Program
University of Michigan Department of Dermatology
Ann Arbor, MI
This article was underwritten, in part, by a grant from
Neutrogena Corporation, Los Angeles, California.
Copyright 2013-2014 Linus Pauling Institute
The Linus Pauling Institute Micronutrient Information Center provides scientific information on the health aspects of dietary factors and supplements, foods, and beverages for the general public. The information is made available with the understanding that the author and publisher are not providing medical, psychological, or nutritional counseling services on this site. The information should not be used in place of a consultation with a competent health care or nutrition professional.
The information on dietary factors and supplements, foods, and beverages contained on this Web site does not cover all possible uses, actions, precautions, side effects, and interactions. It is not intended as nutritional or medical advice for individual problems. Liability for individual actions or omissions based upon the contents of this site is expressly disclaimed.
Thank you for subscribing to the Linus Pauling Institute's Research Newsletter.
You should receive your first issue within a month. We appreciate your interest in our work.