The Risks and Benefits of Cutaneous Sunlight Exposure
A thesis submitted to The University of Manchester
for the degree of Doctor of Medicine
in the Faculty of Biology, Medicine and Health
2016
Sarah Jane Felton
School of Biological Sciences
Division of Musculoskeletal & Dermatological Sciences
2 CONTENTS
List of contents...... 3
List of figures ...... 8
List of tables ...... 10
List of abbreviations ...... 11
Abstract ...... 15
Declaration ...... 16
Copyright statement ...... 17
Acknowledgements ...... 18
CHAPTER 1: INTRODUCTION
1.1 Electromagnetic Radiation ...... 19
1.1.1 The electromagnetic spectrum ...... 19
1.1.2 Ultraviolet radiation ...... 21
1.1.3 Ultraviolet radiation reaching the Earth’s surface...... 22
1.2 Vitamin D ...... 23
1.2.1 Sources of vitamin D ...... 23
1.2.1.1 Dietary ...... 23
1.2.1.2 Cutaneous synthesis ...... 24
1.2.2 Vitamin D metabolism ...... 25
1.2.2.1 Control of vitamin D synthesis ...... 26
1.2.3 Functions of vitamin D ...... 26
1.2.3.1 Musculoskeletal health ...... 27
1.2.3.2 Systemic diseases and malignancies...... 28
1.2.4 Vitamin D supplementation ...... 29
1.2.5 Age and vitamin D production ...... 30
3 1.3 Skin pigmentation, delayed tanning and phototype ...... 31
1.3.1 Constitutive pigmentation ...... 31
1.3.2 Facultative pigmentation and epidermal thickening ...... 32
1.3.3 Skin phototype...... 33
1.3.4 Minimal erythemal dose ...... 33
1.4 Skin cancer ...... 34
1.4.1 Incidence ...... 34
1.4.2 Risk factors ...... 35
1.5 Cutaneous Deoxyribonucleic acid (DNA) ...... 36
1.5.1 Structure: Base pairing ...... 36
1.5.2 Direct DNA damage ...... 37
1.5.2.1 Cyclobutane pyrimidine dimers (CPD) ...... 37
1.5.2.2 Pyrimidine-6-4-pyrimidone photoproducts (6-4 PP)...... 38
1.5.2.3 Oxidative DNA damage: 8-Oxo-2'-deoxyguanosine ...... 38
1.5.3 Estimation of DNA damage by urinary excretion ...... 40
1.5.3.1 Urinary 8-oxodG ...... 40
1.5.3.2 Urinary CPD ...... 43
1.5.4 DNA damage and repair ...... 45
1.6 UVR-induced erythema ...... 48
1.7 Photoageing ...... 50
1.8 UVR effects on immunity...... 51
1.8.1 Photo-immunosuppression ...... 51
1.8.2 Antimicrobial peptides...... 52
1.9 Endocannabinoid system ...... 54
1.9.1 Anandamide and 2-arachidonyl glycerol ...... 54
1.9.2 The cutaneous endocannabinoid system ...... 54
4 1.9.3 Involvement in skin disease...... 55
1.9.4 Response to UVR...... 56
1.9.5 Involvement in mood ...... 57
1.10 Psychological effects of sunlight exposure ...... 58
1.10.1 Seasonal affective disorder...... 59
1.10.2 Dependency on suntanning ...... 59
1.10.3 Opioids and suntanning ...... 60
1.11 Artificial sunlight ...... 62
1.11.1 Sunbeds ...... 62
1.11.2 Psoriasis and phototherapy ...... 63
1.12 Guidance on sunlight exposure ...... 64
1.12.1 National campaigns ...... 64
1.12.2 Apparent discordance between national guidance and health
benefits ...... 65
1.13 Aims of this thesis ...... 66
Contribution of the author and her co-authors ...... 68
Rationale for submitting in alternative format ...... 69
CHAPTER 2: METHODOLOGY ...... 71
2.1 Intervention study ...... 71
2.1.1 Study design and approval ...... 71
2.1.2 Subjects ...... 71
2.1.3 Protocol overview ...... 72
2.1.4 Baseline MED assessment ...... 74
2.1.4.1 Waldmann UV6 lamp ...... 74
2.1.4.2 UVR output measurements ...... 74 5 2.1.4.3 MED testing procedure ...... 74
2.1.5 Simulated summer sunlight exposures ...... 75
2.1.5.1 Whole body irradiation cabinet ...... 75
2.1.5.2 UVR characterisation ...... 75
2.1.5.3 Cutaneous exposure in the irradiation cabinet ...... 77
2.1.5.4 Acute 2X MED exposures ...... 77
2.1.6 Diaries of dietary vitamin D intake ...... 78
2.1.7 Blood sampling ...... 78
2.1.7.1 Vitamin D status...... 79
2.1.7.2 Serum PTH and biochemistry analysis ...... 80
2.1.7.3 Serum endocannabinoid and N-acyl ethanolamine (NAE)
quantification ...... 80
2.1.8 Skin colour measurements ...... 81
2.1.8.1 L*a*b* recordings ...... 82
2.1.8.2 Individual typology angle...... 82
2.1.8.3 Erythema meter ...... 82
2.1.9 Cutaneous sampling ...... 83
2.1.9.1 Immunostaining for cutaneous anti-thymine dimer ...... 83
2.1.9.2 Immunostaining for cutaneous neutrophil elastase ...... 84
2.1.10 Urinalysis...... 85
2.1.10.1 Urinary 8-oxodG quantification ...... 86
2.1.10.2. Urinary thymine dimer quantification...... 88
2.2 Survey of tanning addiction in a cross-section of psoriasis patients ...... 91
2.2.1 Study design and approval ...... 91
2.2.2 Survey participants ...... 91
2.2.3 Psoriasis assessment ...... 92
6 2.2.4 Assessment of tanning ...... 92
2.2.5 Assessments of addictive-like tanning behaviour ...... 93
2.3 Statistical analyses ...... 94
CHAPTER 3: Manuscript 1
Concurrent beneficial (vitamin D production) and hazardous (cutaneous DNA damage) impact of repeated low-level summer sunlight exposures ...... 95
CHAPTER 4: Manuscript 2
Photoprotection conferred by low level summer sunlight exposures against pro- inflammatory UVR insult ...... 130
CHAPTER 5: Manuscript 3
Serum endocannabinoids and N-acyl ethanolamines and the influence of simulated solar UVR exposure in humans in vivo ...... 163
CHAPTER 6: Manuscript 4
The significant health threat from tanning bed use as a self-treatment for psoriasis ...... 200
CHAPTER 7: CONCLUSIONS...... 231
7.1 The principal risk and benefit of summer sunlight exposure ...... 231
7.2 UVR-induced photoprotection from summer sunlight exposure...... 233
7.3 Circulating endocannabinoids and their responses to UVR...... 235
7.4 The addictive behavioural component to tanning ...... 236
7.5 Synthesis of thesis findings ...... 237
7.6 Strengths and limitations...... 239
7.7 Implications for future research ...... 240
7
BIBLIOGRAPHY ...... 242
APPENDICES
1 Case Report Form ...... 257
2 Diet diary...... 265
3 Simplified Psoriasis Index ...... 269
4 Research letter: The significant health threat from tanning bed use as a self-
treatment for psoriasis...... 270
WORD COUNT 43,495 words
List of Figures
1.1A The electromagnetic Spectrum ...... 20
1.1B Solar radiation ...... 20
1.1C Ultraviolet penetration of skin...... 22
1.2A Chemical structures of vitamin D2 and vitamin D3...... 23
1.2B Vitamin D formation and its regulation ...... 25
1.5A Pyrimidine dimers ...... 38
1.5B Structure of 8-oxo-2'-deoxyguanosine (8-oxodG)...... 39
1.5C UVB-induced 8-oxodG formation in normal human epidermal
keratinocytes ...... 40
1.5D Competitive ELISA analysis of urinary 8-oxodG and thymine dimers ...... 42
1.5E Urinary thymine dimer levels before and after sunbathing on the beach..43
1.5F Urinary thymine dimer excretion as a function of UVR dose...... 44
1.8 Dual action of UVR on innate and adaptive immune systems...... 53
1.9A Chemical structures, anandamide and 2-arachidonoylglycerol...... 54
8 1.9B Effect of low- and high-dose UVB irradiation on a) AEA and b) 2-AG in
human keratinocyte cultures in comparison to control (CT)...... 56
2.3 Flow chart to demonstrate an individual’s progression throughout the
study period...... 73
2.5 Pre-vitamin D-weighted irradiance emission from the irradiation cabinet
versus that of a clear Manchester, UK (53.5°N) summer day at noon...... 77
Manuscript 1
1 25(OH)D levels during the simulated summer sunlight exposures ...... 123
2 Representative epidermal DNA damage in phototype II and V individuals
under varying conditions of UVR exposure...... 124
3 Urinary 8-oxodG damage (pmol/µmol creatinine) in skin type II and V
volunteers...... 125
Manuscript 2
1 Weekly skin colour measurements taken from buttock skin during the
simulated summer’s sunlight exposure...... 159
2 Effect of UVR exposure on epidermal thickness ...... 160
3 Cutaneous erythema following acute 2X MED UVB-challenge ...... 161
4 Neutrophil infiltration under varying conditions of UVR exposure...... 162
Manuscript 3
1 Schematic of endocannabinoid and NAE metabolism...... 195
2 Flow chart demonstrating an individual’s progression through the study
protocol...... 196
3A Serum endocannabinoid and NAE levels at baseline ...... 197
9 3B A representative UPLC-MS/MS chromatogram ...... 198
4 Serum 2-AG levels following UVR exposures...... 199
List of Tables
1.2A Examples of dietary sources of vitamin D...... 24
1.2B Proposed circulating 25(OH)D levels for health...... 27
1.3 Fitzpatrick skin phototype definitions...... 33
1.5A DNA complementary base pairs...... 36
1.5B Quantification of CPD and 6-4 PP in UVB-irradiated human keratinocytes
by radioimmunoassay ...... 46
Manuscript 1
1 Subject Demographics...... 120
Manuscript 3
1 Subject demographics...... 189
2 Serum endocannabinoid and NAE levels during week one of UVR-
exposure* for white Caucasians (n=10) and South Asians (n=6)……….....190
3 Serum endocannabinoid and NAE levels at baseline (week 0) and during
the six weeks of simulated summer UVR-exposures* for:
A. white Caucasians (n=10; upper) ...... 191
B. South Asians (n=6; lower)...... 192
Manuscript 4
1 Demographics and Clinical Characteristics...... 220
2 Characteristics of non-tanners and tanners...... 221
3 Characteristics of those who commenced tanning bed usage as a treatment
for psoriasis...... 222
10 4 Reasons for tanning as a self-treatment for psoriasis from 383
respondents ...... 223
5 Tanning characteristics of tanning bed users ...... 224
6 Distribution of responses adapted from the Diagnostic Statistical Manual
(DSM-V) criteria for Substance Use Disorder endorsed by current tanners
(n=301) ...... 225
List of Abbreviations
1,25[OH]2D 1,25-dihydroxy vitamin D (calcitriol)
25(OH)D 25-hydroxy vitamin D (calcidiol)
2-AG 2-arachidonoyl glycerol
8-oxodG 8-oxo-2’-deoxyguanosine
A Adenine
AEA Anandamide (N-arachidonoylethanolamide)
AMP Antimicrobial Peptide
APA American Psychiatric Association
APC Antigen Presenting Cell
BAD British Association of Dermatologists
BCC Basal Cell Carcinoma
BMI Body Mass Index
C Cytosine
CB Cannabinoid
CD Cluster of Differentiation molecule
CI Confidence Interval
CPD Cyclobutane pyrimidine dimmers
DGLEA N-Dihomo-γ-linolenoyl ethanolamine
11 DHEA N-Docosahexaenoyl ethanolamine
DH Department of Health
DHC Dehydrocholesterol
DNA Deoxyribonucleic Acid
DOH Department of Health
DPEA N-Docosapentaenoyl ethanolamine
DSM Diagnostic and Statistical Manual of Mental Disorder
EI Erythema Index
ELISA Enzyme-linked immunosorbent assay
EMR Electromagnetic Radiation
EMS Electromagnetic Spectrum
EPEA N-Eicosapentaenoyl ethanolamine
G Guanine
HPA Health Protection Agency
Hpf High power field
HPLC High pressure liquid chromatography
IARC International Agency for Research on Cancer
IFN-γ Interferon-gamma
IL Interleukin
IP Internet Protocol
ITA Individual Typology Angle
IOM Institute of Medicine iu International unit
LC Langerhan cell
LC-MS/MS Liquid chromatography coupled to tandem mass
spectrometry
12 LEA N-Linoleoyl ethanolamine mb Millibase
MEA N-Myristoyl ethanolamine
MED Minimal erythemal dose
MMP Matrix metalloproteinase mtDNA Mitochondrial DNA
NAE N-acyl ethanolamine
NER Nucleotide Excision Repair
NF-κB Nuclear Factor Kappa Beta
MAPK Mitogen activated protein kinase
NMSC Non-melanoma skin cancer
NPF National Psoriasis Foundation
NRPB National Radiological Protection Board
OEA N-Oleoyl ethanolamine
PBS Phosphate-buffered saline
PEA N-Palmitoyl ethanolamine
PHE Public Health England
Pmol Picomoles
PreD Pre-vitamin D
PP Pyrimidine-6-4-pyrimidone photoproducts
PTH Parathyroid hormone
PUVA Psoralen-UVA
RNI Reference Nutrient Index
ROS Reactive oxygen species
SACN Scientific Advisory Committee for Nutrition
SAD Seasonal affective disorder
13 SCC Squamous Cell carcinoma
SD Standard Deviation
SED Standard Erythemal Dose
SEM Standard Error of Mean
SPE Solid phase extraction
SPF Sun Protection Factor
SRD Substance-related disorder
SPI Simplified Psoriasis Index
SSR Solar-simulated radiation
STEA N-Stearoyl ethanolamine
T Thymine
Tan Tangent
TBS Tris-buffered saline
Th Helper T cell
Treg Regulatory T cell
TT Thymidine-thymidine dimer
T<>T Cyclobutane thymine dimer
TpT Thymidylyl-3’-5’thymidine
T<>pT Cyclobutane thymidylyl-3’-5’thymidine
UHPLC-MS/MS Ultra-high-performance liquid chromatography-
tandem mass spectrometry
UK United Kingdom
USA United States of America
UV Ultraviolet
UVR Ultraviolet Radiation
WHO World Health Organization
14 Abstract
The University of Manchester Sarah Jane Felton, Doctor of Medicine The Risks and Benefits of Cutaneous Sunlight Exposure, 2016
Background: Recommendations to restrict summer sunlight exposure to prevent skin cancer apparently conflict with requirements to protect bone health through adequate 25-hydroxyvitamin D (25[OH]D) levels, as provided by cutaneous ultraviolet (UV)B exposure. Furthermore, sunlight exposure promotes a feeling of euphoria that is felt to drive further sun-seeking behaviour. Objectives: My principal objective was to examine health risk (DNA damage) and health benefits (25[OH]D gain, and potential cutaneous photoprotection) following low-level summer sunlight exposures in people of light (phototype II) and darker (phototype V) skin. A further objective was to evaluate serum endocannabinoid levels, potential drivers of mood elevation, following these exposures, and to assess for evidence of tanning addiction in a cross-section of psoriasis patients who had received similar low dose UV exposures, as medical phototherapy. Methods: During wintertime, 10 white Caucasians and 6 South Asians aged 18 to 60 years, from Greater Manchester, UK, received a simulated summer's sunlight exposures, specifically 1.3 standard erythemal dose, thrice weekly for 6 weeks, whilst casually dressed. Serum and urine samples and skin colour measurements were taken at baseline, Monday, Wednesday and Friday of week 1 and then weekly, and buttock skin that had received differential UVR exposures was biopsied for immunohistochemical analysis. Phototype II individuals, who are at higher risk of sunburn, were subsequently challenged with 2X minimal erythema dose (MED) UVB on small areas of simulated summer-exposed and photoprotected skin. Separately, a link to an online tanning questionnaire survey was sent to all members of the National Psoriasis Foundation (USA) during my USA field trip. Results: The simulated summer resulted in 50% gain in 25(OH)D for both phototype groups, but significantly more cutaneous DNA damage (cyclobutane pyrimidine dimers, CPD) in phototype II than V (p<0.0001). There was no accumulation of cutaneous CPD after 6 weeks compared with a single UVR exposure in either group, while phototype V individuals had repaired a greater proportion of their CPD 24 hours after final UVR exposure (p<0.0001). Urinary oxidative DNA damage was higher in phototype II throughout the simulated summer (p=0.002) and unaffected by UVR. All individuals had significant skin darkening, and in phototype II, stratum corneum thickness increased significantly (p<0.05). This tanning response provided significant photoprotection against a pro-inflammatory UVB (2X MED) challenge, as shown by reduced erythema and neutrophil influx in skin exposed to the simulated summer than in photoprotected skin (p<0.05 for both). Serum endocannabinoid (2-arachidonoyl glycerol) levels increased significantly during the simulated summer in both phototype groups (p<0.01), peaking at week 2-3. The cross-sectional study of 1,832 psoriasis patients revealed 34% had used sunbeds; 11% of current users fulfilled diagnostic criteria for addictive-like tanning behaviour: female sex, younger age, younger age at psoriasis diagnosis, severe disease and prior phototherapy were significant risk factors for addiction. Discussion: These findings should assist public health guidance on safe sunlight exposure and highlight the need for distinct guidance targeted to different phototype groups. Furthermore, individuals with psoriasis, in particular those who previously received regular UVR exposure, are at high risk of tanning addiction that may be driven by the endocannabinoid system. 15 DECLARATION
Method development and mass spectrometric analysis of endocannabinoid and related lipid species was undertaken by A. Almaedani at the University of
Manchester, and reported in his PhD thesis, under the supervision of Professor
Anna Nicolaou.
16 Copyright Statement i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and she has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and
Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trade marks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property
University IP Policy (see http://documents.manchester.ac.uk/display.aspx?DocID=24420), in any relevant
Thesis restriction declarations deposited in the University Library, The University
Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The University’s policy on Presentation of Theses.
17 Acknowledgements
I should like to particularly thank my supervisor Professor Rhodes for her enduring patience, enthusiasm and inspiration.
This thesis is dedicated to my daughter, who makes everything worthwhile.
18 CHAPTER 1: INTRODUCTION
The skin forms an interface between the body and the environment. It is faced with the constant challenge of maintaining internal homeostasis despite external challenges. Perhaps the most relentless of these insults is electromagnetic radiation (EMR) in the form of ultraviolet radiation (UVR) from the sun. This research project will examine both the health benefits and the hazards posed by repeated sunlight exposure incident on the skin as experienced during the course of a summer, and in skin of different colour.
1.1 Electromagnetic Radiation
Electromagnetic Radiation (EMR) is the energy, travelling at the speed of light, which is emitted and absorbed by charged particles.
1.1.1 The Electromagnetic Spectrum
The electromagnetic spectrum (EMS) encompasses the entire range of EMR and is classified by order of increasing frequency and concomitant decreasing wavelength into radio waves, microwaves, infrared radiation, visible light, UVR, X- rays and gamma rays (Fig. 1.1A).
19 Figure 1.1A: The Electromagnet Spectrum. Components are displayed in order of increasing frequency and thus decreasing wavelength, from left to right.
Although the sun emits EMR across most of the EMS, infrared, UVR and visible radiation constitute its largest components (Fig. 1.1B). The Earth’s atmosphere filters out the harmful rays of UVC, and some UVB.
Figure 1.1B: Solar radiation, focusing on its UVR, visible and infrared components categorised in order of increasing wavelength.
20 1.1.2 Ultraviolet Radiation
The sun emits UVR, comprising approximately 10% of its total irradiance. UVR is subdivided by wavelength into three types: UVA (400-320nm), UVB (320-290nm) and UVC (290-100nm). UVC is wholly, and UVB partially, absorbed by the Earth’s ozone layer. Therefore UVA, and to a lesser extent, UVB reach the Earth’s surface.
The effects of EMR on biological systems depend on the amount of energy it is carrying and its frequency/wavelength: UVA, having a longer wavelength than
UVB, penetrates deeper into the dermis than UVB, which principally reaches the epidermis and to a much lesser extent the upper dermis (Fig 1.1C). Through its greater energy, UVB is around 1,000 times more effective at causing erythema
(redness) of the skin, i.e. sunburn, than UVA (Parrish et al., 1982). The epidermis additionally contains several molecules, including 7-dehydrocholesterol (7-DHC) and deoxyribonucleic acid (DNA), which significantly absorb UVB (Sutherland et al., 1980; Drobetsky et al., 1995). It is this UVR reaching and penetrating the skin that provokes the beneficial and harmful effects of sunlight that are the focus of this thesis.
21
Figure 1.1C: Ultraviolet penetration of skin. UVA is able to penetrate deeper into the dermis than UVB, which is predominantly absorbed by the epidermis.
1.1.3 Ultraviolet radiation reaching the Earth’s surface
The amount of UVR emitted from the sun that reaches the Earth's surface is dependent on its relative absorption, reflection and transmission whilst passing through the atmosphere. UVB is primarily absorbed by stratospheric ozone, but is additionally absorbed and reflected by clouds and air pollution. UVA is absorbed less by the ozone layer and scattered less by the atmosphere than UVB, so more
UVA reaches the Earth’s surface, year-round. Ozone depletion, which is a growing environmental concern, is therefore associated with greater penetration of UVB to the Earth’s surface (Madronich et al., 1998). Latitude, altitude, season and time of day all affect the amount of UVB at a particular location: The closer a place is to the sun, the more UVB reaching that location. Indeed, every 1,000 meter increase in altitude is associated with an approximate 5% increase in UVB irradiance [World
22 Health Organization (WHO) 2012]. Many surfaces also reflect the sun’s rays, increasing the overall UVR exposure: For example, grass and soil reflect less than
10% of UVR, whilst beach sand reflects 15% and fresh snow up to 80% (WHO
2012).
1.2 Vitamin D
Vitamin D is a fat-soluble prohormone that was originally identified in 1919 by its key role in canine bone health (Mellanby, 1919). It exists in two forms that differ in the structure of their side chains, vitamin D2 (ergocalciferol) and vitamin D3
(cholecalciferol; Fig 1.2A). Both forms are secosteroids, i.e. steroids in which one of the steroid rings is broken.
Figure 1.2A: Chemical structures of vitamin D2 (left) and vitamin D3 (right)
1.2.1 Sources of Vitamin D
1.2.1.1 Dietary Vitamin D is present naturally in few food sources including small amounts in eggs and milk, with larger amounts found only in oily fish. Both vitamin D2 and D3 can additionally be obtained from vitamin supplements and fortified foods including cereals and margarine (Table 1.2A).
23 Table 1.2A: Examples of dietary sources of vitamin D
Food source Approximate vitamin D
content (iu)*
1 tablespoon cod liver oil 1360
100g mackerel/salmon 400-800
40 One bowl (28g) of Kellogg’s corn flakes cereal (fortified since 2011)
One egg (vitamin D is found in the yolk) 26
* 1µg is equivalent to 40 international units [iu]
Overall, only around 10% of the body’s store of vitamin D is thought to originate from dietary intake, as vitamin D2 and/or D3, with the majority of vitamin D supplied from cutaneous synthesis, in the form of vitamin D3.
1.2.1.2 Cutaneous synthesis
Vitamin D3 is principally synthesised as a prohormone in the skin following cutaneous UVR exposure. The action spectrum for the rapid conversion of cutaneous 7-DHC to previtamin D3 was first determined by MacLaughlin et al in
1982, and further extended by the International Commission for Illumination in
2006 (CIE 2006). Based on their experimental studies, the optimum wavelengths were demonstrated to be between 295 and 300nm, with a peak at 297nm and virtually no production below 260nm or above 315nm (i.e. beyond UVB wavelengths).
24 The previtamin D3 then undergoes slower thermal isomerisation to vitamin D3
(cholecalciferol) over a few hours (Tian et al., 1993; Tian and Holick 1999), as illustrated in figure 1.2B.
Figure 1.2B: Vitamin D formation and its regulation
1.2.2 Vitamin D metabolism
Vitamin D, obtained from either cutaneous or dietary sources, is itself biologically inactive. Vitamin D-binding protein within the circulation binds the vitamin D and transports it to the liver where it is metabolised to 25(OH)D (calcidiol). This is the main circulating and storage form, and the form that is most commonly used as a measure of vitamin D status. The renal system further hydroxylates 25(OH)D to
1,25-dihydroxy vitamin D (1,25(OH)2D or calcitriol), the active hormone.
25 1.2.2.1 Control of vitamin D synthesis
The amount of sunlight required to maintain vitamin D status varies according to such factors as geography, season, cloud cover and time of day but, crucially, the photochemistry of vitamin D limits its own photosynthesis. This is because the isomer mixture of previtamin D3 and vitamin D3 within the skin reaches pseudoequilibrium in sunlight, thereby limiting the previtamin D3 formed from a single UVB exposure to 10-20% of the epidermal 7-DHC concentration. This means that additional UVR exposure does not further increase 25(OH)D levels (Webb et al., 1989). Moreover, vitamin D is itself photolabile such that any vitamin D that is not transported by vitamin D-binding protein is degraded to 5,6-transvitamin D and the suprasterols I and II by UVA and UVB on prolonged sunlight exposure or the next time the skin is exposed (Fig 1.2B).
Together these mechanisms limit the amount of vitamin D that can accumulate in the skin and enter the circulation, thus preventing an overdose/toxicity through this route. In contrast, dietary vitamin D bypasses these control mechanisms and so can, when taken excessively for a prolonged period of time, lead to toxic effects that include nausea, vomiting, renal stones and in extreme circumstances, death
(Ozkan et al., 2012).
1.2.3 Functions of vitamin D
Evidence is amassing to indicate potentially diverse functions of the active hormone of vitamin D (1,25(OH)2D) within body systems.
26 1.2.3.1 Musculoskeletal health
The key known function of 1,25(OH)2D is the regulation of calcium and phosphate levels via parathyroid hormone (PTH) control, intestinal absorption of calcium, renal reabsorption and bone remodelling. It is well documented that vitamin D deficiency, defined as a circulating 25(OH)D level of less than 5-10 ng/ml (12.5-25 nmol/L), can lead to the severe bone complications of rickets and osteomalacia in children and osteomalacia in adults (NRPB 2002; SACN 2016). It is now understood that 25(OH)D levels correlate positively with bone mineral density
(Bischoff-Ferrari et al., 2004) and that levels of less than 10-20 ng/ml (25-50 nmol/L, representing vitamin D insufficiency) are associated with secondary hyperparathyroidism, bone loss, osteoporotic fractures and muscle weakness
(Zittermann, 2003).
It has further been proposed that an optimal vitamin D status is reflected by
25(OH)D levels of 32 ng/ml (80 nmol/L) and above (Hollis, 2005). This has not gained general acceptance. In support of this cut-off level, signs of reduced bone mineral density subside and PTH is maximally suppressed as levels approach 32 ng/ml. Proposed 25(OH)D values for the various health categories are illustrated in table 1.2B.
Table 1.2B: Proposed circulating 25(OH)D levels for health
Category 25(OH)D level ng/ml nmol/L Deficiency <5-10 <12.5-25
Insufficiency 10-20 25-50
Sufficiency >20 >50
Optimal ≥30-32 ≥75-80
27 1.2.3.2 Systemic diseases and malignancies
A volume of indirect evidence, including from epidemiological studies, is also amassing to implicate the potential importance of vitamin D in a wider range of health benefits, including protection against malignant and immune-mediated diseases: Reduced responsiveness to insulin, high blood pressure and an increased risk of multiple sclerosis and solid tumours have been reportedly associated with what is regarded by some as “suboptimal” vitamin D status (i.e. serum 25(OH)D
<32 ng/ml; Grant 2006, Vieth 2006). However, results are somewhat inconsistent and studies frequently confounded by other factors (IOM, 2010; SACN 2016).
In vitro work has demonstrated that 1,25(OH)2D inhibits proliferation of cancer cells and modulates the immune system. Indeed 1,25(OH)2D can inhibit cellular proliferation, induce cellular maturation, inhibit angiogenesis and ultimately cause cell apoptosis (Thorne and Campbell 2008). Moreover, cells of both the innate and adaptive immune systems possess vitamin D receptors (van Etten and Mathieu,
2005) so enabling, for instance, vitamin D3-induced suppression of antigen presenting cell (particularly dendritic cell) maturation and T cell proliferation, in addition to the modulation of cytokine secretion and induction of regulatory T cells
(May et al., 2004). Indeed, variations in vitamin D receptor haplotypes (DNA sequences) have been associated with increased cancer risk for carcinomas of the colon, breast, prostate, vulva and other organs (Skowronski et al., 1993; Flugge et al. 2007; Engel et al. 2012; Salehin et al., 2012). Overall, the association between vitamin D deficiency and colon cancer has the highest evidence (IARC, 2008), while it is acknowledged that associations may be attributable to reverse causality. Thus, whilst vitamin D formation is an accepted health benefit from cutaneous sunlight exposure, other postulated benefits are currently speculative.
28 1.2.4 Vitamin D supplementation
Guidance on vitamin D supplementation has been limited and conflicting. The
WHO recommends that at northerly latitudes, particularly in winter when UVB at the Earth’s surface and thus cutaneous vitamin D production are negligible, all adults aged 19-50 years should consume 5 µg/day (200 iu) vitamin D, while 10
µg/day (400 iu) should be taken by those aged 51 to 65 years and 15 µg/day (600 iu) by people aged 65 or over (WHO 2004).
The UK Department of Health (DH) suggested a Reference Nutrient Intake (RNI) of
0 for most of the white Caucasian population (aged 4-64 years), since sunlight exposure alone was assumed to provide adequate vitamin D; only those who have pigmented skin, extensively cover their skin, or have restricted exposure to sunlight were advised to take vitamin D in the form of supplements (DH 1998; DH
2010).
The Institute of Medicine (for USA and Canada) convened an expert panel to review worldwide literature regarding the health outcomes of vitamin D and consequently to specify its dietary reference intake for white-skinned individuals over a range of latitudes across North America. In terms of health benefits, the crucial involvement of vitamin D was confirmed only for bone health. Having excluded “groups under medical care whose needs or sensitivities are affected by rare genetic disorders or diseases and their treatments” and assuming minimal sun exposure, they recommend an intake of 600 iu/day for individuals aged 1-70 years and 800 iu/day for those aged 71 and over, with the goal of reaching a circulating
25(OH)D level of ≥20 ng/mL in the North American population (IOM 2011).
29 More recently, Public Health England released new advice in 2016 based on recommendations from the Scientific Advisory Committee on Nutrition (SACN
2016). They suggested a RNI of 400 iu/day (10 µg/day) for everyone aged 4 years and above, particularly during autumn and winter, or year-round for those at increased risk of deficiency, such as those with darker skin, those who mostly cover their skin or spend majority of time indoors (PHE 2016). SACN themselves found convincing evidence only for links between vitamin D and musculoskeletal health, and at 25(OH)D levels above 25 nmol/L. Consequently they recommended year-round supplementation for all aged over 4 years, including pregnant and lactating women, to maintain at least this level year-round. They could not offer recommendations on durations of sun exposure to attain adequate vitamin D status because of the number of factors affecting endogenous synthesis including time of day, clothing coverage, air pollution, season, cloud cover and sunscreen
(SACN 2016).
1.2.5 Age and vitamin D production
The availability of 7-DHC for previtamin D3 formation in the skin appears to diminish with age: a small study comparing two young patients (of ages 8 and 18 years) and two older patients (aged 77 and 82 years) reported that the ability to produce 7-DHC in the older individuals was half that of the younger pair
(MacLaughlin and Holick, 1985). In addition, elderly persons, particularly if institutionalised, would be less likely to receive ambient UVR exposure from outdoor activities, further contributing to their risk of vitamin D deficiency/insufficiency. SACN did not offer separate guidance for elderly individuals, but did categorise frail or institutionalised people as being at increased
30 risk of deficiency (SACN 2016), and for these groups year-round supplementation was suggested by PHE (PHE 2016).
1.3 Skin pigmentation, delayed tanning and phototype
Variations in the levels of cutaneous melanin are responsible for the differences in skin colour seen between individuals. As constitutive and facultative pigmentation both contribute to skin colour and phototype, their relationships with UVR and vitamin D shall be here discussed.
1.3.1 Constitutive pigmentation
Constitutive pigmentation is a reflection of the melanin that intrinsically resides in an individual’s skin. It is primarily determined by genetic factors, and is responsible for the ‘baseline’ skin colour in non-UVR-exposed skin. Whilst all individuals possess roughly the same number of melanocytes, it is variations between the size and distribution of their melanosomes which account for skin colour, being smaller and more clustered in people with paler skin (Thong et al.,
2003). The wavelengths of UVR that stimulate pre-vitamin D production are also absorbed by cutaneous melanin. Thus, in people of darker skin colours, their higher constitutive levels of melanin absorb a greater proportion of incident UVB, so reducing the amount that reaches epidermal 7-DHC, predominantly found in the stratum basale and spinosum (Webb and Holick 1988). They consequently require more UVB to produce the same change in circulating 25(OH)D as in light-skinned people (Lo et al., 1986). However, they can also tolerate higher levels of UVB exposure before skin erythema is seen, and have a lower skin cancer risk. Studies examining impact of melanin on vitamin D synthesis in vivo, however, have shown conflicting results (Armas et al., 2007; Bogh et al., 2010; Farrar et al., 2011;
31 Springbett et al., 2010), potential reasons including differences in skin site, baseline 25(OH)D, UVR dose and emission spectrum (Bjorn, 2010).
1.3.2 Facultative pigmentation and epidermal thickening
Tanning produces facultative pigmentation, which is additional to the intrinsic skin colour. Delayed tanning involves spatial rearrangement of melanosomes, with the development of protective supranuclear caps of melanosomes over the basal keratinocyte nuclei, along with increased eumelanin synthesis as a photoprotective response to DNA damage (Eller et al., 1996). However, this melanisation is believed to be triggered by UVR-induced DNA damage (Young et al., 1998), thus suntanning intrinsically involves skin hazard. Skin thickening also occurs, with epidermal hyperplasia, which can involve both viable epidermal, and particularly stratum corneum thickening, which additionally provide photoprotection by decreasing the amount of UVR penetrating into the deeper layers of the epidermis/dermis (Lavker et al.1995; de Winter et al., 2001). UVB wavelengths may provide more effective photoprotection than UVA wavelengths, in view of a greater element of epidermal thickening (Rosen et al., 1987). There may also be skin type differences in tanning, as greater facultative pigmentation and a wider distribution of pigment is reported in people with darker skin (Duval et al., 2001).
However, overall an UVR-induced tan offers only a low sun protection factor (SPF), in the region of 3 to 4 at most (Cripps 1981). Therefore, what is sometimes popularly referred to as gaining a pre-holiday “base tan” provides only modest protection against the skin damage caused by further UVR exposure; rather appropriate clothing, sunscreens, and use of shade remain important for sun protection in lighter skin types.
32 1.3.3 Skin phototype
The skin phototype classification system was originally defined by Fitzpatrick in
1988. This crudely assigns individuals according to their sunburning and tanning histories along with their associated phenotypical characteristics into one of six categories, as illustrated in table 1.3. Phototype, therefore, is associated with an individual’s constitutive and facultative pigmentation.
Table 1.3: Fitzpatrick skin phototype definitions
Phototype Response to sunlight (UVR) exposure
I Always burns/never tans (light skin)
II Usually burns, sometimes tans (white skin)
III Usually tans, sometimes burns (white skin)
IV Always tans, never burns (olive skin)
V Brown skin
VI Black skin
Taken from Fitzpatrick, 1988.
1.3.4 Minimal erythemal dose
It is recognised that the Fitzpatrick phototype classification system (Fitzpatrick
1988) is deficient, in that white-skinned individuals are subcategorised into one of four groups dependent upon their tanning versus burning abilities, whilst brown and black skin hold only one group each. Indeed, any skin phototype may burn under specific conditions. A more accurate but time-consuming method to determine an individual’s UVR-sensitivity does exist, and involves the direct assessment of the skin’s erythemal (i.e. reddening) response to UVR in terms of
33 minimal erythemal dose (MED). The MED is the lowest dose of UVR required to give a ‘just-perceptible’ erythema 24-hours following the exposure (Harrison and
Young, 2002; see section 2.1.4.3).
1.4 Skin cancer
The major health risk to the population from cutaneous UVR exposure is skin cancer development.
1.4.1 Incidence
There is currently a surge in the incidence of skin cancer, making it the most common form of cancer in white-skinned individuals in the UK and many other countries where individuals of Northern European ancestry reside (Leiter and
Garbe, 2008; www.skincancer.org). Skin cancers can be categorised into melanoma and non-melanoma skin cancers (NMSC), the latter being specifically squamous cell carcinomas (SCC) and basal cell carcinomas (BCC).
In 2010, the national cancer registry cited 13,000 people in the U.K. as having been diagnosed with malignant melanoma, 100,000 with NMSC and nearly 3,000 deaths from skin cancer (Cancer Research UK, 2010); even these figures represent an under-estimate due to recognised widespread under-registration of cases.
Melanomas, SCC and rarely even BCC can metastasize and unfortunately cause death in addition to significant morbidity.
34 1.4.2 Risk factors
UVR exposure from sunlight is the most important risk factor for both melanoma and NMSC (IARC, 1992). In general, for melanoma high-risk behaviour includes intermittent high-level exposures particularly with sunburn at a young age, whilst
SCC risk reflects more an individual’s long-term chronic exposure, and BCC appears related to both sunburn and chronic exposure. However, it is evident that not all individuals demonstrate the same tendency towards skin cancer development. Additional risk factors in terms of patient phenotype are, therefore, highlighted below:
Patient characteristics associated with increased risk of skin cancer
o Pale skin: tans poorly, burns easily
o Red hair colour
o Multiple freckles/moles
o Personal or family history of skin cancer
o Sunburn at young age
o Sunbed usage
Adapted from Rigel DS, Dermatology; chapters 108 and 113; Bolognia, Rapini Eds
2008.
Inevitably, whilst some of these risk factors are avoidable, others represent an inherited predisposition and are at least partly characterised by Fitzpatrick’s phototype classification system. Amongst white skinned people, phototypes I and
II have a substantially higher risk of NMSC and melanoma than phototypes III and
IV.
35 1.5 Cutaneous Deoxyribonucleic acid (DNA)
A carcinogen is an agent/substance that can trigger cancer development. UVR fulfils this criterion by its efficiency in initiating the DNA damage within skin cells that can ultimately lead to mutagenesis and thus initiate skin cancer development.
In fact, UVR is classed as a “complete carcinogen,” by virtue of its dual action in both triggering such DNA damage (initiator), and in suppressing the body’s immune system to enable the propagation of cancerous cells (promoter).
1.5.1 Structure: Base pairing
Nucleotides are the building blocks of DNA. Each is composed of a base, a sugar (2- deoxyribose for DNA) and a phosphate group. (By contrast, the combination of base and sugar without the phosphate is a nucleoside.) Bases are nitrogen-based molecules that, according to their structure are classified into two subgroups, pyrimidines and purines. Usually, hydrogen bonds link pyrimidines to purines, i.e. cytosine to guanine, or thymine to adenine respectively (Table 1.5A), to form the well-recognised double-helix structure.
Table 1.5A: DNA complementary base pairs, i.e. purine to pyrimidine
PURINE PYRIMIDINE
Adenine (A) ------Thymine (T)
Guanine (G) ------Cytosine (C)
A key event in the pathogenesis of skin cancer is the development of DNA damage in skin cells exposed to UVR from sunlight/artificial sources.
36 1.5.2 Direct DNA damage
UVR is one of the most effective mutagenic agents to DNA: The absorption of UVB by the nuclei of skin cells can result in direct DNA damage with the formation of pyrimidine dimers, namely cyclobutane pyrimidine dimers (CPD), and pyrimidine-
6-4-pyrimidone photoproducts (6-4 PP). If not recognised and rectified by the DNA repair mechanisms (such as NER and mismatch repair enzymes), both CPD and 6-4
PP can halt the progress of the DNA polymerase enzymes essential for DNA formation, or alternatively result in misreading of the DNA (transcription errors) and cell death.
Although UVA is much more abundant than UVB at the Earth’s surface, it is poorly absorbed by cutaneous DNA and the exact mechanism of UVA-induced mutagenesis is still a matter of debate. Nevertheless UVA penetrates deep into the skin and remains capable of causing significant DNA damage, particularly via the generation of singlet oxygen that can indirectly damage DNA in addition to CPD
(Mouret et al., 2006).
1.5.2.1 Cyclobutane pyrimidine dimers (CPD)
CPD are the most mutagenic form of UVR-induced DNA damage, due to their high abundance, sluggish repair and their distinct mutagenicity (Yoon et al., 2000). They are formed when UV irradiation induces the formation of covalent bonds between two adjacent pyrimidine nucleotides, most commonly T-T, thus altering the DNA structure (Fig 1.5A).
37 1.5.2.2 Pyrimidine-6-4-pyrimidone photoproducts (6-4 PP)
Another detrimental effect of UVR on the skin is the formation of the 6-4 PP between adjacent pyrimidines (Fig 1.5A). These occur particularly when the base cytosine is located downstream to a pyrimidine (Mitchell, 1988). The rate of production of 6-4 PP is thought to be about 10% that of CPD generation (Franklin and Haseltine, 1984) and additionally, 6-4 PP are more readily repaired than CPD, and so have less impact on cell survival.
Figure 1.5A: Pyrimidine dimers: CPD (upper) and 6-4 PP (lower) are formed following UV irradiation of pyrimidine nucleotides.
Taken from www.photobiology.info
1.5.2.3 Oxidative DNA damage: 8-Oxo-2'-deoxyguanosine
Indirect (oxidative) DNA damage is caused both by UVB and UVA. Guanine is the most susceptible of the DNA bases to excitation by reactive oxygen species, as are generated following cutaneous UV irradiation, because it has the lowest ionisation potential. When oxidation occurs at the C8 (carbon-8) position of the nucleoside, 8-
Oxo-2'-deoxyguanosine (8-oxodG) is formed (Kasai et al., 1984; Fig 1.5B).
38
Figure 1.5B: Structure of 8-oxo-2'-deoxyguanosine (8-oxodG)
As the major product of DNA oxidation, and one of the most intensively studied oxidised DNA bases, 8-oxodG is consequently used as biomarker of a cell’s oxidative stress. 8-oxodG is additionally mutation-prone and can induce an inversion of the mismatch repair mechanism that normally acts to proofread DNA; it can pair not only with cytosine but also with adenine, inducing G:C to T:A transversions with high frequency (Shibutani et al., 1991). The abnormal 8- oxodG:dA, resembling a standard base pairing, is exempted from removal by the
DNA repair mechanisms (Hsu et al., 2004). Consequently, if 8-oxodG remains unrepaired before DNA replication, it is highly likely to result in this G:C to T:A transversion. Furthermore, given that these transversions have been commonly detected in human cancers (Kamiya et al., 1995), 8-oxodG is believed to contribute significantly to carcinogenesis and, accordingly, is also an important biomarker for assessing cancer-risk associated with exposures to carcinogens/environmental pollutants, including UVR exposure and smoking (Asami et al., 1997).
Pelle et al (2003) demonstrated dose-dependent increases in 8-oxodG with UVB irradiation in primary normal human epidermal keratinocytes (NHEK), as illustrated in figure 1.5C.
39
Figure 1.5C: UVB-induced 8-oxodG formation in normal human epidermal keratinocytes. p<0.001; *p<0.01, **p<0.002, ***p<0.001. Data expressed as mean +/-SEM; n=11, 2, 9, 5, and 4 for 0, 62.5, 125, 250, and 500 mJ/cm2, respectively. Reproduced from Pelle et al., 2003.
When the same group pretreated the keratinocytes with mannitol, a hydroxyl radical scavenger, the production of 8-oxodG was diminished by up to 80%, suggesting that hydroxyl radicals are necessary for UVR-induced oxidative DNA base damage.
1.5.3 Estimation of DNA damage by its urinary excretion
DNA repair mechanisms, including the nucleotide excision repair (NER) enzymes, are responsible for recognising and excising altered segments of DNA before DNA polymerase synthesizes a complementary strand to this excised region. In this way, the DNA damage is rectified and cell division can proceed. It is these excised segments that appear in the urine.
1.5.3.1 Urinary 8-oxodG
The repair products of oxidised DNA, being reasonably water-soluble, are generally excreted in the urine without further metabolism. Measuring urinary
40 excretion in this way represents the average rate of damage within that individual’s whole body (rather than direct skin sampling which pinpoints an exact time and site). Although measurements have shown large differences between techniques, they are now becoming more reliable and consensus between laboratories is being reached (Moller et al., 2012). 8-oxodG in particular has consequently been investigated as a urinary biomarker to permit examination of the impact of environmental measures, lifestyle and certain diseases upon oxidative stress, and to do so in a non-invasive manner.
Cooke et al (2001) were the first group to report the presence of 8-oxodG and T-T dimers in the urine of healthy humans following UVR exposure, as would be experienced for instance during a single therapeutic exposure to UVA for the treatment of psoriasis: Seven volunteers of skin type II (aged 23-56 years) were exposed to a suberythemal dose (15 J/cm2) of UVA (320-405 nm) and compared to an age and sex matched unexposed control group. First void, mid-stream urine samples were collected pre-exposure (i.e. day 0) and daily post-UVR, for 13 days for analysis of urinary 8-oxodG and thymine dimers (Fig 1.5D) by competitive enzyme-linked immunosorbent assay (ELISA).
41
Figure 1.5D: Competitive ELISA analysis of urinary 8-oxodG (left) and thymine dimers (right). *p≤ 0.05; **p < 0.001; ***p < 0.0001 Taken from Cooke et al., 2001.
The maximal increase in urinary 8-oxodG levels occurred 4 days following the UVR exposure, before gradually returning to baseline. However, a significant decrease in 8-oxodG compared to baseline occurred at day 10. The authors propose that this may be due to over-compensation by insult-induced DNA repair systems. T-T dimer levels similarly peaked on the third day post-exposure but in contrast to the
8-oxodG analysis, a second peak was noted at days 9-11, before again returning to baseline. The authors hypothesised that these two peaks were accountable by two
42 separate repair mechanisms, one being more rapid than the other. It is proposed that such measurements could be used as a noninvasive method for biomonitoring
UVR exposures, and specifically psoralen-UVA (PUVA) treatments, which are themselves known to be potentially carcinogenic. Although this group was the first to quantify direct and oxidative DNA damage in the urine of healthy humans, their study only examined for UVA-induced damage and, as stated above, UVB is much more mutagenic than UVA.
1.5.3.2 Urinary CPD
In a more recent study, Liljendahl et al (2012) examined urinary thymine dimer levels in Swedish children and adults following recreational sunlight exposure, quantified by personal dosimeters and diaries detailing skin exposure and sunscreen usage. Twelve children and 11 adults spent 1-2 days on a Swedish beach in July-August. Urine samples were collected pre-exposure and for 3 days post- exposure for analysis of DNA damage (Fig 1.5E).
Figure 1.5E: Urinary thymine dimer levels before (day 0) and for 3-4 days after sunbathing on the beach for 1–2 days (day 0 and 1). *p<0.05, **p<0.001 in comparison to day 0. Taken from Liljendahl et al., 2012.
43 Results confirmed significantly increased urinary thymine dimer levels above baseline following the sunbathing. Recorded levels were maximal on the third day post-sunbathing in a similar pattern to the results of Cooke et al (2001) described above, where the maximal level of damage was detected on the fourth day. It is unfortunate that Leljendahl et al do not have longer follow-up data, especially to identify whether any secondary peaks in DNA damage exist.
As duration of sunlight exposure varied considerably between volunteers, from 2-
14 hours, UVR dosage was calculated from personal dosimeters and adjusted for skin surface area exposed. UVR dosage correlated strongly with urinary thymine dimer levels, for both adults and children (Fig 1.5F).
Figure 1.5F: Urinary thymine dimer excretion as a function of UVR dose (adjusted for area of skin exposed), for A) adults and B) children. Both associations were statistically significant. Taken from Liljendahl et al., 2012.
44 In this study, individuals were requested to refrain from sunbathing for only one week prior to the investigations. As Cooke et al (2001) demonstrated a time-scale of 11 days for urinary DNA damage to decline to baseline following UVR exposure, it is possible that Liljendahl et al’s results are confounded by previous sunlight exposures, and this may also account for the disparities in baseline thymine dimer levels between individuals.
Additionally, as three of the adult volunteers received their two days of sunlight exposure split over consecutive weekends, their UVR doses and urinary results were therefore pooled over the two weekends. Such results may not be representative of the levels of DNA damage that would be experienced following exposures on two consecutive days. There is thus much scope for further investigation of urinary DNA damage (direct and/or oxidative) following interventional (quantified) and natural UVR exposures.
A positive association between urinary T-T and vitamin D gain has been reported three days following six days of high-level sunlight exposures in individuals of phototypes I-IV (Petersen et al., 2014). That study looked at intense UVR exposures experienced during sun (28°N 16°W) and ski holidays (mean UVR 60-
101 KJ/m2). However, the effects of low-level (suberythemal) casual sunlight exposures were not explored.
1.5.4 DNA damage and repair
Perhaps equally pertinent to skin cancer risk as DNA damage is DNA repair, that is, the extent to which the altered DNA is recognised and subsequently rectified.
Indeed, the inherited disorder xeroderma pigmentosum epitomises the role of
45 sunlight-induced DNA damage in the development of skin cancer. In this disorder, the enzymes needed to repair UVR-induced DNA damage are mutated and thus are non, or only partially, functional. Consequently, melanoma and NMSC develop to such an extent that affected individuals rarely live beyond the age of 20 years, without strict sunlight avoidance (DiGiovanna and Kraemer, 2012).
Chouinard et al (2001) examined UVR-induced DNA damage/repair in vitro, scrutinising the effects of 20 seconds of low-level UVB irradiation (8 mJ/cm2, about two-fold lower than the MED for skin phototype I) on human keratinocyte cultures: After one irradiation, they identified the presence of both CPD and, to a much lesser extent, 6-4 PP. Twenty four hours after the irradiation, most 6-4 PP had been repaired, in contrast to very few CPD. After these low-level exposures were repeated for 8 consecutive days, CPD actually accumulated whereas 6-4 PP were efficiently repaired, and remained at a lower level than that provoked by one single UVB exposure (Table 1.5B).
Table 1.5B: Quantification of CPD and 6-4 PP in UVB-irradiated human keratinocytes by radioimmunoassay.
Treatments Photoproduct frequency* (No/mb)
CPD 6-4 PP No UV 4.2 +/- 1.0 2.5
UVB 66.8 +/- 17.5 13.8 +/- 1.3
UVB and 24 h repair 59.4 +/- 9.9 6.1 +/- 0.1
8 days UVB and 24 h repair 160.8 +/- 15.8 3.9 +/- 0.9
Mean +/- SD of three experiments conducted with cell isolated from different skin donors. UVB dose was 8 mJ/cm2; mb denotes millibase of DNA Reproduced from Chouinard et al., 2001.
46 This group, therefore, concluded that repeated exposure to suberythemal doses of
UVB cause serious, unrepaired DNA damage that, if accumulated over time, could lead to the development of skin cancer. Murine models exploring the effects of repeated low-level UVB exposures (0.5 kJ/m2 every 24 hours for 40 consecutive days) add support to this hypothesis: Excisional repair of these CPD and 6-4 PP became significantly reduced following chronic exposures, so resulting in an accumulation of increasing irreparable DNA damage in the dermis and epidermis induced by each low-level challenge (Mitchell et al., 1999).
In this way, chronic low-dose exposures to sunlight may significantly enhance the predisposition of mammalian skin to such sunlight-induced carcinogenesis.
Furthermore, when UVB-induced skin tumours in mice were scrutinised for DNA damage, 20% had base substitutions, all at dipyrimidines, and these were usually in the form of C-T or CC-TT substitutions in the p53 tumour suppressor gene
(Kress et al., 1992).
Actinic keratoses are pre-malignant lesions commonly found on the sun-exposed skin of older white Caucasians in particular. They are indicative of an individual’s preceding UVR exposure, and can progress into invasive skin cancer if untreated.
Studies of such lesions can therefore provide insight into cutaneous carcinogenesis: Mutations within the p53 tumour suppressor gene of actinic keratoses have been identified, particularly as C-T substitutions or CC-TT double base changes in adjacent pyrimidines (Ziegler et al., 1994), consistent with the above findings in murine skin cancers. When invasive skin cancers on sun-exposed sites of human skin were evaluated, mutations in p53 dipyridimines were similarly detected with C to T predominating 62% of mutations (Brash et al., 1991).
47 Consequently, these C-T and CC-TT transitions that are characteristic of UVR- induced damage are also known as ‘UVR signature mutations’ (Wikonkal and
Brash 1999).
In terms of DNA damage/repair in individuals of different skin phototypes,
Sheehan et al (2002) irradiated healthy human volunteers with 0.65 x MED of solar-simulated radiation (SSR) daily (Monday to Friday) for two weeks. Of these, six individuals were of phototype II and six of phototype IV; the individuals of phototype IV accordingly received a higher cumulative physical dose of UVR (but the cumulative biologic dose was the same for each volunteer, i.e. 10 doses, each of
0.65 x MED). A comparison of pre- and post-irradiation biopsies confirmed DNA damage even at these suberythemal doses. Moreover, levels of thymine dimers appeared higher in the individuals of phototype IV, but differences were not found to be significant. In the phototype IV group, 69% of the DNA damage present on the day of final exposure had been repaired by one week post-irradiation compared to 37% in the phototype II group; the decrease was only found to be significant in the phototype IV group. These results imply that physical rather than biological dose may determine the degree of DNA damage and also suggest that more resistant skin types may be more efficient at repairing damaged DNA.
1.6 UVR-induced erythema
UVR-induced erythema, known colloquially as ‘sunburn’, is the reddening of the skin from excessive UV irradiation; it is part of an acute inflammatory response and consequently the skin can also become painful, hot and swollen with increasing UVR exposure. The action spectrum of erythema induction is very similar to that of CPD induction (see section 1.5.2), in-keeping with the hypothesis
48 that UVR-induced DNA damage is its initiator (Parrish et al., 1982; Young et al.,
1998). Thus UVB is much more effective than UVA at inducing erythema (Willis and Cylus, 1977) and, consequently, the UVB component of sunlight, although smaller, is actually responsible for the vast majority of sunburn.
This acute inflammatory response to UVR is secondary to the complex release of many inflammatory mediators and sequential eicosanoid profiles (Strickland et al.,
1997; Rhodes et al., 2001; Rhodes et al., 2009). It includes UVR–induced hydrolysis of membrane phospholipids and dermal mast cell degranulation (Walsh et al.,
1991; Norris et al, 2002), which result in the cutaneous vasculature dilation that produces erythema. Chemotactic factors including interleukin (IL)-8, leukotriene-
B4 and complement factor C5a promote neutrophil adherence to the vascular endothelium, diapedesis and their migration towards the area of UVR-induced damage. Whilst the time course of the erythemal response is modified by UVB dosage, it is usually maximal around 24 hours post-exposure, accompanied by peak neutrophil influx at this time (Lee et al., 2008). Functions of the neutrophils include release of reactive oxygen species (ROS) and proteolytic enzymes, such as neutrophil elastase and matrix metalloproteinases (MMP) that can degrade cutaneous elastin and collagen fibres, as well as the phagocytosis of damaged cells to assist recovery (Hawk et al., 1988; Shapiro 2002; Lee et al., 2008).
Lymphocytes and macrophages also infiltrate the area, predominantly between 24 and 72 hours post-exposure (di Nuzzo et al., 1996). The T cells are mainly of Th2 subtype (see section 1.8.1) and the macrophages, in addition to their role in phagocytosis, release immunomodulatory mediators including IL-10 that acts to inhibit cell-mediated immune responses. Neutrophils themselves also release such
49 Th2-associated cytokines as IL-4 and IL-10, which act in a predominantly immunosuppressive manner to ultimately dampen the inflammatory response
(Teunissen et al., 2002; Piskin et al., 2005). Together these factors contribute to a coordinated cascade of cytokine expression that control immune and inflammatory responses to UVR-insult.
1.7 Photoageing
Sun exposure has a cumulative effect on the skin, resulting in changes that are superimposed on the appearance of intrinsically aged skin. This process of photoageing is manifested by a variety of clinical signs including wrinkles
(rhytides), dyspigmentation, telangiectasia, laxity, a leathery texture and a yellow hue (NRPB 2002). Both direct and indirect effects of UVR are believed to contribute to these changes, including UVA and UVB-induced activation of MMP to degrade matrix proteins, particularly via induction of the mitogen-activated protein kinase (MAPK) pathway (Fisher et al., 1996). UVR may also damage mitochondrial DNA (mtDNA; Birch-Machin et al., 1998) directly or indirectly, augmenting the production of ROS (Rizwan et al., 2011) by the mitochondrial
‘power houses’ (Birch-Machin and Swalwell, 2010). ROS may themselves damage mtDNA, and consequently lead to further mitochondrial dysfunction and ROS generation in a vicious cycle that may contribute to photoageing processes
(Hudson et al., 2016). Histologically, there is an accelerated loss of collagens I and
III (El-Domyati et al., 2002) with prominent elastosis, i.e. an abnormal accumulation of fragmented elastic fibres within the dermis (Kligman, 1969), increased thickness of the basal membrane and chronic inflammation (Lavker and
Kligman, 1988).
50 It is postulated that the higher levels of neutrophils seen to be residing in chronically photoexposed mouse skin (Takeuchi et al., 2010) may contribute to the photoageing process via their release of proteolytic enzymes, particularly neutrophil elastase and MMP, that can damage elastin fibres and collagen networks. Overall, the depth and severity of dermal changes reflect the amount of
UVR damage.
1.8 UVR effects on immunity
1.8.1 Photoimmunosuppression
Sunlight exposure has direct effects on both the skin’s innate and adaptive immune systems. An example of the immunosuppressive aspect of UVR’s action on the skin is seen in those susceptible individuals who develop cold sores following sunlight exposure: The UVR has suppressed the adaptive immune system such that the cold sore (Herpes simplex) virus is able to replicate and cause clinical disease. The immunosuppressive effects of UVR can increase skin cancer risk in a similar fashion to the increased risk seen in transplant recipients in whom, as a result of their immunosuppressive medications, skin cancers present much more commonly
(Bouwes Bavinck et al., 1996).
The key molecular trigger for UVR-induced immunosuppression is believed to be
UVR-induced DNA damage (Kripke et al., 1992). UVB has a potent immunosuppressive action on cell-mediated immunity, particularly by the induction of regulatory T cells (Treg; Schwarz and Schwarz, 2011). These cells are lymphocytes that suppress activation of the immune system and, therefore, switch off the immune response once an invading microorganism, for instance, has been
51 eliminated. UVR-induced Tregs express the protein markers CD4 and CD25 (hence their alternative name of CD4+ CD25+ Treg), and upon antigen-specific stimulation secrete IL-10 (Schwarz, 2008). IL-10 is an essential mediator of UVR-induced immunosuppression and is additionally secreted from irradiated keratinocytes; pretreatment of mice with an IL-10 blocker prevents photoimmunosuppression from UV irradiation (Rivas and Ullrich, 1992).
Those T cells expressing CD4 are also known as helper T cells. They can be further subdivided into Th1 and Th2 groups. In brief, with Interferon-gamma (IFN-γ) as the main Th1 cytokine, Th1 responses are mainly proinflammatory, killing intracellular parasites. Inevitably, excessive proinflammatory responses could lead to uncontrolled tissue damage, and so to counteract this, the Th2-type cytokines include amongst others IL-10, which has more of an anti-inflammatory response.
Overall, UVR exposure results in a shift in the activation of T cells from Th1 to Th2 immune response via alteration of antigen presenting cell function (APC), specifically Langerhan cells (LC), whose emigration from the epidermis is characteristic for photoimmunosuppression (Schwarz, 2005). Similarly, UVR- irradiated LC present only to Th2 (Simon et al., 1990) and treatment of LC with IL-
10 blocks antigen presentation to Th1 and induces tolerance in within these Th1 cells (Rivas and Ullrich, 1992).
1.8.2 Antimicrobial peptides
In direct contrast to the provocation of labial herpes following UVR exposure, reports describing secondary infection after sunburn from prolonged sunbathing, or prescribed UVR-phototherapy treatments for skin disease, are virtually non-
52 existent. The question then arises how this is possible, given that UVR can impair the skin barrier and is an effective immunosuppressant?
Recent work has demonstrated the direct upregulation of antimicrobial peptides
(AMP) in healthy skin by cutaneous UVR exposure (Glaser et al., 2009; Felton et al.,
2013). AMP are small peptides that are key components of the innate immune system, originally identified by their vital role in protecting the body-environment interface from infection. However, it has now become clear that AMP have more extensive actions, including the provision of pivotal links with the adaptive immune system. Their upregulation may, therefore, serve to protect the skin from certain risks imposed by both the biophysical barrier-compromise and the cell- mediated immunosuppression that are attributable to UVR exposure (Fig 1.8).
Figure 1.8: Dual action of UVR on innate and adaptive immune systems.
Upregulation of antimicrobial peptides (AMP) may potentially serve to counteract the promotion of the Th2 responses which themselves have an anti-inflammatory effect.
53 1.9 The endocannabinoid system
The endocannabinoid system encompasses the physiological ligands for cannabinoid (CB) receptors; these G-protein coupled receptors were originally identified as the target for the biologically active components of the cannabis plant
(Matsuda et al., 1990; Munro et al., 1993; Howlett, 2002).
1.9.1 Anandamide and 2-arachidonoyl glycerol
Physiological endocannabinoids include the endogenous lipid agonists of CB receptors, the best-characterised of which are anandamide (N-arachidonoyl ethanolamine, AEA) and 2-arachidonoyl glycerol [2-AG; Fig 1.9A]). These are signalling molecule derivatives of arachidonic acid (Mechoulam et al., 1998;
Mechoulam and Hanus, 2000; Bisogno, 2008). Their main actions are reduction in pain perception via CB1 and transient receptor potential vanilloid-1 receptors
(Guindon and Hohmann, 2010), and predominantly anti-inflammatory/immune- modulatory effects via peripheral CB2 receptors with reduced IFN-δ and increased
IL-10 secretion (reviewed by Boorman et al., 2015).
Figure 1.9A: Chemical structures, anandamide (left) and 2- arachidonoylglycerol (right).
1.9.2 The cutaneous endocannabinoid system
CB1 receptors were originally identified in the central nervous system and CB2 peripherally in immune cells, but recently it has become evident that their
54 distribution is more variable and widespread throughout organ systems, including the skin (reviewed by Biro et al. 2009): Human keratinocytes are themselves able to synthesise endocannabinoids and, similarly, endocannabinoid binding to keratinocyte CB1 receptors regulates cell growth, function and death (Pertwee,
2006; Paradisi et al., 2008). Moreover, human melanocytes have been shown to possess fully functional endocannabinoid systems, including expression of AEA, 2-
AG and their cannabinoid receptors CB1 and CB2. Indeed, at low concentrations,
AEA upregulated tyrosinase (the gene responsible for melanin synthesis) via CB1, whilst at high doses, AEA actually induced DNA fragmentation, expression of the key tumour suppressor gene p53, and melanocyte apoptosis (Pucci et al., 2012).
1.9.3 Involvement in skin disease
The endocannabinoid system is proposed to have a number of functions within the skin, including mediation of pain, itch and inflammation via CB and possibly other receptors, such as transient receptor potential cation channels in sensory nerve terminals (Guindon and Hohmann, 2010). Indeed disruption of the endocannabinoid system is thought to contribute to a number of skin diseases including acne, eczema, psoriasis and hair disorders (reviewed by Biro et al. 2009).
More specifically, in vivo mouse studies have shown that genetic deletion or pharmacologic blockade of keratinocyte CB1 and CB2 receptors enhances allergic contact dermatitis, whereas their activation by increased AEA or synthetic agonists reduces this inflammation (Karsak et al., 2007). Furthermore, cannabinoid receptor (CB1/CB2)-deficient mice have reduced UVR-induced activation of MAPK and Nuclear Factor-κB, accompanied by increased skin cancers (Zheng et al.,
2008), suggesting that endocannabinoids may be protective against cutaneous carcinogenesis.
55 1.9.4 Response to UVR
An involvement of endocannabinoids in cellular responses to external stresses, such as UVR exposure, was suggested by their increased production (including of
AEA) following the UVB irradiation (4 KJ/m2 for 16 minutes) of mouse epidermal cells in vitro (Berdyshev et al., 2000). Similarly, Magina et al (2011) confirmed increased endocannabinoid levels in human keratinocyte cell lines following UVB exposure (Fig 1.9B): At baseline (pre-irradiation) low levels of endocannabinoids were present. Low-level (30 mJ/cm2) UVB irradiation increased AEA to an insignificant degree, whilst high-level (60 mJ/cm2) irradiation caused significant increase. In contrast, 2-AG was significantly increased and to a greater extent than
AEA by both UVB doses.
Figure 1.9B: Effect of low- and high-dose UVB irradiation on a) AEA and b) 2-
AG in human keratinocyte cultures in comparison to control (CT) p<0.05 in comparison to non-irradiated control. Taken from Magina et al., 2011.
56 A study quantifying serum endocannabinoid levels in humans following a six-week course of UVB phototherapy treatment for the skin condition psoriasis identified a significant decrease in plasma AEA, implying an UVR-responsive mechanism
(Magina et al., 2014). However, a recent exploratory study examining their cutaneous expression in skin biopsies taken 24 hours after a single exposure of 2x
MED of principally UVB (275-380, peak 305nm) to a localised area of the buttock, found no alteration (Kendall et al., 2016), suggesting that repeated or larger surface area exposures may be necessary for quantifiable endocannabinoid responses. Alternatively since endocannabinoid levels may be altered by skin conditions including psoriasis and itching (Wilkinson et al., 2007; Sugawara et al.,
2012), Magina et al’s results may be confounded in comparison with healthy volunteer studies.
1.9.5 Involvement in mood
Involvement of the endocannabinoid system in control of mood was suggested following its identification within the central nervous system and by virtue of the
‘feel-good’ effects of cannabis itself on mood. In support of this, models of depression in rat studies have identified reduced AEA levels and differential changes in CB1 receptor binding density within the brain (Hill et al., 2008).
Similarly, endocannabinoid system activation mediated anti-depressive effects through activation of CB1 receptors in further rat models (Adamczyk et al., 2008).
Studies in patients with untreated depression revealed significantly reduced serum 2-AG and AEA levels compared to controls (Hill et al., 2009). One possible mechanism underyling mood modulation by endocannabinoids may be their activation of CB1 receptors on brain GABA-ergic neurons, increasing dopamine
57 release in central reward centres and causing mood enhancement (Lupica 2005).
Other factors such as increased neutrophin brain-derived neurotrophic factor, and peripheral CB1 and CB2 receptor activation may also be involved (Heyman et al.,
2012; Fuss et al., 2015).
Similarly, involvement of the endocannabinoid system in the ‘runner’s high,’ (i.e. mood elevation from exercise) has also been proposed: Sparling et al. (2003) demonstrated that 50 minutes of moderate exercise in 16 healthy male college students resulted in significant elevation in serum AEA levels (accompanied by a similar but non-significant rise in 2-AG), in comparison to eight sedentary controls.
Raichlen et al., (2013) subsequently monitored serum AEA levels in ten healthy, regular runners after walking or running on the treadmill for 30 minutes at differing intensities. They found that moderate intensity exercise significantly elevated serum AEA over pre-exercise levels, but low and high intensity training did not. The authors postulated that low and high intensity training may not associate with mood enhancement, although mood itself was not assessed.
By virtue of their widespread distribution in body tissues and range of biological activities, there is scope for the further investigation of endocannabinoid involvement in human responses, including to external insults affecting the skin.
1.10 Psychological effects of sunlight exposure
Whilst not an established health benefit of sunlight exposure, people commonly report that they ‘feel good’ when in the sun. It is unclear whether this may be attributable to UVR or visible radiation, or even an infrared/heat-related effect.
58 The pathway underlying this mood enhancement is unknown, although this could potentially be conveyed through cutaneous hormone synthesis and/or visually.
1.10.1 Seasonal affective disorder
Seasonal affective disorder (SAD) or the “winter blues,” is a commonly recognised psychological condition in the Diagnostic and Statistical Manual of Mental Disorder
(DSM-IV, American Psychiatric Association, 2000). Sufferers, who have normal mental health throughout most of the year, report a depressed feeling when they have not experienced enough sunlight; however most evidence points to the likelihood of a visible radiation effect (Lurie et al., 2006). Although there exist no firm links between vitamin D deficiency and SAD, the former has been associated with both low mood and deterioration in cognitive performance (Wilkins et al.,
2006). Treatments applied for SAD have particularly included bright light therapy i.e. visible radiation, with effects postulated to operate through visual (rather than cutaneous) pathways, spending more time outdoors, and also vitamin D supplementation, melatonin and anti-depressant medication (Lurie et al., 2006).
1.10.2 Dependency on suntanning
Despite ongoing efforts to educate the public on the dangers of UVR exposure, specifically a heightened risk of skin cancer, recreational sun tanning behaviour continues to increase. It has therefore been hypothesised that there may be an addictive component to tanning. Studies have indeed confirmed that repeated UVR exposures may result in similar behaviour patterns to those seen in other forms of substance-related disorders (SRD), such as alcohol and drug abuse, as reported below. These behaviours have been commonly assessed by two validated written
59 measures used to evaluate SRD, namely the CAGE (Cut down, Annoyed, Guilty, Eye opener) questionnaire and the DSM-IV criteria.
In 2005, a survey of 145 beachgoers in Texas highlighted 26% of participants as fulfilling the CAGE and 53% the DSM-IV criteria for SRD (Warthan et al., 2005).
Addictive behaviours included the continuation of tanning despite attempts to stop, persistent tanning in the presence of adverse consequences and the neglect of other responsibilities in order to maintain a tan. Similar results have been demonstrated in other studies, including one of 229 college students in New York in 2006, which demonstrated that 39% met DSM-IV criteria and 31% CAGE criteria for addiction to indoor tanning. Interestingly, these individuals also reported greater symptoms of anxiety and greater consumption of alcohol and marijuana
(Mosher and Danoff-Burg, 2010).
Harrington et al (2011) found comparable results in their study and also demonstrated that early age at first tanning and female gender were more likely to be associated with addiction. Moreover, when exposed to UVR and non-UVR emitting sun beds under blinded conditions, frequent tanners can distinguish between these and actively choose to receive further UVR exposure from only
UVR-emitting booths (Feldman et al., 2004). These results suggest, therefore, that
UVR itself is a reinforcing stimulus.
1.10.3 Opioids and Suntanning
Reported motivations for tanning behaviour include increased attractiveness, relaxation and mood enhancement (Robinson et al., 2008); most individuals would state that sunlight exposure, and having a suntan gives them a ‘feel good’ factor
60 (Schneider et al., 2012). This has been suspected to be secondary to cutaneous production of endorphins, endogenous opioids that are produced in the hypothalamic-pituitary axis and are increased after exercise, so potentially contributing to the “runner’s high” (Carr et al., 1981). In line with this hypothesis, in 1983 Levins et al demonstrated a significant increase in plasma β-endorphin and its immediate precursor β-lipotrophin, after repeated broad-band UVA (320-
400 nm) exposures. However, subsequent studies have shown inconsistent results:
Belon (1985) confirmed significantly increased β-lipotrophin following UVA exposures that was blocked by sunscreen usage but more recently studies by
Wintzen et al (2001) and Gambichler et al (2002) did not find elevations in β- endorphin following UVA or UVB exposures, although neither study performed β- lipotrophin assays.
More recently, Fell et al (2014) demonstrated that repeated low-dose UVR exposures (equivalent to 20 minutes of Florida sunlight 5 days per week for 6 weeks) significantly elevated plasma β-endorphin levels in rodents after one week of exposures, and that administration of naloxone (a blocker of endogenous opioid receptors), resulted in symptoms of withdrawal. It is, however, somewhat difficult to draw comparisons between humans and murine models of UVR exposure, particularly with the latter being nocturnal and possessing furry skin.
Nevertheless, a human trial of eight frequent and eight infrequent tanners who were pretreated with naltrexone (another blocker of endogenous opioid receptors) was halted prematurely when two of the three frequent tanners (but none of the infrequent tanners) developed opioid withdrawal symptoms including nausea, jitteriness and shaking. The frequent tanners also showed preference for
UVR-emitting sunbeds over non-UVR sunbeds when pretreated with placebo, in-
61 line with the results of Feldman et al (2004). Similarly, following the opioid blockade by naltrexone, preference for UVR decreased (Kaur et al., 2006). Taken together, such results might imply not only addictive responses to tanning, but also involvement of endogenous opioids, although the latter have been challenging to directly demonstrate.
1.11 Artificial Sunlight
1.11.1 Sunbeds
Tanning lamps have been in existence for nearly a century and their basic structure has remained fairly constant over this time. Sunbeds use fluorescent bulbs that emit UVA and UVB in varying proportions, often around 97% UVA/3%
UVB. Smaller tanning beds designed for at-home usage consist of fewer lamps, these being of lower wattage (W [i.e. Joules/second]; often 12 to 28 lamps of 100
W each) than those frequently found in tanning salons (which may possess around
24 to 60 lamps of 100-200 W each). There additionally exist newer “high pressure” tanning beds which use fewer lamps and have a greater UVA:UVB ratio with claims by the tanning industry to provide a faster suntan (www. europeantanning.com).
The surface that physically separates the user from the underlying tanning apparatus is an UVR-transmitting polymethyl methacrylate with specific spectral transmittance between 290 nm and 400 nm (i.e. UVA and UVB). These materials are subject to deterioration and, as is the case for the tanning bulbs, without regular maintenance or replacement, will affect the radiation penetrating to the user. For these reasons, it can be difficult to identify the exact UVR dose (UVB and
UVA) that has been received in commercial tanning sessions. UVR exposure from indoor sunbeds heightens skin cancer risk, with as few as ten tanning sessions
62 being strongly linked to the development of melanoma, and usage under the age of
30 years in particular associated with a significantly increased risks of both melanoma and non-melanoma skin cancers (Colantonio et al., 2014; Wehner et al.,
2014). Consequently, sunbeds and other UVR-emitting tanning devices are classified in the top carcinogenic category (Group 1) to humans (IARC, 2009). In
2009, Brazil was the first country to place a nationwide ban on sunbeds, and several states in Australia are now following suit (Sinclair et al., 2014). In the UK, as in many other countries and some states in USA, usage by minors under age 18 years is prohibited (DH, 2011).
1.11.2 Psoriasis and phototherapy
Psoriasis is a skin condition known usually to improve with sunlight exposure.
Indeed phototherapy is a form of medically prescribed UVR treatment commonly used in the treatment of psoriasis. It is usually given in the form of narrow-band
UVB phototherapy, specifically 311 nm, administered three times a week for six to ten weeks at gradually increasing but measured doses (to remain suberythemal), until skin clearance or plateau of response (Ibbotson et al., 2004). In this closely regulated manner, efforts are made to reduce the cumulative UVR dose and minimise an individual’s future risk of skin cancer.
Given psoriasis improvement with UVR, there exists the possibility that some sufferers may choose to self-treat their psoriasis with sunbeds rather than attend a medical setting for prescribed and controlled phototherapy. Studies have suggested that around 15% to 20% of the general population use sunbeds (Boyle et al., 2010; Guy et al., 2014). However, the prevalence of sunbed usage in psoriasis patients has not been investigated, despite the potential for their increased usage.
63 1.12 Guidance on sunlight exposure
Since the incidence of skin cancer continues to rise in white-skinned populations, these neoplasms are a significant health problem, both in terms of patient morbidity and healthcare expenditure. Therefore, national campaigns are in place to highlight to the general public the dangers posed by UVR exposure, specifically heightened skin cancer risk, with the aim of improving their sun–protective behaviour.
1.12.1 National campaigns
The SunSmart campaign, headed by leading UK cancer charity Cancer Research UK, began in 2003 (Cancer Research UK: Sunsmart). It was designed to educate the public on reducing their sun exposure, particularly in the summer when UV index
(i.e. the measurement of the strength of the UVR from the sun at a specific place on a particular day) is highest. Its key messages are to seek shade, avoid the midday sun, cover up with clothing and use sunsceen to reduce the risk of sunburn and consequently skin cancer. Other charities and health promotion groups, including the British Association of Dermatologists’ Sun Awareness Campaign and its Mole and Sun Advice Roadshow, similarly support this strategy (BAD 2012). Their objectives are to “make even more people aware of skin cancer risks and how to enjoy the sun more safely”.
The Health Protection Agency (HPA), previously known as the National
Radiological Protection Board (NRPB), designed a Sunsense guide (HPA, 2009) to educate regarding sun-safe behaviour. Its consensus is consistent with their report detailing the ‘Health effects of Ultraviolet Radiation’ (NRPB 2002), recommending that casual exposures of limited areas of skin to sunlight several times a week,
64 whilst avoiding sunburn, are adequate to maintain vitamin D status without the need to deliberately sunbathe. However, these statements were based on projections from the data of limited research, and on conclusions that 25(OH)D levels of 5-10 ng/ml or above are adequate for health benefit, whereas the
USA/Canadian and European agencies now recommend that 20 ng/ml should be reached to achieve sufficiency (IOM 2011; European Food Safety Authority 2016).
Moreover, the advice on sunlight exposure has been geared towards white-skinned individuals, with little specific information or advice for the UK’s increasing proportion of darker-skinned populations. Indeed, there are anecdotal reports of darker skin type people following the recommendations for lighter skin types, with potentially deleterious impact on health.
1.12.2 Apparent discordance between national guidance and health benefits
National guidance on summer sunlight exposure therefore emphasises the importance of reducing sun exposure and usage of photoprotection, in order to decrease the risk of skin cancer. Such measures could potentially increase the risk of low vitamin D status. Nevertheless, the main message is to avoid over-exposure to UVR, i.e. exposures resulting in sunburn (skin reddening, as a surrogate for DNA damage and skin cancer), and more than mild tanning. Since vitamin D is formed following short exposures to UVR, longer exposures serve only to increase risk of skin cancer, without the increasing benefit of further vitamin D formation.
However, it is clear that the “one size fits all” approach to sunlight exposure recommendations is unlikely to serve the UK public well, bearing in mind the increasing proportion of people with pigmented skin.
65 1.13 Aims of this thesis
I hypothesise that repeated suberythemal UVR exposures, despite being low-level, provide both health benefit and health detriment. The overall objective of this study is, therefore, to evaluate indicators of beneficial and hazardous health effects of cumulative low levels of sunlight exposure to human skin. Such effects shall additionally be compared between white Caucasians of phototype II, and South
Asians of phototype V. I predict that the lighter skin type subjects, having less constitutive melanin, will exhibit evidence of greater health benefit, but simultaneously larger health risk from these sunlight exposures.
My detailed aims are to examine:
1. The principal risk and benefit of summer sunlight exposures in terms of the
UVR-induced DNA damage and the 25(OH)D gain concurrently attained in
skin type II (white-skinned) and V (brown-skinned) individuals.
2. Photoprotection by summer sunlight exposures in skin type II individuals,
in terms of skin darkening and epidermal thickening responses, and
subsequent protection conferred against sunburn inflammation, assessed
clinically as erythema and histologically by neutrophilic infiltration.
3. Potential for mood modulation by summer sunlight exposures, in terms of
circulating endocannabinoid levels, in skin types II and V.
4. The addictive behavioural component to tanning, namely why some
patients continue to seek UVR exposure despite knowledge of the adverse
consequences including skin cancer.
To fulfil aims 1, 2 and 3, I shall analyse the effects of an intervention study, in which individuals of phototypes II and V were exposed to the same measured, absolute, low doses of UVR. The UVR course was designed to simulate a summer’s
66 sunlight exposures, as could be experienced if following the advice given through national policy for white-skinned people. Circulating levels of 25(OH)D and endocannabinoids will thus be measured, the tanning and photoprotection conferred will be evaluated through non-invasive and histological assessments, and cutaneous and urinary DNA damage/repair will be quantified pre-, per- and post-UVR exposure in all volunteers. To fulfil aim 4, I will conduct a nationwide cross-sectional survey designed to examine suntanning histories and, using validated questions adapted from the Diagnostic and Statistic Manual of Mental
Disorders, any addictive behavioural components.
67 Contribution of the author and her co-authors
I was the lead investigator and main author on all papers. Prof Rhodes, my MD supervisor, was the principal supervisor of all projects and manuscripts, with the exception of manuscript 4, where the study was performed during my field trip to
Dallas, USA, and along with the publication, was supervised by Prof Jacobe and Prof
Adinoff.
For manuscripts 1, 2 and 3, the research was designed by Prof Rhodes, whilst I, myself, wrote the detailed protocols and gained the ethical approvals. I recruited the volunteers, consented them, and performed the clinical interventions and invasive procedures. I prepared the skin biopsies and performed the immunohistochemical studies (manuscripts 1 and 2). For manuscript 4, the research was designed by Prof Jacobe together with myself. I wrote the protocol and gained ethical exemption approval by the Institutional Review Board. I arranged approval by the National Psoriasis Council (USA) for the research to be performed and for them to disseminate the survey. I designed the survey on
Survey Monkey® and analysed all the responses including calculation of the
Simplified Psoriasis Index. I performed all data analyses (manuscripts 1-4) and wrote all the manuscripts (1-4), making adjustments according to my co-authors’ comments.
I acknowledge and am grateful for the input of my other co-authors, who in addition to their input to relevant aspects of study design and their comments on the manuscripts, contributed the following specific inputs: Dr Berry for performance of the 25-hydroxyvitamin D assays (manuscript 1); Prof Cooke for supervision of direct and indirect urinary DNA damage analysis and Dr Lam for
68 mass spectrometric analysis of urinary 8-oxodG (manuscript 1); Prof Vail
(manuscripts 1 and 3) and Prof Jeon-Slaughter (manuscript 4) for statistical analysis and research design advice; Dr Watson and Dr Shih for assistance with skin section assessment (manuscript 2); Prof Nicolaou, Dr Urquhart, Dr Almaedani and Dr Kendall respectively for supervision and performance of endocannabinoid assays (manuscript 3); Prof Webb and Dr Kift for input to the cabinet irradiation spectroradiometry and performance of irradiations (manuscripts 1, 2 and 3).
Rationale for submitting in alternative format
This thesis follows the alternative literature review and journal paper format. I felt this appropriate because this research project covers novel areas of the impact of
UVR on the skin following a simulated summer’s sunlight exposure in the UK.
Indeed this is the first concurrent examination of risk and benefit of a simulated summer sunlight intervention, and moreover is performed for people of different skin types. Thus dissemination of outcomes through peer-reviewed publication could impact on national public health information and recommendations, which are under revision. Thus I have been keen to quickly publicise my results in this fast moving area, and wish this research to assist national policy.
The thesis is constructed with an introductory literature review section that, together with the introductions within the manuscripts, details the rationale behind the investigations, followed by detailed methodology. I have four manuscripts, all of which are accepted (papers 1 and 4) or finalised for submission
(papers 2 and 3) for publication, covering the range of data explored during this
69 study. As my data are detailed within these publications, these four publications constitute the results chapters. The first manuscript examines the benefits of a simulated summer’s sunlight exposures, specifically 25(OH)D production, and the risks, in terms of cutaneous/urinary DNA damage in people of light and dark skin.
The second manuscript focuses on the photoprotective benefits against acute sunburn provided by a simulated summer’s sunlight exposure. The third paper encompasses the response of serum endocannabinoids and their related fatty acid
N-acyl ethanolamines to the simulated summer’s sunlight, exploring for any potential role they may have in the ‘feel good factor’ or addiction that some experience from tanning. The fourth manuscript subsequently covers my assessment of tanning addiction in a large cohort of patients with the UVR- responsive skin condition psoriasis. Ultimately, the thesis is summarised and further discussed in an overall concluding section.
70 CHAPTER 2: METHODOLOGY
2.1 Intervention study
2.1.1 Study design and approval
A pilot intervention study was designed to examine the impact of a simulated summer’s sunlight exposures in people of phototypes II and V, in terms of indicators of health risk and benefit, when following an approximation of national guidance on summer sunlight exposure (HPA, 2007). Ethical approval was obtained from North Manchester Research Ethics Committee (reference 09/
H1014/73) and regulatory approval from the Research & Development department of the study site (Salford Royal Hospital, Greater Manchester, UK).
2.1.2 Subjects
Eighteen healthy volunteers aged 18 to 60 years from Greater Manchester, UK, were recruited: Ten white Caucasians of skin phototype II and eight South Asians of phototype V. Recruitment methods involved advertisements in the local community, Salford Royal Hospital and The University of Manchester. Subjects were excluded from participation if pregnant or breastfeeding, taking vitamin D supplements or photoactive medication, if they had a personal history of skin cancer, photosensitivity, or if they had used a sunbed/sunbathed in the three months prior to commencement of the study, or during the study period. All received and read the study information leaflets at least one week prior to their decision to participate. They subsequently attended for an initial assessment visit where the study was discussed and they were given the opportunity to ask questions before providing their consent to participate. Written informed consent was obtained from the participants following good clinical practice guidelines, and the study adhered to the Declaration of Helsinki Principles.
71 2.1.3 Protocol overview
The flow chart below illustrates the assessments and interventions performed throughout the main study (Fig 2.3). Each participant’s age, sex, phototype, medical history, medications, height and weight at baseline, and study progress were recorded in individual Case Report Forms (Appendix 1). Body mass index
(BMI) in kg/m2 was thus calculated as weight (kg) divided by the square of their height (m). Volunteers were given short, suberythemal exposures to UVR three times weekly for six weeks, whilst wearing informal summer clothing
(standardised T-shirts and knee-length shorts) thus exposing ~35% skin area.
Serum biochemistry was checked at baseline, serum 25(OH)D assessed at baseline then weekly, and circulating parathyroid hormone (PTH) was measured at study commencement and end. Serum endocannabinoids were collected Monday,
Wednesday and Friday during the first week and then weekly during the study period (prior to irradiation at the same time of day ± 1 hr). Daily dietary logs recorded volunteers’ dietary vitamin D-containing foods during the first and last weeks of the study. Measurements of skin colour were made at baseline and subsequently at the end of each week of exposures. Urine samples were collected daily during week 1 and at the end of each subsequent study-week. Skin samples were taken at study-end from both UVR-exposed and unexposed buttock skin as detailed below (Section 2.1.9).
72
Figure 2.3: Flow chart to demonstrate an individual’s progression throughout the study period.
73 2.1.4 Baseline MED assessment
2.1.4.1 Waldmann UV6 lamp
Each participant’s MED was assessed at study-outset: A geometric series of 10 doses (7-80 mJ/cm2 for phototype II; 26.6–271 mJ/cm2 for phototype V) of erythemally weighted UVR was applied over two horizontal rows to buttock skin using a Waldmann UV 236B unit with Waldmann CF-L 36W/UV6 lamps (peak emission: 313 nm; range: 290–400 nm; Waldmann GmbH, Villingen Schwenningen,
Germany). The peak emission of the continuum spectrum was about 320 nm
(range 280-360nm).
2.1.4.2 UVR output measurements
The irradiance of the Waldmann lamp was routinely confirmed to be 1.25 to 1.45 mW/cm2 using a Waldmann UVR-meter, type 585 200 000 (Herbert Waldmann
GmbH, Villengen-Schwenningen, Germany). These irradiance readings were also regularly validated with an International Light IL730A UVR-radiometer
(International Light Technologies Inc, Peabody, MA, USA). In turn, this radiometer underwent biannual calibration for UV6-spectrum UVR sources with a Kratos
Schoeffel Instruments GM200 double-grating UVR spectroradiometer by Dr D
Allan (Principal Medical Physicist, The Christie NHS Foundation Trust and Salford
Royal NHS Foundation Trust, Manchester), working within a BSI-registered
ISO9001:2008 quality system.
2.1.4.3 MED testing procedure
As a photoprotected site, buttock skin was used for MED assessment. An MED filter plate, an UVR-opaque plate containing a series of 10 apertures, one aperture completely open and the others fitted with a series of metal-foil neutral density
74 filters that control the UVR irradiance reaching the skin, was secured with tape and the surrounding skin covered with UVR-protective drapes. Both researcher and subject wore UVR-protective eyewear during the testing-procedure.
After 5 mins warm-up, the UV6 lamp was secured 10 cm from the filter plate, and the UVB doses applied to the skin. 15 mins exposure time was given, with an erythemally-weighted irradiance of 0.14 mW/cm2 at the open aperture: UVR dosage (mJ/cm2) at each skin testing site was consequently the product of the UVR irradiance (mW/cm2) and total time (secs) for which it was applied. The test area was outlined on the skin with a surgical skin-marker pen and the subjects asked to keep that area dry for the following 24 hrs. MED readings were made 24 hrs later by visual assessment of the minimal (i.e. just perceptible) erythemal response.
2.1.5 Simulated summer sunlight exposures
2.1.5.1 Whole body irradiation cabinet
A Philips HB588 Sunstudio irradiation cabinet (Eindhoven, The Netherlands) was used to deliver whole body UVR exposures. This cabinet was re-fitted with a combination of 11 Arimed B (Cosmedico GmbH, Stuttgart, Germany) and 13 Cleo
Natural (Philips, Eindhoven, The Netherlands) fluorescent tubes in an alternating pattern, to provide an UVR emission spectrum as close as possible to summer sunlight (95% UVA: 320–400 nm, 5% UVB: 290–320 nm).
2.1.5.2 UVR characterisation
Emission from the Sunstudio radiation cabinet was characterised using a Bentham
DTM300 spectroradiometer (Bentham, Reading, UK) and monitored using an
Ocean Optics S2000 spectroradiometer (Ocean Optics, Dunedin, FL, USA) by Dr R.
75 Kift (School of Earth, Atmospheric and Environmental Sciences, University of
Manchester).
The course of simulated summer sunlight was given three times a week in January and February when ambient UVB is negligible at UK latitudes and people have their lowest vitamin D status (Webb and Engelsen, 2006). An exposure of 1.3 standard erythemal dose (SED; Diffey et al., 1997) was given to each subject at every visit. The time required to deliver this dose was found to be 6.5 mins after accurate measurement of cabinet UVR irradiance (Taylor et al., 2002); a constant
UVR dose was maintained throughout the study by adjusting delivery time for any change in irradiance.
In terms of the pre-vitamin D irradiance dose for one exposure in the cabinet, this was equivalent to 13 to 17 mins of sunlight exposure on a clear June midday in
Manchester, UK (53.5°N; Fig 2.5). This takes into account (i) unlike in the cabinet, ventral and dorsal surfaces are not irradiated simultaneously in sunlight when lying horizontally and (ii) in daily life postures range from the horizontal to the vertical randomly orientated to the sun (Webb et al., 2011).
76
Figure 2.5: Pre-vitamin D-weighted irradiance emission from the irradiation cabinet versus that of a clear Manchester, UK (53.5°N) summer day at noon.
Taken from Rhodes et al, 2010.
2.1.5.3 Cutaneous exposure in the irradiation cabinet
To enable comparisons between UVR-exposed and UVR-protected sites, a 10 cm2 aperture was made in the shorts material overlying the right buttock to ensure complete UVR exposure. In contrast, the entire contralateral buttock was completely covered with an UVR-opaque material. A skin marker pen was used to outline the margin of the aperture in the shorts material on the buttock skin and this was reinforced on each visit. Volunteers were requested not to remove these marks. The aperture in the shorts material was then fixed to the outline on the skin with Micropore® tape (3M, St Paul MN, USA) once the subjects were lying prone in the cabinet prior to irradiation, to ensure that the UVR-exposed area was in exactly the same position on each occasion.
2.1.5.4 Acute 2X MED exposures
Following the simulated summer sunlight exposures, buttock skin from the
77 phototype II participants who, by definition, are more at-risk of sunburn, was exposed to a pro-inflammatory challenge. This was individually dosed for each participant as twice their baseline MED (2X MED) of UVB. 2X MED was applied using a Waldmann UV 236B unit (Waldmann GmbH, Villingen Schwenningen,
Germany) to two 1 cm2 areas of buttock skin: one area had received the simulated summer’s UVR exposures whilst the other was a control site that had been photoprotected. These sites were used for tanning and erythema measurements and skin sampling after 24 hrs, as detailed below (sections 2.1.8, 2.1.9).
2.1.6 Diaries of dietary vitamin D intake
During the first and last weeks of the simulated summer’s sunlight exposures, volunteers completed daily dietary logs of vitamin D-fortified foods, and six key food categories: cheese, butter/margarine/oily spreads, milk/milk-containing products, red meat, oily fish, eggs/egg dishes, as well as vitamin D supplements (an exclusion criterion for the study). Records were made of the amount and type of foodstuff consumed (Appendix 2). The vitamin D content of these foodstuffs was subsequently obtained from food package labelling and the 6th edition of McCance and Widdowson’s The Composition of Food (2002).
2.1.7 Blood sampling
Venous blood samples were taken at baseline for measurement of vitamin D status
(25(OH)D), endocannabinoids and PTH. Serum biochemistry (including calcium, kidney and liver function tests) was also confirmed to be normal at study-onset.
Endocannabinoids were checked Monday, Wednesday and Friday of week 1, and then weekly. Sampling was also weekly for 25(OH)D, and at the end of the study
25(OH)D, endocannabinoids and PTH were rechecked. Samples were spun in a
78 centrifuge at 2400 rpm for 15 mins and once the serum had separated it was aspirated using a pipette and stored at -20 °C until analysis.
2.1.7.1 Vitamin D status
Serum 25(OH)D was measured by high pressure liquid chromatography (HPLC) using UV detection in the Vitamin D Research Laboratory, Manchester Royal
Infirmary, under the supervision of Dr J Berry. This laboratory is accredited to ISO
9001:2000 and ISO 13485:2003 standards, and certified proficient by the national
Vitamin D quality assurance scheme (Berry et al., 2007): In brief, thawed samples were vortexed before 1 ml of serum was used to extract 25(OH)D. This volume was subsequently made up to 3 ml using 0.9% saline and 20 µl 25-hydroxy [26,27-
3 methyl- H]cholecalciferol (25(OH)D3) recovery tracer (TRK 655, GE Healthcare UK
Ltd, Amersham, Buckinghamshire). This was allowed to equilibrate for 30 mins before 3 ml of acetonitrile was added and the mixture vortexed at 2800 rpm at 4 °C for 15 mins (Sigma Labs 4K15). 1 ml of distilled water was added before application to C18 Sep Pak Cartridges (Waters Ltd, Elstree, Hertfordshire, UK), preconditioned with 2 ml of methanol, 5ml of water. These cartridges were washed with 3 ml of 65% methanol/water before 3 ml of acetonitrile added. After drying under a stream of nitrogen, samples were transferred to limited volume inserts using HPLC column solvent. Once blown down, 250 µl of column solvent was added to each vial and these then capped. Separation was using straight phase
HPLC overnight with 5 µm, 4.6 x 255 mm Hewlett-Packard Zorbax-Sil column
(Hicrom, Reading, Berkshire, UK), eluted with hexane:propan2ol (98:2), quantified by UVR absorbance at 265 nm, corrected for recovery. To ensure good separation, run-time was set around 16 mins. The sensitivity was 2 ng/ml and interassay variation 6%.
79 2.1.7.2 Serum PTH and biochemistry analysis
Serum PTH level was measured using the OCTEIA immunoenzymometric assay,
(Immunodiagnostic Systems, Boldon, Tyne and Wear, UK), with sensitivity 0.06 pmol/l and intra- and inter-assay coefficients of variation 4% and 6%, respectively.
Serum biochemistry was measured using the Hitachi 917 autoanalyser (Hitachi,
Tokyo, Japan).
2.1.7.3 Serum endocannabinoid and N-acyl ethanolamine (NAE) quantification
Blood samples were taken Monday, Wednesday and Friday of the first week of irradiation and each subsequent Monday (i.e. 3 days after last irradiation) until course-end for endocannabinoid and NAE quantification. Collections were prior to
UVR exposure at the same time of day on each occasion to within 1 hr. Serum was stored at −20 °C until study completion whereupon they were defrosted in ice and
3 ml of ice-cold 2:1 (v/v) chloroform/methanol was added. Internal standards were 20 ng of anandamide-d8 and 40 ng of 2-arachidonoyl glycerol-d8 (Cayman chemicals). The samples were mixed and incubated on ice for 30 mins. To each sample, 500 µl of water was added before centrifugation at 5000 rpm, 4 °C, for 5 mins. The organic phase was subsequently dried under a steam of nitrogen and the lipids extract reconstituted in 100 µl of HPLC-grade ethanol. The collection was centrifuged for 10 secs and all extracts then transferred to an insert in a labelled amber vial with septa and lid. Parafilm was placed around the lid to stop evaporation of solvent and they were stored on ice until finished. The syringe was washed with three ethanol washes and the extracts maintained at -20 °C awaiting liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.
80 An isocratic system composed of two solvents (A and B) mixed at constant ratio of
30:70 (v/v) was used to separate analytes. Solvent A was acetonitrile: water: acetic acid, 2:98:0.5 (v/v/v) and solvent B acetonitrile: water: acetic acid, 98: 2: 0.5
(v/v/v), with run-time of 69 mins. A C18 column (Luna 5 µl, 150 x 2.0 mm;
Phenomenex) was used for separation. Injection volume for standards and the biological extracts was 3 µl. The instrument is operated in the positive ionisation mode, and the MS/MS settings for all compounds were: capillary voltage 4500 V, source temperature 100 °C, desolvation temperature 400 °C, cone voltage 35 Dwell time 0.2 secs. The transitions were split at 10 mins so that those that are eluted before 10 mins and those that are eluted after 10 mins have more focused analyses. LC/ESI-MS/MS analysis was performed on a Waters Alliance 2695 HPLC pump with a Waters 2690 autosampler coupled to an electrospray ionisation (ESI) triple quadruple Quattro Ultimo mass spectrometer. The Mass LynxTM V 4.0 was used as operating software to control the instrument and data acquisition.
Standards were used to generate calibration lines to cover a range of 1-200 pg/µl, which showed a linear response and samples were analysed within this range before normalisation against protein content.
2.1.8 Skin colour measurements
All participants had assessments of their buttock skin colour taken non-invasively at baseline (i.e. prior to UVR exposure) using a Minolta CM-2500d hand-held spectrophotometer (Konica Minolta, Tokyo, Japan). Readings were taken when the volunteers had been lying prone for ~5 mins prior to irradiation, to reduce postural variations in circulation as confounding factors. UVR-exposed and nearby
UVR-protected buttock skin subsequently underwent weekly spectrophotometric measurements in this manner. Readings were also taken 24 hrs after the 2X MED
81 UVB-challenge from exposed and UVR-protected skin.
2.1.8.1 L*a*b* recordings
Data were recorded in standard, three-dimensional Commission International de l’Eclairage L*a*b* format (Robertson, 1977): L* represents white-black differentiation, whereby an L* of 100 is pure white and conversely L* of 0 is pure black, a* values reflect the balance between green (negative) and increasing redness (positive) whilst b* represents the differentiation between blue (negative) and yellow (positive). The spectrophotometer was ‘zero calibrated’ at study commencement by Dr D Allan, in addition to being calibrated using its ‘white calibration’ system prior to each participant’s measurements, to guarantee accuracy of readings. Readings were made in triplicate at each site and the mean calculated.
2.1.8.2 Individual Typology Angle
Individual Typology Angle (ITA) values were calculated from spectrophotometer readings as the vector direction in the L*b* plane, as arctangent [(L*-
50)/b*]x(180/π); readings of > 55 ° are classified as ‘very light’, 10-28 ° as ‘tanned’ and > -30 ° as ‘dark’ (Chardon et al., 1991; Del Bino et al., 2006; Del Bino and
Bernerd, 2013).
2.1.8.3 Erythema meter
A reflectance instrument (Erythema meter, Diastron, Andover, UK) was used for non-invasive measurements of buttock skin erythema in phototype II individuals.
After calibration, erythema index (EI) readings were taken in triplicate from buttock skin that had been exposed to the simulated summer’s UVR exposures, and
82 skin 24 hrs after 2X MED UVB, and to obtain background readings for both scenarios, from adjacent non-irradiated skin. Mean values were calculated.
2.1.9 Cutaneous sampling
Following the six weeks’ simulated summer sunlight exposures, all participants had four 4 mm punch biopsies taken under local anaesthetic (1% lignocaine) from buttock skin as follows: Photoprotected skin (untreated control), skin immediately post-exposure to a single 1.3 SED, skin immediately following the course of 18X 1.3
SED (i.e. the simulated summer), and skin 24 hrs following the complete course of
18X 1.3 SED irradiations. For the ten phototype II participants, to whom there had been an additional 2X MED UVB exposure, two additional biopsies were taken from 2X MED exposed buttock skin, one site having also received, the other having been photoprotected from the simulated summer sunlight. Biopsies were formalin- fixed and paraffin-embedded prior to histological analysis.
2.1.9.1 Immunostaining for cutaneous anti-thymine dimer
4 μm biopsy sections were prepared for immunohistochemical analysis including being deparaffinised with xylene, rehydrated through graded alcohol and incubated with 0.1% trypsin water at 37 °C for 25 mins. After rinsing in water for
10 mins, DNA was denatured with 70mM NaOH in 70% ethanol for 4 mins before repeated rinsing under running water for 5 mins, and 3% hydrogen peroxide diluted in methanol added for 5 mins at room temperature, to inhibit endogenous peroxidase. Slides were subsequently rinsed under running water for a further 5 mins. Blocking buffer (Vector Laboratories, Cambridgeshire, UK) was added for 30 mins followed by monoclonal antibody at 1:2,000 (TDM-2 [CosmoBio, Tokyo,
Japan] which binds CPD; Mori et al., 1991) at room temperature for 1 hr. For
83 control purposes, primary antibody was not added to one slide in each staining- cycle. Slides were rinsed three times with phosphate buffered saline (PBS)/Tween-
20 before being incubated with biotinylated secondary antibody. Two further
PBS/Tween washes were performed before the addition of ABC solution (Vector
Laboratories) for 30 mins. They were developed with Vector SG solution, rinsed and counterstained with Nuclear Fast Red (Vector Laboratories), before being dehydrated and permanently mounted with Clarion (Sigma-Aldrich Inc; St. Louis,
USA). Three sections were mounted per slide.
Quantification of staining and epidermal thickness/area measurements
Images were scanned using Pannoramic 250 Flash II, 3DHisTech Ltd. (Budapest,
Hungary) and analysed for epidermal thickness and area using Pannoramic Viewer
1.15.2 (Budapest, Hungary) and Image J 1.48 (National Institutes of Health,
Bethesda, MD). Counts were made of the number of positively-staining nuclei per high power field (hpf; original magnification x40), with three hpf per section (9 hpf per slide), whilst blinded to slide identity, and the means calculated.
2.1.9.2 Immunostaining for cutaneous neutrophil elastase
3 μm skin sections were prepared for immunohistochemical analysis including being deparaffinised with xylene, rehydrated through graded alcohol and solubilised with 0.5% Triton X-100 for 10 mins. After washing in tris-buffered saline (TBS), DNA was denatured with 70 mM NaOH in 70% ethanol for 4 mins before repeated washing. 0.3% hydrogen peroxide diluted in PBS was added at room temperature for 5 mins to inhibit endogenous peroxidase before repeated washing in TBS for 2 mins. Primary antibody was added (Neutrophil Elastase
NP57, Dako, Glostrup, Denmark; dilution 1:100) in primary antibody diluent
84 (Diagnostic Biosystems, California, USA) for 30 mins at room temperature. For control purposes, the primary antibody was not added to one slide for each cycle of staining. After washing twice in TBS, ImmPress reagent (Vector Laboratories,
California, USA), was added for a further 30 mins before repeated rinsing in TBS.
Slides were developed with NovaRed (Vector Laboratories, California, USA) for 5 mins, rinsed and counterstained with Mayer's haematoxylin for 5 mins. After washing they were dehydrated and permanently mounted using Distyrene,
Plasticiser and Xylene solution (DPX; Sigma-Aldrich Inc, St. Louis, USA).
Quantification of staining and epidermal thickness measurements
Images were captured and scanned using Pannoramic 250 Flash II, 3DHisTech Ltd.
(Budapest, Hungary) scanner. Counts were made of the number of positively staining nuclei per hpf (original magnification X20), in triplicate per section, with three hpf taken per section (nine hpf per slide) whilst blinded to slide identity.
Thresholds of detection were chosen after examination of initial samples, such that all dense staining cells were identified without identification of non-staining cells.
Cellular epidermal (from stratum basale to distal stratum granulosum) and stratum corneum thicknesses were measured using the Pannoramic Viewer, again nine per slide and the means calculated. Epidermal and dermal areas per hpf were measured using Image J 1.48 (National Institutes of Health, Bethesda, MD) whilst blinded to sample identity for counting and measurements.
2.1.10 Urinalysis
First-void urine samples were collected daily (Monday-Friday) during week one and then at the end of each week of UVR exposure, and stored at -20 °C until processing.
85 2.10.1 Urinary 8-oxodG quantification
Samples were analysed for 8-oxodG using ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS; Lam et al., 2012):
Samples were thawed on the day of extraction and then spun at 16,000 g for 10 mins at 4 °C. 0.5 ml urine was then spiked with 10 pmol of 8-[15N5]oxodG internal standard and diluted 1:1 with deionised water. Samples were then microcentrifuged at 16,000 g for 1 min at room temperature. Env + Isolute (1 ml,
50 mg) cartridges (Biotage, Uppsala, Sweden) were preconditioned with 1 ml methanol and 1 ml H2O. Samples were introduced onto the cartridges and drawn through at a flow rate of approximately 1 ml/min. The cartridges were washed twice with 300 ml H2O and 8-oxodG was eluted in 2 µl x 300 µl of 20% (v/v) acetonitrile in methanol. The eluents were dried under nitrogen before reconstitution in 50 ml mobile phase and centrifuged at 1000 g for 1 min before transferring to Eppendorf tubes. Samples were then centrifuged at 16,000 g for 10 mins before the supernatants were transferred to HPLC vials.
To determine the extraction efficiency, urine separately collected from one healthy individual was centrifuged at 16,100 g at 4 °C for 10 mins. The recovery of 8-oxodG was assessed by the addition of five different amounts of 8-oxodG (0, 5, 10, 15, and
20 pmol) to 500 ml of the urine supernatant, before SPE. In parallel, 5 500-ml aliquots of the same urine supernatant underwent SPE; however, these were spiked with the same amount of 8-oxodG after SPE. All 10 samples were spiked with 10 pmol IS after SPE. The peak area ratio (PAR) of unlabeled 8-oxodG to IS in all 10 samples was calculated. The percentage recovery of 8-oxodG was then calculated as (PARPre/PARPost) x 100%, where PARPre represents samples with the addition of unlabeled 8-oxodG before SPE, and PARPost represents samples with the
86 addition of unlabeled 8-oxodG after SPE.
UHPLC–MS/MS analysis
The UHPLC–ESI–MS/MS system comprised an Acquity UPLC in line with a Quattro
Premier tandem mass spectrometer (Waters, Elstree, UK). The column was an
Acquity UPLC BEH C18 (2.1 mm x 100 mm) maintained at 40 °C. An isocratic mobile phase of 5% methanol, 0.1% formic acid was used with a flow rate of 0.25 ml/min.
Samples were maintained at 4 °C throughout. Analytes were quantified using tandem electrospray ionization mass spectrometry in positive-ion mode. Source parameters were capillary voltage 3.2 kV, cone voltage 20 V, source temperature
110 °C, desolvation temperature 350 °C, cone gas flow 50 L/h, desolvation gas flow
800 L/h. MS/MS conditions for each precursor [M+H]+ ion comprised entry, collision, and exit energies of 2, 15, and 0 eV, respectively. Product ions were monitored in multiple-reaction monitoring mode. Mass transitions were 8-oxodG
15 (m/z 284 >167.9) and 8-[ N5]oxodG (m/z 289 > 173). A secondary or qualifier transition (m/z 284 > 116.8) was used to confirm the identity of 8-oxodG in samples. Injection volume was 5 ml. Seven-point calibration curves spiked with IS were generated. Spectrograms were integrated using MassLynx software version
4.1. QuanLynx software calculated the concentration of 8-oxodG using calibration curves of concentration against relative response calculated as follows: relative response (y) = (peak area)/(IS area/[IS]). The UHPLC–MS/MS method was validated according to FDA guidelines (http://www.fda.gov). Results were normalised to urinary creatinine levels.
87 2.1.10.2 Urinary thymine dimer quantification
Two approaches to the mass spectrometric analysis of urinary thymine dimers were adopted: the first quantified the dimer as a dinucleotide monophosphate (the dimerised form of thymidylyl-3’-5’thymidine, T<>pT; Le Curieux et al., [2001]), the second utilised formic acid hydrolysis of urine to render all oligomeric forms down to the nucleobase form of the dimer (thymine-thymine dimer, T<>T).
Materials
Acetonitrile (HPLC gradient grade), water (HPLC gradient grade) and mass spectrometry grade formic acid were from Fisher Scientific (Loughborough, UK).
Ammonium formate (>99.0%) and acetophenone were from Sigma-Aldrich (Poole,
15 13 UK). Isotopically labelled ([ N2, C10]; CK Gas Products, Ibstock, UK) and unlabelled thymidylyl-(3’-5’)-thymidine (dTpdT) were from Carbosynth Limited
(Berkshire, UK). Oasis WAX 50mg (Waters Corporation, Manchester, UK), and
ISOLUTE ENV+ 50 mg (Biotage, Uppsala, Sweden) solid-phase extraction (SPE) cartridges were obtained from their manufacturers.
Preparation of unlabelled and stable isotope labelled thymine dimers
Due to possible loss of analyte during sample workup, isotopically labelled internal standards were included for each target compound. The synthesis of both labelled
2 (α,α,α,6- H4) and unlabelled T<>T was based on the method of Wang (1961; described by Podmore et al., 1996). A frozen aqueous solution of thymine (2 mg/ml) was stored at -80 °C for 12 hrs and then UVC irradiated (254 nm UVS-18 EI series UV lamp) on dry ice for 4 hrs. The solution was thawed and its precipitate isolated by centrifugation (1200 x g) for 20 mins, and dried at 40 °C. Such
88 methodology has been shown previously to produce a standard free from impurity by thymine (Podmore et al., 1996).
13 Labelled thymidylyl-3’-5’thymidine (TpT) was custom-synthesised from [ C10,
15 N2]-dThy by Carbosynth Ltd (Compton, UK). One molecule of dThy per TpT was
13 15 labelled. Labelled ([ C10, N2]-T<>pT) and unlabeled T<>pT were generated by
UVR exposure in the presence of a photosensitizer (Liu and Yang, 1978). Oxygen was purged from a solution of TpT (5 mg/22.7 ml) in HPLC gradient grade water, and 13.2 mM acetophenone, by bubbling nitrogen through this solution. This was then UVB irradiated (Philips TL-01 lamp) for 3.5 hrs under nitrogen, at 10 °C. The resulting photoproducts (cis-syn and trans-syn T<>pT) were isolated in the
Department of Chemistry (University of Leicester) using preparative, reversed phase HPLC with a Dionex Ultimate 3000 system and Phenomenex Gemini NX 5µm
C18 110Å AXIA column (250 x 21.20mm). Separation was achieved by a 35 mins isocratic run with 99.9% water and 0.1% trifluoroacetic acid, with a flow rate of
10.6 ml/min. Only two peaks were seen in the chromatogram, at 13.9 and 18.3 mins. Both peaks were collected and subsequently freeze-dried, with 2.7 mg and
<1 mg yields respectively.
Urine pre-treatment and solid phase extraction
All urine samples were kept frozen at -20 °C until processing, when they were thawed at room temperature and then centrifuged. Any pellet was discarded and the supernatant used subsequently. For T<>T, urine supernatants (10 µl) were spiked with internal standard prior to formic acid hydrolysis, and solid phase extraction on ISOLUTE ENV+ SPE columns. Following SPE, the resulting eluate was evaporated to dryness under nitrogen and then reconstituted in 50 µl mobile
89 phase. For T<>pT, 150 µl of urine supernatant was spiked with internal standard and diluted 1:1 with 2% formic acid. Oasis WAX SPE columns were conditioned with 1 ml methanol, then equilibrated with 1 ml of water, before adding the 300µl sample (vacuum sufficient to achieve 1 ml/min). The columns were washed with 1 ml 2% formic acid, eluted with 1 ml of methanol and then 300 µl of 5% (w/v) ammonium hydroxide in methanol (x2). The resulting eluate was also evaporated to dryness under nitrogen before reconstitution in 50 µl mobile phase.
UHPC-MS/MS analyses
Experiments were performed on a Quattro Premier spectrometer (Waters
Corporation, Manchester, UK), operated in negative-ion mode with ESI capillary voltage of 3 kV and cone voltage 60 V. The nitrogen desolvation and cone gas flow rates were 1000 and 50Lh−1 respectively, with a source temperature of 120 °C and desolvation gas temperature 300 °C. The mass spectrometer was operated in multiple-reaction monitoring mode. For cis,syn T<>T three transitions were
2 identified: 253>210, 253>210 and 253>139 ([ H8]-T<>T: 261>159). For cis,syn
T<>pT, three mass transitions were identified 545>447 545>253, and 545>195
13 15 (([U- C10, U- N2]-T<>pT: 557>431, 557>333, 557>298, and 557>125). Conditions for each were as follows: entry and exit voltage of 30eV with a collision energy of
31eV and an Argon collision cell pressure of 5.9e-003. Masslynx version 4.1 (Waters
Corporation, Manchester, UK) was used to control the IM-MS instrument and for data acquisition and processing. UHPLC was performed on a Waters Acquity system (Waters Corporation, Manchester, UK) with an Acquity UHPLC HSS T3®
(Waters Corporation, Manchester, UK) column (2.1 x 50 mm). The UHPLC system was coupled to a post-column T-piece fitting (Waters Corporation, Manchester,
UK) and then a further Dionex HPLC system (Thermo Fisher UK Ltd, Hemel
90 Hempstead, UK) with the outlet to the ESI ion source of the mass spectrometer.
Cleaned-up samples (5 µl injected) were eluted with the following gradient: 99.9%
A (0-2 mins), increased to 30% B (2-4 mins) and then to 99.9% A (4-6 mins), where A=2.5 mM ammonium formate and B=acetonitrile. The mobile phase flow rate was 0.1 ml/min. The HPLC flow rate was set to 0.1 ml/min employing an isocratic flow of acetonitrile to aid ionisation. The limit of detection was 396 fmol.
2.2 Survey of tanning addiction in a cross-section of psoriasis patients
2.2.1 Study design and approval
This was a cross-sectional survey. Participants were identified by their membership in the National Psoriasis Foundation (NPF): An invitation was incorporated into February 2014’s e-newsletter. The NPF reports that approximately 14,000 of their members open these emails. The invitation included a link to the online survey, administrated by Survey Monkey®. A reminder was sent one and two months later. To prevent analyses of duplicate responses from the same individual, Internet Protocol (I.P.) addresses were recorded. Other identifying features were not collected. This study was qualified as exempt by the
Institutional Review Board of the University of Texas Southwestern Medical
Center, Dallas, U.S.A, where the study was performed.
2.2.2 Survey participants
To be eligible for the survey, respondents were asked to confirm that they had received a diagnosis of psoriasis from a physician. Introductory questions included basic demographic assessment including age, gender and health insurance status.
Survey questions are included in manuscript 4.
91 2.2.3 Psoriasis assessment
With regards to their psoriasis, questions encompassed the age at diagnosis, indicators of severity (including previous hospitalisation or erythroderma) and both previous and currents treatments, specifically topical, systemic, biologic and/or photo- therapies. Current psychological impact of psoriasis was assessed according to its interference with daily activities on an ascending scale of 0 to 10 with increasing interference. Participants were asked to score the overall state of their psoriasis that day, from specific choices ranging from clear to intensely inflamed with pus spots, in addition to the extent/severity of their psoriasis on ten named body areas (scalp, face, arms, hands, chest/stomach, back, groin, buttocks/thighs, knees/lower legs and feet). These responses were inputted to the previously validated self-assessed Simplified Psoriasis Index (SPI; Appendix 3;
Chularojanamontri et al., 2013), to calculate current psoriasis severity from 0 to 50 whereby an overall score of <10 was considered mild, 10 to 20 as moderate and
>20 classified as severe psoriasis (Chularojanamontri et al., 2014).
2.2.4 Assessment of tanning
Participants were asked if they had ever used UVR outside of the medical setting, such as sunbed usage. If a negative response was provided, the survey ceased. If affirmative, the survey continued, querying firstly whether tanning had commenced in order to treat their psoriasis and, if so, what factors had contributed to this self-treatment (e.g. difficulty accessing, cost of, or perceived inefficacy of physician-provided therapy or conversely previous successful medical phototherapy). Participants could also provide any other applicable reasons in a free-text area.
92 For all tanners, regardless of whether they had commenced to treat their psoriasis or for alternate reasons, questions covered their age at commencement, total number of times tanning and last tanning session. Those who endorsed using a sunbed within the past 12 months were asked how often they typically use a sunbed (more than once a week, once a week, 1 to 4 times a month, less than once a month), and those who have tanned within the past five years were asked about any seasonal variations (i.e. spring/summer, fall/winter, year-round or other). In addition, five reasons were presented in a randomly-generated order, namely to improve mood, to look good, to prevent sunburn, to treat their psoriasis, or prevent their psoriasis from recurring, and participants were asked to rank the perceived importance of these reasons in relation to their tanning. An additional free-text area was provided to enable respondents to describe any other reasons for tanning.
2.2.5 Assessments of addictive-like tanning behaviour
Respondents who acknowledged ongoing tanning were asked questions adapted from the Diagnostic Statistical Manual (DSM-IV) for Substance Use Disorder (APA
2000). To limit the time required to complete the questionnaire, and thus decrease the dropout rate, only seven of the eleven DSM-V criteria were asked:
1. Are there times your indoor tanning is more important to you than most other
things in your life such as friends, family, school or job?
2. Has your indoor tanning caused you any problems with other people such as
family members, friends, romantic partners or people at work or school?
3. Did you find that when you started using indoor tanning you ended up using it
much more than you were planning to?
4. Have you tried to cut down or stop indoor tanning but been unsuccessful?
93 5. Have you found that you needed a lot more indoor tanning sessions in order to
get the feeling you wanted (for example, feel good, reduce stress, feel relaxed,
etc) than when you first started indoor tanning?
6. When you haven’t tanned for a while, do you have a strong desire to tan?
7. Have you been diagnosed with a skin cancer and continued to use a tanning bed
after this diagnosis?
Positive responses were summed to calculate severity of an addictive disorder as described in DSM V, i.e. 2 to 3 criteria as mild addictive-like behaviour, 4 to 5 as moderate, and >5 as severe. To compare personal perceptions, they were also asked if they considered their tanning to be a problem.
2.3 Statistical analyses
Data analyses, specifically paired and unpaired t-tests, Chi-square, linear regressions, and repeated measures ANOVAs, were performed using SPSS statistical software (version 21.0.0; IBM SPSS Statistics, Portsmouth, UK) and
GraphPad Prism (version 6; GraphPad Software, La Jolla, CA, USA). Ratio measures were logarithmically transformed to make them normally distributed. Results were considered statistically significant if p<0.05.
94 CHAPTER 3: Manuscript 1
Published as Br J Dermatol. 2016 Jul 14. doi: 10.1111/bjd.14863. [Epub ahead of print]
95 Concurrent beneficial (vitamin D production) and hazardous (cutaneous
DNA damage) impact of repeated low-level summer sunlight exposures
Short title: Benefit and hazard of summer sunlight exposures
SJ Felton,1 MS Cooke,2 R Kift,3 JL Berry,4 AR Webb,3 PMW Lam,5 FR de Gruijl,6 A
Vail,7 LE Rhodes1
1. Dermatology Research Centre, Institute of Inflammation and Repair, Faculty of
Medical and Human Sciences, University of Manchester, Manchester Academic
Health Science Centre, Salford Royal NHS Foundation Trust, Manchester, UK
2. Oxidative Stress Group, Department of Environmental and Occupational Health,
Florida International University, Miami, USA
3. School of Earth Atmospheric and Environmental Sciences, University of
Manchester, Manchester, UK
4. Department of Clinical Biochemistry, Manchester Royal Infirmary, Central
Manchester NHS Foundation Trust, Manchester Academic Health Science Centre,
Oxford Road, Manchester, UK
5. Oxidative Stress Group, Department of Cancer Studies and Molecular Medicine, University of Leicester, Leicester, UK
6. Department of Dermatology, Leiden University Medical Centre, Leiden, The
Netherlands
7. Centre for Biostatistics, Institute of Population Health, University of Manchester,
Manchester Academic Health Science Centre, Salford Royal NHS Foundation Trust,
Manchester, U.K.
96 Correspondence to: Prof L E Rhodes, Photobiology Unit, Dermatology Centre,
University of Manchester, Salford Royal Hospital, Manchester, M6 8HD, UK
Tel 00 44 161 1150, Fax 00 44 161 1156,
Email: [email protected]
Word count: 2969
Tables: 1
Figures: 3
What’s already known about this topic?
Repeated low-level exposures to simulated UK sunlight can produce vitamin D sufficiency in light-skin people but concurrent impact on cutaneous DNA damage is unknown.
What does this study add?
Low-level simulated sunlight exposures in people of phototype II confer vitamin D sufficiency concurrent with DNA damage that shows partial clearance at 24h and no evidence of accumulated damage after 6-weeks’ exposures.
The same exposures produce minimal DNA damage but less vitamin D in brown- skin people (phototype V).
Findings are informative for sun-exposure guidance.
97 Abbreviations 25(OH)D 25-hydroxy vitamin D
8-oxodG 8-oxo-7,8-dihydro-2’-deoxyguanosine
CPD Cyclobutane pyrimidine dimers
ITA Individual Typology Angle
MED Minimal Erythemal Dose
SED Standard Erythemal Dose
SPE Solid-phase extraction
UVR Ultraviolet radiation
UHPLC-MS/MS Ultra-high-performance liquid chromatography-tandem mass
spectrometry
TpT Thymidylyl-3’-5’thymidine
T<>T Cyclobutane thymine dimer
T<>pT Cyclobutane thymidylyl-3’-5’thymidine dimer
98 ABSTRACT
Background: Concurrent impact of repeated low-level summer sunlight exposures on vitamin D production and cutaneous DNA damage, potentially leading to mutagenesis/skin cancer, is unknown.
Objectives: Experimental study (i) determining dual impact of repeated low-level sunlight exposures on vitamin D status and DNA damage/repair (via both skin and urinary biomarkers) in light-skin adults (ii) comparing outcomes following same exposures in brown-skin adults.
Methods: Ten white Caucasians (phototype II) and six South Asians (V), 23–59y, received 6-weeks’ simulated summer sunlight exposures (95% UVA/5% UVB,
1.3SED x3 weekly) wearing summer clothing exposing ~35% surface.
Assessments made were: circulating 25(OH)D; immunohistochemistry for cyclobutane pyrimidine dimer (CPD)-positive nuclei in skin: (i) unexposed, (ii) immediately-post single UVR, (iii) immediately-post and (iv) 24h-post x18 UVR- exposures; urinary biomarkers of direct and oxidative (8-oxo-dG) DNA damage.
Results: Serum 25(OH)D rose from mean 36.5 (SD13.0) to 54.3 (10.5) nmol/L
[14.6 (5.2) to 21.7 (4.2) ng/mL] in phototype II versus 17.2 (6.3) to 25.5 (9.5) nmol/L [6.9 (2.5) to 10.2 (3.8) ng/mL] in V (p<0.05). Phototype II skin showed
CPD-positive nuclei immediately post-course, mean 44% (range 27-84) cleared after 24h, contrasting with minimal DNA damage and full-clearance in V
(p<0.0001); findings did not differ from following single UVR-exposure. Urinary
CPD remained V (p=0.002), but unaffected by UVR. Conclusions: Low-dose summer sunlight exposures confer vitamin D sufficiency in light-skin people concurrent with low-level, non-accumulating DNA damage. 99 The same exposures produce minimal DNA damage but less vitamin D in brown- skin people. This informs tailoring of sun-exposure policies. 100 INTRODUCTION Solar UVR exposure has the established benefit to health of vitamin D synthesis whilst skin cancer is a major hazard. Studies using various protocols have examined impact of single- and repeated-dose ultraviolet radiation (UVR) on vitamin D status,1-5 but research examining accompanying UVR-induced DNA damage is scarce. Recently, the impact of high intensity UVR-exposures attained through a sunbathing holiday (Canary Islands, 28°N) on circulating 25- hydroxyvitamin D (25(OH)D) and cyclobutane pyrimidine dimer (CPD) excretion in urine, as a proxy for UVR-induced cutaneous DNA damage, was explored in white Caucasians.5 There was increase in both vitamin D status and urinary CPD, and conclusion that under high-level UVR-exposure conditions, vitamin D benefit is inevitably derived at DNA damage cost. However, this might differ with UVR- exposure pattern/dose, and between phototypes.6,7 Skin cancer is prevalent and causes substantial health burden in white Caucasian populations. The main exogenous risk factor, UVR, is a complete carcinogen, initiating DNA damage and suppressing skin immunity.8 UVB induces pyrimidine (6-4) pyrimidone photoproducts9 and CPD,10 the dominant mutagenic form of direct UVR-induced DNA damage11 with thymine-containing dimers most common.10 If not repaired, these photoproducts form “UVB signature” mutations present in skin cancers.12 Recently, UVA was also shown to induce thymine- containing dimers in human epidermis in vivo.10,13 UVR also induces oxidatively- generated damage to nucleic acids.14 UVR-induced DNA damage stimulates melanogenesis, although this provides only modest protection against further UVR-damage.15,16 Urinary excretion of UVR-induced DNA damage may act as a convenient proxy for cutaneous DNA damage;17 however, to date, skin and urinary damage have not been directly compared. 101 UVB triggers conversion of 7-dehydrocholesterol to pre-vitamin D, the body's principal vitamin D source, with usually only small amounts obtained from diet.18 Vitamin D undergoes hepatic hydroxylation to 25(OH)D, the major circulating form and current best indicator of vitamin D status, and subsequent renal hydroxylation to active 1,25-dihydroxyvitamin D. There is associative evidence of diverse health benefits of vitamin D,19-21 while its established benefit is musculoskeletal, including prevention of rickets and osteomalacia.22,23 Public health guidance recommends sun protection in individuals at high risk of skin cancer,24 while also considering vitamin D benefit. It is generally assumed regular brief sun exposures to skin produce adequate vitamin D.25 Guidance is geared for light-skin individuals, and supported by intervention study in 109 white Caucasians where simulated low-level sunlight exposures, while casually dressed, produced vitamin D sufficiency, defined as 25(OH)D ≥50nmol/L (20ng/mL).1 This study’s objectives are to examine impact on cutaneous DNA damage/repair (skin and urinary biomarker assessment) alongside 25(OH)D gain with regular low level UVR-exposures, in both white- and brown-skin people. We exposed 10 white Caucasians and six South Asians to a simulated summer’s brief exposures (95%UVA/5%UVB, x3 weekly for six weeks). Skin biopsies were examined for CPD-positive nuclei: induction by a single 1.3 SED exposure; accumulation over 6 weeks’ UVR-exposures; clearance 24h post-course. Urine was analysed for CPD and 8-oxodG DNA damage.26 Through performance under known exposure conditions, the data gained are informative for sun-exposure guidance. 102 METHODS Subjects Experimental study in healthy volunteers. People of phototype II (white skin; sunburns easily, tans minimally) and V (South Asian, brown skin), 18–60y, from Greater Manchester, UK were recruited by advertisement (January 2010). Exclusion criteria: history of skin cancer/photosensitivity, use of sunbed/sunbathing within 3 months, taking photoactive medication/vitamin D supplements, pregnancy/breastfeeding. North Manchester Research Ethics Committee provided ethical approval (reference 09/H1014/73). The study adhered to the Declaration of Helsinki; subjects gave written, informed consent. Minimal Erythemal Dose (MED) assessment Individuals’ MED were assessed pre-course, as the lowest UVR-dose producing a visually discernible erythema 24h post-UVR. A geometric series of 10 doses (7-80 mJ/cm2 for phototype II; 26.6–271 mJ/cm2 for V) of erythemally-weighted UVR was applied to buttock skin using a Waldmann UV236B unit with CF-L 36W/UV6 lamps (peak emission 313 nm; range 290–400 nm; Waldmann GmbH, Villingen Schwenningen, Germany). Simulated summer sunlight exposures Volunteers were given a six-week course of UVR-exposures, concordant with the length of the UK school summer holiday when the population is most exposed to sunlight, as described.1 A Philips HB588 Sunstudio irradiation cabinet (Eindhoven, The Netherlands) delivered whole-body UVR-exposure after fitting with alternating Arimed B (Cosmedico GmbH, Stuttgart, Germany) and Cleo Natural (Philips, Eindhoven, The Netherlands) fluorescent tubes, providing UVR-emission 103 close to UK summer sunlight (95% UVA: 320–400 nm, 5% UVB: 290–320 nm). Cabinet emission was characterised using a DTM300 spectroradiometer (Bentham, Reading, UK) and monitored using an Ocean Optics S2000 spectroradiometer (Ocean Optics, Dunedin, USA). Wearing protective eye-goggles, standardised T- shirts and knee-length shorts, volunteers lay prone, exposing ~35% skin surface in total. 1.3 SED UVR-exposures were given x3 weekly in January/February, when ambient UVB is negligible at UK latitudes and people are at trough vitamin D status,27 with exposure-time adjusted to maintain constant dosing.28 Doses took ~6.5min to administer, equating to 13-17min exposure to UK June midday sunlight exposure x6 weekly, which takes account that (i) when horizontal, ventral and dorsal surfaces are exposed sequentially in sunlight, not simultaneously as in the cabinet (ii) in daily life postures range from horizontal to vertical randomly- orientated to the sun.29 To compare UVR-exposed/protected sites, a 10x10 cm2 aperture was made in the shorts material over one buttock; the contralateral buttock was covered with UVR-opaque material. Dietary vitamin D logs Volunteers completed daily dietary-logs of vitamin D-fortified foods, and six key food categories: cheese; butter/margarine/oily spreads; milk/milk-containing products; red meat; oily fish; eggs/egg dishes; during first and last study weeks.30 Vitamin D content was obtained from food package labelling and McCance and Widdowson’s “The Composition of Food”.31 Vitamin D, parathyroid hormone, and serum biochemistry Blood samples were taken weekly, and serum stored at −20 °C until study- 104 completion. Serum 25(OH)D was measured by HPLC-UV, as reported.32 The laboratory was accredited to ISO 9001:2008 and 13485:2003 standards, and certified proficient by the national vitamin D quality assurance scheme (DEQAS). PTH was measured at course-beginning and -end, and serum biochemistry analyzed.1 25(OH)D levels of 25nmol/L (10ng/ml) and 50nmol/L (20ng/ml) were defined as deficiency and sufficiency cut-offs respectively.23,33,34 Skin Colour Measurements UVR-exposed/protected buttock skin colour was measured at baseline and weekly (CM-2500d spectrophotometer, Konica Minolta, Tokyo, Japan). Triplicate standard L*a*b* data were recorded.35 Individual Typology Angle (ITA) was calculated as the vector direction in L*b* plane, as arctangent [(L*-50)/b*]x(180/π).36-38 Cutaneous sampling Following the UVR-course, all participants had four 4mm punch biopsies from buttock skin: photoprotected skin, skin immediately-post 1x 1.3 SED, immediately- post 18x 1.3 SED, and 24h following the 18 exposures. Biopsies were formalin- fixed and paraffin-embedded for histological analysis. Cutaneous CPD immunostaining Immunostaining used a modification of Tewari et al’s method:13 4μm sections were treated with 0.1% trypsin; hydrogen peroxide (0.3% in methanol) was added to inhibit endogenous peroxidase; blocking buffer (Vector Laboratories, Cambridgeshire, UK) was added followed by monoclonal antibody incubation (TDM-2 [CosmoBio, Tokyo, Japan], 1:2,000).39 Primary antibody was omitted from one slide/staining-cycle as negative control. Slides were incubated with 105 biotinylated secondary antibody before addition of ABC solution, developed with Vector SG solution and counterstained with Nuclear Fast Red (Vector Laboratories) before dehydration and mounting. Images were scanned (Pannoramic 250 Flash II, 3DHisTech Ltd, Budapest, Hungary) and analysed for epidermal thickness and area (Image J 1.48, NIH, Bethesda, MD). Positively- staining nuclei were counted per hpf (original magnification x40; 3 hpf/section, 9 hpf/slide). The researcher (SJF) was blinded to slide-identity. Urinary analyses for DNA damage First-void urine samples, collected daily (Monday-Friday) during week one to assess for early impact, and then weekly to assess for accumulation of DNA damage, were stored at -20 °C until processing. Quantification of urinary 8-oxodG Samples were analysed for 8-oxodG using UHPLC-MS/MS as described.40 Results were normalised using urinary creatinine. Quantification of urinary thymine dimers We developed a UHPLC-MS/MS assay for cis,syn T<>pT in urine, which benefits from stable isotope labelled internal standardisation, is more rapid than the HPLC 32P-postlabelling method, avoids need for 32P, and provides absolute quantification, unlike ELISA. CPD are removed from DNA by NER, as a lesion- containing single-stranded oligomer approximately 24-32 nucleotides long.41 These oligomers are subject to 5’→3’ exonucleolytic attack, generating lesion- containing 6- and 7-mers, with some 2-mers. Current methodology for measuring CPD in urine is HPLC pre-purification followed by 32P-postlabelling. This approach 106 quantifies the dimer as a dinucleotide monophosphate (the dimerised form of thymidylyl-3’-5’thymidine, T<>pT).42 However, potential exists for dimers to be present in urine as other oligomeric forms. Therefore we adopted two approaches: the first quantifies T<>pT, the second utilizes formic acid hydrolysis of urine to render all oligomeric forms down to the nucleobase form of the dimer (thymine-thymine dimer, T<>T). Methods are detailed in supplementary material. Statistical analyses Paired and unpaired t-test, repeated measures analyses and linear regressions, were performed using SPSS statistical software (version 21.0.0; Chicago, USA) and GraphPad Prism (version 6; La Jolla, USA). Ratio measures, logarithmically transformed to make normally-distributed, were considered statistically significant if p<0.05. 107 RESULTS Volunteers Volunteers were compliant with study procedures and all completed the study. Table 1 displays baseline characteristics; general serum biochemistry was normal. Baseline serum PTH appeared lower (non-significant) for phototype II than V, and did not change significantly. Dietary vitamin D was low, 80% of phototype II and 83% of V ingesting <5µg/day, and constant between weeks. Serum 25(OH)D gain The 6-week course produced greater mean serum 25(OH)D gain in phototype II volunteers: 17.8 (SD 4.8)nmol/L versus 8.3 (10.5)nmol/L for phototype V (p<0.05; Fig 1). Gain was inversely associated with baseline 25(OH)D for phototype II (R2=0.4; p=0.049) but not phototype V. However, proportional gain in 25(OH)D from baseline was almost identical, mean increase 49% in phototype II, from 36.5 (13.0) at baseline to 54.3 (10.5)nmol/L at course-end, and 48% from 17.2 (6.3) to 25.5 (9.5)nmol/L in V. Post-UVR level was positively associated with baseline 25(OH)D (p<0.001), consistent with previous studies.1,43 Skin darkening At baseline, mean L* (skin lightness) was 69 (2.8) in phototype II and 41 (12.8) in V, with mean ITA of 52 (5.7)° and -22 (33.3)°, respectively. The 6-weeks’ exposures produced significantly greater darkening in phototype V than in phototype II as indicated by reduction in L* (p=0.02), although this did not reach significance for ITA. ITA decrease (darkening) was positively associated with 25(OH)D gain for phototype II (R2=0.54, p=0.016) but not V, in whom there was wide inter-individual variation in ITA and less 25(OH)D gain. 108 Cutaneous CPD Skin section examination showed UVR did not induce epidermal thickening in either phototype (data not shown). In the absence of UVR-exposure, no CPD were detectable in any individual (Fig 2A, E). One 1.3 SED exposure caused a range of CPD levels in phototype II (median count 200 [range 16.5-284] CPD-positive nuclei/mm2, Fig 2B), whilst only two phototype V volunteers showed any evidence of CPD (counts of 4 and 16/mm2; Fig 2E). Skin receiving cumulative UVR (18x 1.3 SED) showed elevated CPD-positive nuclei counts in phototype II (234 [125- 314]/mm2; Fig 2C) versus V (12 [0-148]/mm2; p<0.001). No significant difference was seen in CPD after cumulative versus single exposure for either phototype. At 24h after 6-weeks’ exposures, phototype II had cleared mean 44% (range 27-84%) of their cutaneous CPD-positive nuclei, while V had cleared 97% (84-100%; p<0.0001; Fig 2D, E). Phototype II showed a positive association of CPD-positive nuclei induction with baseline ITA (R2=0.49; p=0.02), but weak, non-significant associations with baseline L* (R2=0.29), age (R2=0.33) and 25(OH)D gain (R2=0.23). Urinary DNA damage CPD (T<>T and T<>pT) were undetectable for both phototypes, at baseline and after the UVR-course. At baseline, phototype II had higher urinary 8-oxodG (mean 2.72 [SD 0.97] pmol/µmol creatinine) than V (0.96 [0.28] pmol/µmol creatinine), p<0.001, with no significant increase during any of the days measured in week 1 (Fig 3A). Moreover, whilst 8-oxodG levels were higher in phototype II at all time points (repeated measures analysis; p=0.001; Fig 3B), there was no accumulation in urinary 8-oxodG over the 6-week course. 109 DISCUSSION To our knowledge, this is the first study examining vitamin D benefit and cutaneous DNA damage/repair concurrently following low-level UVR-exposures. Employing radiation similar to summer solar UVR-emission and protocols simulating repeated casual exposures, UVR-doses were equivalent to 13-17 min UK June midday exposure, on most days of the week (latitude 53.5°N).29 Such exposures have been assumed to provide adequate vitamin D status, and were shown to attain serum 25(OH)D levels equating to sufficiency (50nmol/L)23 in white Caucasians,1 while concurrent DNA damage outcome-assessments (cutaneous and urinary) has awaited exploration. We demonstrated that low-level exposures readily induced CPD in keratinocytes in white Caucasian skin (phototype II) and, to a much lesser extent, South Asian skin (brown, phototype V), in vivo. Induction was significantly positively associated with skin pallor (baseline ITA), consistent with a recent ex vivo human skin study.38 Comparison of DNA damage induced by one 1.3 SED with that following 18 doses revealed close similarity. There was no evidence for regular low-level exposures leading to DNA damage accumulation, indicating effective repair between exposures. Sheehan et al described accumulation of CPD-positive nuclei with repeated 0.65 MED exposures in phototypes II and IV.44 However, this involved an MED- adjusted not absolute UVR-dose, and exposures at shorter intervals (Monday- Friday for two weeks). Another human study reported 48-72h for CPD-positive nuclei levels to return to baseline following a single higher (1.2 MED) exposure.45 In human keratinocytes in vitro, following low-level UVB (8 mJ/cm2, ~two-fold lower than phototype I MED) on eight consecutive days, very few CPD-positive nuclei had been repaired 24h post-UVR.46 Similarly, mice given repeated low-level 110 UVB (0.5 kJ/m2 every 24h for 40 consecutive days) showed CPD repair lagged behind formation, leading to damage accumulation.9 The low-level UVR we employed may cause insufficient DNA damage to overwhelm repair,44,47 or the 48h intervals between exposures could provide sufficient repair-time. It is also feasible that repair mechanisms are upregulated by repeated low-level exposures. Since CPD persistence can lead to mutagenesis, and repair-kinetics in human skin are most rapid within 24h,48 we quantified CPD-positive nuclei in biopsies taken 24h post-UVR. In phototype II, a mean of 44% of CPD-positive nuclei were cleared versus virtually all (97%) in phototype V, where initial level of damage was much lower. Sheehan et al.’s cumulative UVR study also found more complete repair in skin type IV than II at one-week post-UVR.44 The decrease in CPD-positive nuclei we observed at 24h showed significant inter-individual variation within phototype II (27-84%). From a human health perspective, it was encouraging CPD did not accumulate over the UVR-course; nevertheless a substantial % damaged cells were still present 24h post-UVR, and the potential remains for mutagenesis after each DNA-damaging event. Interestingly, following both single and repeated (x18) low-level UVR- exposures, urinary CPD remained below detection limit, and oxidatively-damaged DNA did not increase from baseline, in either phototype. Concurrent skin section analysis confirmed CPD-induction, but lack of urinary CPD detection suggests damage was relatively small, and/or number of cells affected was insufficient to generate a signal in urine. This conclusion is supported by the urinary 8-oxodG findings. Our previous study showed urinary 8-oxodG increases four days following single, whole-body suberythemal (15 J/cm2) UVA exposure in vivo,17 suggesting our levels of UVR-exposure (reflecting UVR-dose and surface area exposed) were insufficient to induce urinary 8-oxodG, a sensitive biomarker of 111 oxidative stress. Intriguingly, phototype II had greater urinary 8-oxodG than V at all time points, implying a non-UVR explanation, such as differences in metabolism, repair and/or antioxidant intake; this warrants future exploration. Studies examining impact of melanin on vitamin D synthesis in vivo show conflicting results,2,4,6,7,49 potentially through differences in skin-site, baseline 25(OH)D, UVR-dose and spectrum.50 We found most phototype II participants reached vitamin D sufficiency [25(OH)D≥50nmol/L] consistent with our larger sample of white Caucasians.1 Over half the South Asian volunteers left deficiency status, attaining ≥25nmol/L but none reached ≥50nmol/L, as in previous investigation in 15 South Asians.30 In addition to the higher constitutive pigmentation in phototype V, they had significantly greater skin darkening during the UVR course than II and this may be responsible for the lower 25(OH)D gain/plateau in phototype V. Facultative pigmentation includes involvement of higher epidermal levels,51 limiting UVB penetration to 7-dehydrocholesterol and hence initiation of vitamin D synthesis. A positive association between urinary T<>pT and 25(OH)D gain was reported following intense UVR-exposures (mean 60-101KJ/m2) during sun/ski holidays in phototypes I-IV.5 Liljendahl et al also identified significantly increased urinary T<>pT 3-5 days after two days’ beach-sunbathing in Sweden, urinary DNA damage strongly correlating with personal UVR-dosage (up to 1400J/m2).52 These high-dose exposures contrast with our brief sub-erythemal exposures, where association between cutaneous CPD-positive nuclei and 25(OH)D gain was weak and non-significant. Building on the present study, application of a dose-range of low-level UVR-exposures could assess whether there are doses where vitamin D benefit is gained with minimal DNA damage in light-skin adults, and similarly, a 112 dose-range of higher-level exposures4 could examine whether brown-skin individuals can achieve higher serum 25(OH)D gain, still with limited DNA damage. Strengths of this study are the original, concurrent examination of cutaneous CPD with urinary DNA-damage biomarkers and 25(OH)D gain, following low-level UVR. This simulation of northerly-latitude summer sunlight exposures employed UVR-emission close to midday sunlight, and examined 25(OH)D gain after repeated exposures to commonly-exposed skin sites. Completion of dietary logs indicated no alteration in vitamin D intake over the study. Future studies may explore findings in a wider range of phototypes, using differing patterns of UVR and natural sunlight exposure. Our findings indicate tailoring of public health policies on safe sun-exposure for different phototypes. Brown-skin individuals who experienced almost negligible DNA-damage but generated low amounts of 25(OH)D, could be advised on less-limited sun-exposure practice,4 while caution is required for phototype II, as unrepaired cutaneous DNA damage was seen at 24h following even the low UVR-doses employed, in these easy-burning individuals. Acknowledgements The authors acknowledge assistance of Marie Durkin in volunteer recruitment, Vilas Mistry in 8-oxodG measurement and Emma Harry for contribution to development of mass spectrometric assays for T<>T and T<>pT. Funding This project was in part funded by Cancer Research UK, project C20668/A10007. The funder had no involvement in study design, data collection, data analysis, manuscript preparation or publication decisions. 113 References 1. Rhodes LE, Webb AR, Fraser HI et al. Recommended summer sunlight exposure levels can produce sufficient (≥20 ng ml-1) but not the proposed optimal (≥32 ng ml-1) 25(OH)D levels at UK latitudes. J Invest Dermatol. 2010; 130:1411-8. 2. Bogh MK, Schmedes AV, Philipsen PA, et al. Vitamin D production after UVB exposure depends on baseline vitamin D and total cholesterol but not on skin pigmentation. J Invest Dermatol. 2010; 130:546-53. 3. Mckenzie R, Scragg R, Liley B, et al. Serum 25-hydroxyvitamin-D responses to multiple UV exposures from solaria: inferences for exposure to sunlight. Photochem Photobiol Sci. 2012; 11:1174-85. 4. Farrar MD, Webb AR, Kift R, et al. Efficacy of a dose range of simulated sunlight exposures in raising vitamin D status in South Asian adults: implications for targeted guidance on sun exposure. Am J Clin Nutr. 2013; 97:1210-6. 5. Petersen B, Wulf HC, Triguero-mas M, et al. Sun and Ski Holidays Improve Vitamin D Status, but Are Associated with High Levels of DNA Damage. J Invest Dermatol. 2014; 134:2806-13. 6. Clemens TL, Adams JS, Henderson SL, Holick MF. Increased skin pigment reduces the capacity of skin to synthesise vitamin D3. Lancet. 1982;1:74-6. 7. Lo CW, Paris PW, Holick MF. Indian and Pakistani immigrants have the same capacity as Caucasians to produce vitamin D in response to ultraviolet irradiation. Am J Clin Nutr. 1986;44:683-5. 8. IARC (1992). IARC Monographs on the evaluation of carcinogenic risks to humans. Solar and ultraviolet radiation. IARC Monogr Eval Carcinog Risks Hum, 55:1–316. 9. Mitchell DL, Greinert R, de Gruijl FR et al. Effects of chronic low-dose ultraviolet B radiation on DNA damage and repair in mouse skin. Cancer Res 1999; 59:2875- 114 84. 10. Kappes UP, Luo D, Potter M, et al. Short- and long-wave UV light (UVB and UVA) induce similar mutations in human skin cells. J Invest Dermatol. 2006; 126:667-75. 11. Yoon JH, Lee CS, O’Connor TR et al. The DNA damage spectrum produced by simulated sunlight. J Mol Biol 2000; 299:681-93. 12. Ziegler A, Jonason AS, Leffell DJ et al. Sunburn and p53 in the onset of skin cancer. Nature 1994; 372:773-6. 13. Tewari A, Sarkany RP, Young AR. UVA1 induces cyclobutane pyrimidine dimers but not 6-4 photoproducts in human skin in vivo. J Invest Dermatol. 2012; 132:394- 400. 14. Kvam E, Tyrrell RM. Induction of oxidative DNA base damage in human skin cells by UV and near visible radiation. Carcinogenesis. 1997; 18:2379-84. 15. Eller MS, Ostrom K, Gilchrest BA. DNA damage enhances melanogenesis. Proc Natl Acad Sci U S A 1996; 93:1087-92. 16. Brenner M and Hearing VJ. The protective role of melanin against UV damage in human skin. Photochem Photobiol. 2008; 84:539–549. 17. Cooke MS, Evans MD, Burd RM et al. Induction and Excretion of Ultraviolet- Induced 8-Oxo-2'-deoxyguanosine and Thymine Dimers In Vivo: Implications for PUVA. J Invest Dermatol 2001; 116:281-285. 18. Hollis BW. Circulating 25-hydroxyvitamin D levels indicative of vitamin D sufficiency: implications for establishing a new effective dietary intake recommendation for vitamin D. J Nutr 2005; 135:317-22. 19. Grant, WB. Epidemiology of disease risks in relation to vitamin D insufficiency. Prog Biophys Mol Biol 2006; 92:65-79. 20. Vieth R. What is the optimal vitamin D status for health? Prog Biophys Mol Biol 2006; 92:26-32. 115 21. IARC (2008). Vitamin D and Cancer. IARC Working Group Reports, World Health Organization. 22. Berry JL, Davies M, Mee AP. Vitamin D metabolism, rickets, and osteomalacia. Semin Musculoskelet Radiol. 2002; 6:173-82. 23. IOM (Institute of Medicine). 2011. Dietary Reference Intakes for Calcium and Vitamin D. Washington, DC: The National Academies Press. 24. Ziegelberger G, Repacholi M, McKinlay A. International commission on non- ionizing radiation protection. Prog Biophys Mol Biol 2006; 92:1–3. 25. NRPB (2002). Health Effects from Ultraviolet Radiation: Report of an Advisory Group on Non-Ionising Radiation. Documents of the NRPB 13: 207-17. 26. Cooke MS, Olinski R, Loft S. Measurement and meaning of oxidatively modified DNA lesions in urine. Cancer Epidemiol Biomarkers Prev. 2008; 17:3-14. 27. Webb AR, Kift R, Durkin MT, et al. The role of sunlight exposure in determining the vitamin D status of the U.K. white adult population. Br J Dermatol. 2010; 163:1050-5. 28. Taylor DK, Anstey AV, Coleman AJ et al. Guidelines for dosimetry and calibration in ultraviolet radiation therapy: a report of a British Photodermatology Group workshop. Br J Dermatol 2002; 146:755–63. 29. Webb AR, Kift R, Berry JL, et al. The vitamin D debate: translating controlled experiments into reality for human sun exposure times. Photochem Photobiol 2011; 87:741–5. 30. Farrar MD, Kift R, Felton SJ, et al. Recommended summer sunlight exposure amounts fail to produce sufficient vitamin D status in UK adults of South Asian origin. Am J Clin Nutr. 2011; 94:1219-24. 31. Food Standards Agency. McCance and Widdowson's The Composition of Foods. 6th ed. Cambridge, United Kingdom: Royal Society of Chemistry, 2002. 116 32. Berry JL, Selby PL, Davies M, et al. Observations from the UK Supra-Regional Assay Service laboratory for the measurement of Vitamin D metabolites. J Steroid Biochem Mol Biol. 2007; 103:477-9. 33. Department of Health (Great Britain). Nutrition and bone health with particular reference to calcium and vitamin D: Report of the subgroup on bone health, working group on the nutritional status of the population of the committee on medical aspects of food and nutrition policy. Stationery Office, London, 1998. 34. Ashwell M, Stone EM, Stolte H, et al. UK Food Standards Agency Workshop Report: an investigation of the relative contributions of diet and sunlight to vitamin D status. Br J Nutr 2010; 104:603–11. 35. Robertson AR. The CIE 1976 color difference formulas. Color Res Appl. 1977; 2:7-11. 36. Chardon A, Cretois I, Hourseau C. Skin colour typology and suntanning pathways. Int J Cosmet Sci. 1991; 13:191-208. 37. Del bino S, Sok J, Bessac E, et al. Relationship between skin response to ultraviolet exposure and skin color type. Pigment Cell Res. 2006; 19:606-14. 38. Del bino S, Bernerd F. Variations in skin colour and the biological consequences of ultraviolet radiation exposure. Br J Dermatol. 2013; 169 Suppl 3:33-40. 39. Mori T, Nakane M, Hattori T, et al. Simultaneous establishment of monoclonal antibodies specific for either cyclobutane pyrimidine dimer or (6-4) photoproduct from the same mouse immunized with ultraviolet-irradiated DNA. Photochem Photobiol. 1991; 54:225-32. 40. Lam PM, Mistry V, Marczylo TH, et al. Rapid measurement of 8-oxo-7,8- dihydro-2'-deoxyguanosine in human biological matrices using ultra-high- performance liquid chromatography-tandem mass spectrometry. Free Radic Biol Med. 2012; 52:2057-63. 117 41. Galloway AM, Liuzzi M, Paterson MC. Metabolic processing of cyclobutyl pyrimidine dimers and (6-4) photoproducts in UV-treated human cells. Evidence for distinct excision-repair pathways. J Biol Chem. 1994; 269:974-80. 42. Le Curieux F, Hemminki K. Cyclobutane thymidine dimers are present in human urine following sun exposure: Quantitation using 32P-postlabeling and High-performance liquid chromatography. J Invest Dermatol 2001; 117:263-268. 43. Moan J, Porojnicu AC, Dahlback A, et al. Addressing the health benefits and risks, involving vitamin D or skin cancer, of increased sun exposure. Proc Natl Acad Sci USA. 2008; 105:668-73. 44. Sheehan JM, Cragg N, Chadwick CA et al. Repeated ultraviolet exposure affords the same protection against DNA photodamage and erythema in human skin types II and IV but is associated with faster DNA repair in skin type IV. J Invest Dermatol 2002; 118:825-9. 45. De Winter S, Vink AA, Roza L, et al. Solar-simulated skin adaptation and its effect on subsequent UV-induced epidermal DNA damage. J Invest Dermatol. 2001; 117:678-82. 46. Chouinard N, Therrien JP, Mitchell DL et al. Repeated exposures of human skin equivalent to low doses of ultraviolet-B radiation lead to changes in cellular functions and accumulation of cyclobutane pyrimidine dimers. Biochem Cell Biol 2001; 79:507-15. 47. Greinert R, Boguhn O, Harder D, et al. The dose dependence of cyclobutane dimer induction and repair in UVB-irradiated human keratinocytes. Photochem Photobiol. 2000; 72:701-8. 48. Bykov VJ, Sheehan JM, Hemminki K, et al. In situ repair of cyclobutane pyrimidine dimers and 6-4 photoproducts in human skin exposed to solar simulating radiation. J Invest Dermatol. 1999; 112:326-31. 118 49. Armas LA, Dowell S, Akhter M, et al. Ultraviolet-B radiation increases serum 25- hydroxyvitamin D levels: the effect of UVB dose and skin color. J Am Acad Dermatol. 2007; 57:588-93. 50. Björn LO. Vitamin D synthesis may be independent of skin pigmentation only with UV of short wavelength. J Invest Dermatol. 2010; 130:2848-50. 51. Alaluf S, Atkins D, Barrett K, et al. Ethnic variation in melanin content and composition in photoexposed and photoprotected human skin. Pigment Cell Res. 2002; 15:112-8. 52. Liljendahl TS, Kotova N, Segerbäck D. Quantification of ultraviolet radiation- induced DNA damage in the urine of Swedish adults and children following exposure to sunlight. Biomarkers. 2012; 17:634-41. 119 Table 1: Subject demographics Phototype II Phototype V Participants (n) 10 6 Sex: Male, Female (n) 2, 8 4, 2 Mean SD Mean SD Age (y) 45 9 38 11 BMI (kg/m2) 26 4 26 3 MED (mJ/cm2) 37 13 146 64 Baseline PTH* (pmol/L) 2.3 0.9 3.8 1.7 Final PTH* (pmol/L) 2.3 0.7 3.1 1.2 Dietary vitamin D intake 3.1 2.7 2.6 2.5 week 1 (µg/day) Dietary vitamin D intake 3.3 2.6 2.0 1.4 week 6 (µg/day) Baseline 25(OH)D 36.5 13.0 17.2 6.3 (nmol/L) Final 25(OH)D (nmol/L) 54.3 10.5 25.5 9.5 * Normal PTH range 0.8-3.9 pmol/L 120 Figure Legends Figure 1: 25(OH)D levels during the simulated summer sunlight exposures Serum 25(OH)D increased during the six-week simulated summer UVR-exposures, with a plateau in both groups around week four. Values were significantly higher at all time-points in (A) phototype II individuals (n=10) than (B) phototype V (n=6). The 25(OH)D gain between baseline and week six was statistically significant in phototype II. Horizontal bars denote mean values, and horizontal lines represent 25(OH)D level deficiency and insufficiency cut-offs at 25 and 50nmol/L, respectively. * p<0.0001 Figure 2: Representative epidermal DNA damage in phototype II and V individuals under varying conditions of UVR-exposure. CPD-positive nuclei staining (black arrow) from a volunteer of phototype II (left column) and phototype V (right column). Original magnification x40. (a) Photoprotected skin (b) Immediately following one 1.3 SED exposure (c) Immediately following the completion of the six-weeks’ simulated summer sunlight exposures (d) 24h after the completion of the six-weeks’ simulated summer exposures (e) CPD-positive nuclei counts in phototype II (circles; n=10) and V volunteers (triangles; n=6). DNA damage was absent in photoprotected skin in both groups. Median CPD-positive nuclei counts were significantly higher in phototype II than V immediately following a single UVR-exposure, the six-week course of cumulative UVR- exposures, and 24h following the cumulative exposures (p<0.001 for all). In both phototypes, the six-weeks’ simulated summer sunlight exposures caused no statistically significant difference in CPD-positive nuclei compared with a single 1.3 SED exposure. Horizontal bars denote the median. Viable epidermal thickness 121 measurements did not differ for the two phototype groups were and unchanged by the simulated summer’s sunlight exposures. *p<0.0001. Figure 3: Urinary 8-oxodG damage (pmol/µmol creatinine) in skin type II and V volunteers Urinary 8-oxodG concentrations: (A) daily for first five days of week one and (B) weekly during the six weeks’ simulated summer. Subjects of phototype V (n=6) had significantly lower 8-oxodG levels than those of phototype II (n=10) both in the first five days (including prior to exposure) and during the study (p=0.001 and p=0.002 respectively, repeated measures). However, no increase in urinary oxidative DNA damage was seen at any of the time points following a single UVR- exposure during the first five days and no accumulation occurred over the six week course. Data shown are median, interquartile and full range. 122 Figure 1 A B 123 Figure 2 e 124 Figure 3 A. B. 125 Supplementary material Quantification of urinary thymine dimers by UHPLC-MS/MS Materials Acetonitrile (HPLC gradient grade), water (HPLC gradient grade) and mass spectrometry grade formic acid were purchased from Fisher Scientific (Loughborough, UK). Ammonium formate (>99.0%) and acetophenone were 15 13 purchased from Sigma-Aldrich (Poole, UK). Isotopically labelled ([ N2, C10]; CK Gas Products, Ibstock, UK) and unlabelled thymidylyl-(3’-5’)-thymidine (dTpdT) were synthesised by Carbosynth Limited (Berkshire, UK). Oasis WAX 50 mg (Waters Corporation, Manchester, UK), and ISOLUTE ENV+ 50 mg (Biotage, Uppsala, Sweden) solid-phase extraction (SPE) cartridges were purchased from their manufacturers. Preparation of unlabelled and stable isotope labelled thymine dimers Due to possible loss of analyte during sample workup, isotopically labelled internal standards were necessary for each target compound. The synthesis of both 2 labelled (α,α,α,6- H4) and unlabelled T<>T was based on the method of Wang (1961), as described by Podmore et al (1996). A frozen aqueous solution of thymine (2mg/mL) was stored at -80 °C for 12 hrs prior to being UVC irradiated (254 nm UVS-18 EI series UV lamp) on dry ice for 4 hrs. The solution was then thawed and the resulting precipitate isolated by centrifugation (1200 x g) for 20 mins, and then dried at 40 °C. This method has been shown previously to produce a standard free from impurity by thymine (Podmore et al., 1996). Labelled thymidylyl-3’-5’thymidine (TpT) was custom synthesised by Carbosynth 13 15 Ltd (Compton, UK) from [ C10, N2]-dThy. Only one molecule of dThy per TpT 126 13 15 was labelled. Labelled ([ C10, N2]-T<>pT) and unlabeled T<>pT were generated by UVR exposure in the presence of a photosensitizer, based upon the method of Liu and Yang (1978). Oxygen was purged from a solution of TpT (5mg/22.7 ml) in HPLC gradient grade water, and 13.2 mM acetophenone, by bubbling nitrogen through the solution. The solution was then UVB irradiated (Philips TL-01 lamp) for 3.5 hrs under nitrogen, at 10 °C. The resulting photoproducts (cis-syn and trans-syn T<>pT) were isolated in the Department of Chemistry (University of Leicester) using preparative, reversed phase HPLC consisting of a Dionex Ultimate 3000 system and a Phenomenex Gemini NX 5µm C18 110Å AXIA column (250 x 21.20mm). Separation was achieved during a 35 mins isocratic run with 99.9% water and 0.1% trifluoroacetic acid, with a flow rate of 10.6 ml/min. Only two peaks were seen in the chromatogram at 13.9 and 18.3 mins, both peaks were collected and subsequently freeze dried, the resulting yield was 2.7mg and <1 mg respectively. Urine pre-treatment and solid phase extraction All urine samples were kept frozen at -20 °C until analysis, when they were thawed at room temperature and then centrifuged. Any pellet was discarded and the supernatant used in the subsequent steps. For T<>T, urine supernatants (10 µl) were spiked with internal standard prior to formic acid hydrolysis, and solid phase extraction on ISOLUTE ENV+ SPE columns. Following SPE, the resulting eluate was evaporated to dryness under nitrogen before reconstitution in 50 µl mobile phase. For T<>pT, 150 µl of urine supernatant was spiked with internal standard and diluted 1:1 with 2% formic acid. Oasis WAX SPE columns were conditioned with 1mL methanol, then equilibrated with 1 ml of water, prior to addition of the 300 µl sample (vacuum sufficient to achieve 1ml/min). The columns were then washed 127 with 1 ml 2% formic acid, and then eluted with 1ml of methanol then 300 µl of 5% (w/v) ammonium hydroxide in methanol (x2). The resulting eluate was evaporated to dryness under nitrogen before reconstitution in 50 µl mobile phase. UHPC-MS/MS analyses Experiments were performed on a Quattro Premier spectrometer (Waters Corporation, Manchester, UK), operated in negative ion mode with the ESI capillary voltage set to 3 kV and the cone voltage to 60 V. The nitrogen desolvation and cone gas flow rates were set to 1000 and 50 Lh−1 respectively, with a source temperature of 120 °C and desolvation gas temperature set to 300 °C. The mass spectrometer was operated in multiple reaction monitoring mode. For cis,syn 2 T<>T three transitions were identified: 253>210, 253>210 and 253>139 ([ H8]- T<>T: 261>159). For cis,syn T<>pT, three mass transitions were identified 13 15 545>447, 545>253, and 545>195 (([U- C10, U- N2]-T<>pT: 557>431, 557>333, 557>298, and 557>125). Conditions for each was as follows: entry and exit voltage of 30 eV with a collision energy of 31 eV and an Argon collision cell pressure of 5.9e-003. Masslynx version 4.1 (Waters Corporation, Manchester, UK) was used to control the IM-MS instrument and for data acquisition and processing. UHPLC was performed on a Waters Acquity system (Waters Corporation, Manchester, UK) with an Acquity UHPLC HSS T3® (Waters Corporation, Manchester, UK) column (2.1 x 50 mm). The UHPLC system was coupled to a post-column T-piece fitting (Waters Corporation, Manchester, UK) which was also connected to a further Dionex HPLC system (Thermo Fisher UK Ltd, Hemel Hempstead, UK) with the outlet going to the ESI ion source of the mass spectrometer. Cleaned up urine samples (5 µl injected) were eluted with the following gradient: 99.9% A (0-2 mins), increased to 30% B (2-4 mins) and then to 99.9% A (4-6 mins), where A=2.5 mM ammonium formate 128 and B=acetonitrile. The mobile phase flow rate was set to 0.1 ml/min. The additional HPLC flow rate was set to 0.1 ml/min employing an isocratic flow of acetonitrile to aid ionisation. The limit of detection was 396 fmol. Supplementary References Liu F-T, Yang NC. Photochemistry of cytosine derivatives. 1. Photochemistry of thymidylyl-(3’5’)-deoxycytidine. Biochemistry 1978; 17:4865-4876. Podmore ID, Cooke MS, Herbert KE et al. Quantitative Determination of Cyclobutane Thymine Dimers in DNA by Stable Isotope-Dilution Mass Spectrometry. Photochem Photobiol 1996; 64:310-315. Wang SY. Photochemical reactions in frozen solutions. Nature 1961; 190:690-694. 129 CHAPTER 4: Manuscript 2 130 Photoprotection conferred by low level summer sunlight exposures against pro-inflammatory UVR insult Running head: Protective effect of sunlight exposures in UVR-inflammation SJ Felton, BB Shih, REB Watson, LE Rhodes Division of Musculoskeletal and Dermatological Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science Centre, Salford Royal NHS Foundation Trust, Manchester, UK. Correspondence to: Prof L E Rhodes, Photobiology Unit, Dermatology Centre, University of Manchester, Salford Royal Hospital, Manchester, M6 8HD, UK Tel 00 44 161 206 1150 Email: [email protected] Word count: 2996 Figures: 4 Tables: 0 131 What is already known about this topic? Tanning, i.e. increased epidermal pigmentation and thickening, can develop as a photoprotective response to UVR exposure. What does this study add? Low level simulated summer sunlight exposures induce measurable increases in skin pigmentation and epidermal thickness in easy-burning (phototype II), individuals, where we demonstrate this to provide a degree of photoprotection against both clinical (i.e. skin erythema) and histological (i.e. dermal neutrophil infiltration) aspects of UVR-induced inflammation. 132 Abstract Background: Tanning, comprising increased pigmentation and epidermal thickening, develops as a photoprotective response following ultraviolet radiation (UVR) exposure, but it is unclear to what degree this occurs in lightly-pigmented individuals, or how this may impact on the histological as well as the clinical, inflammatory response to UVR. Objectives: To examine whether, in light-skinned people, simulating a summer’s low-level sunlight exposures induced melanogenesis, epidermal thickening and photoprotection against a higher-level, pro-inflammatory UVR challenge, assessed both clinically as erythema and histologically by neutrophil infiltration. Methods: During wintertime, when ambient UVB is negligible, healthy volunteers (n=10; phototype II; median age 47y, range 30-59y) received 1.3 standard erythemal dose (SED; 5% UVB, 95% UVA) thrice weekly for 6 weeks. All were subsequently challenged with 2X minimal erythema dose (MED) UVB on small areas of UVR-exposed and photoprotected buttock skin. Skin colour measurements were taken spectrophotometrically and skin biopsy sections assessed for epidermal thickness, and for neutrophil infiltration by immunohistochemistry. Results: The 6-week UVR course significantly increased both skin pigmentation, skin lightness (L*) reducing from 69.37 (SD 2.8) to 65.52 (2.33) at course-end (p<0.001), and stratum corneum thickness, from 29.3 (9.59) to 41.5 (12.7)µm (p<0.05). Higher–dose UVB (2X MED) challenge produced 18% less erythema and 71% less neutrophil infiltration in skin exposed to the 6-week UVR course than photoprotected skin (both p<0.05). Conclusions: Low level simulated sunlight exposures measurably increased skin pigmentation and epidermal thickness in phototype II individuals, which provided 133 some protection against both clinical and histological aspects of UVR-induced inflammation. 134 Abbreviations EI Erythema Index Hpf High power field ITA Individual Typology Angle MED Minimal erythemal dose SED Standard erythemal dose SSR Solar-simulated radiation UVR Ultraviolet radiation 135 INTRODUCTION Ultraviolet radiation (UVR) from summer sunlight incident on skin comprises approximately 95% UVA (400-320nm) and 5% UVB (320-290nm). Biological effects include vitamin D formation, immunomodulation and DNA damage.1,2 Tanning responses, i.e. increased pigmentation and epidermal thickening, are photoprotective,3 with UVA and UVB components showing differential effects and kinetics.4,5 In general, delayed tanning develops as a protective response 2-3 days after UVR-induced DNA damage, with p53 triggering increased eumelanin production and spatial rearrangement of melanosomes over basal keratinocyte nuclei, providing some protection against further DNA damage.1,6,7 Solar-simulated radiation (SSR) reportedly induces greater delayed pigmentation than UVB alone, suggesting synergism of UVA and UVB.8 Indeed, degree of melanogenesis is influenced by UVR dose, dose interval and emission spectra.9 Epidermal thickening is another photoadaptation against UVR-induced damage, reducing the UVR reaching the dividing basal cells.10,11 Both UVB and, in much higher doses, UVA can cause epidermal hyperplasia, potentially through increased activity of the pentose shunt enzyme, glucose-6-phosphate dehydrogenase.12 More recently, whole genome microarray studies identified alterations in JAK-STAT, p53 and p38 stress-responsive pathways following SSR exposure in ex vivo human skin.13 In mouse models, acceleration of DNA, RNA and protein synthesis, with increased mitosis and basal cell turnover, is observed 24h post-UVB, before epidermal hyperplasia.14 While UVR is noted to induce epidermal hyperplasia, particularly stratum corneum thickening, under certain conditions,15,16 the extent to which repeated low-dose exposures, gained as with casual brief summer sunlight exposure, influences epidermal thickness (both viable epidermis and stratum corneum) in humans in vivo is unknown. 136 Acute UVR-induced inflammation, i.e. ‘sunburn’, presents as cutaneous erythema, accompanied by pain, heat and swelling at higher exposure dose; UVB is orders of magnitude more effective than UVA at inducing erythema.17 This acute inflammatory response follows release of a cascade of inflammatory mediators, including nitric oxide, tumour necrosis factor-α, interleukins and sequential eicosanoid profiles.18-20 Alongside dilatation of dermal vasculature, upregulation of endothelial adhesion molecules and neutrophil chemoattractants results in neutrophil adherence to vascular endothelium, diapedesis and migration to UVR- exposed sites.18,21,22 The erythema time course is modified by UVB dose, although is typically maximal 24h post-exposure, accompanied by peak neutrophil infiltration,19,23 while maximal lymphocytic and macrophage influx occurs between 24-72h.20,24 Neutrophils have manifold functions, 23,25-28 including release of reactive oxygen species and proteolytic enzymes, immunomodulation, and phagocytosis of damaged cells. Sunburn and tanning responses indicate DNA damage has occurred,29,30 with sunburn erythema used as proxy for skin cancer risk.31-33 Presently it is unclear what degree of photoprotection occurs after repeated UVR exposures under everyday conditions, and whether erythema reflects histological response. This study aimed to examine the impact of a simulated summer’s low-level sunlight exposures (6 week course; 95% UVA, 5% UVB) on skin melanogenesis and epidermal thickening in light-skin (phototype II) Caucasians, and their photoprotection against UVR-induced inflammatory challenge, assessed both clinically as erythema and histologically by neutrophil infiltration. 137 METHODS Study Subjects Healthy volunteers, phototype II, were recuited in Greater Manchester, UK (January 2010). Exclusion criteria were pregnancy, breastfeeding, taking photoactive medication or supplements containing vitamin D, history of skin cancer/photosensitivity, and sunbed-usage/sunbathing within 3 months prior to or during the study. Ethical approval was obtained from the North Manchester Research Ethics Committee (reference 09/H1014/73). The study adhered to Declaration of Helsinki principles. Minimal erythemal dose (MED) assessment The MED, the lowest dose of UVR to produced visually discernable erythema at 24h, was assessed for each subject at baseline. A geometric series of 10 doses (7– 80mJ/cm2) of erythemally-weighted UVR was applied over 2 horizontal rows of buttock skin with a Waldmann UV 236B unit containing Waldmann CF-L 36W/UV6 lamps (peak emission: 313nm; range: 290–400nm; Waldmann GmbH, Villinge- Schwenningen, Germany). Simulated summer sunlight exposures Volunteers were given a 6-week course of UVR exposures, concordant with the length of the summer school holiday period, as described.34 A Philips HB588 irradiation cabinet (Eindhoven, The Netherlands), fitted with 11 Arimed B (Cosmedico GmbH, Stuttgart, Germany) and 13 Cleo Natural (Philips, Eindhoven, The Netherlands) fluorescent tubes in an alternating pattern, delivered whole body UVR with emission spectrum close to summer sunlight (95% UVA: 320–400nm, 138 5% UVB: 290–320nm). The cabinet’s emission was characterised using a Bentham DTM300 spectroradiometer (Bentham, Reading, UK) and monitored using an Ocean Optics S2000 spectroradiometer (Ocean Optics, Dunedin, FL, USA). Wearing protective eye goggles, T-shirts and shorts, volunteers lay prone on the sunbed with the canopy closed. One buttock was fully photoprotected whilst contralaterally a cut-out panel revealed 10cm2 buttock skin throughout the UVR course. The simulated summer’s sunlight was given three times weekly in January and February when ambient UVB is negligible at UK latitudes (50–60°N).35 An exposure of 1.3 standard erythemal dose (SED)36 was given at each visit. It took approximately 6.5 mins to deliver this dose after accurate measurement of cabinet UV irradiance;37 a constant UVR dose was maintained throughout by adjusting for any decrease in irradiance by increasing delivery time. This is estimated to be equivalent to 13-17 mins unshaded sunlight exposure on a clear June midday at 53.5°, 6X weekly, which accounts for non-simultaneous exposure of ventral and dorsal surfaces in sunlight, and the range of postures adopted in daily life, from the horizontal to the vertical randomly orientated to the sun.38 2X MED UVB-challenge Following the simulated summer sunlight exposures, buttock skin was exposed to a pro-inflammatory challenge, individually-dosed for each participant to be twice their MED (2X MED) of UVB. This was applied using a Waldmann UV 236B unit to two 1cm2 areas of buttock skin, one area that had received the simulated summer’s UVR exposures, and a control site that had been photoprotected. Skin Colour Measurements 139 A Minolta CM-2500d hand-held spectrophotometer (Konica Minolta, Tokyo, Japan) was used for non-invasive measurement of skin colour. Measurements were made at baseline and weekly from UVR-exposed and UVR-protected buttock skin, and 24h after the 2X MED UVB-challenge. Data were recorded in standard three- dimensional Commission International de l’ Eclairage L*a*b* format; L*represents white-black differentiation, where L* of 100 is pure white and L* of 0 is pure black, a* values reflect the balance between green (negative) and increasing redness (positive) whilst b* is the differentiation between blue (negative) and yellow (positive).39 Readings were made in triplicate at each site. Individual typology angle (ITA) values were also calculated as the vector direction in the L*b* plane, as arctangent [(L*-50)/b*]x(180/π) on exposed and unexposed skin throughout the UVR course.40,41 A reflectance instrument (Erythema meter, Diastron, Andover, UK) was further used for non-invasive measurement of buttock skin erythema. Erythema index (EI) readings were taken in triplicate from 2X MED UVB-challenged skin, the UVR course-treated skin and adjacent non-irradiated skin. To obtain background-corrected data, induced erythema was calculated by subtracting a* or EI readings taken on proximal, untreated buttock from those on 2X MED UVB-challenged buttock skin. Cutaneous sampling At the end of the UVR course, four 4mm punch biopsies were taken, comprising immediately after the last 1.3X SED exposure from buttock skin that was: (i) unexposed and; (ii) exposed to the 6 week simulated summer sunlight, and a second set of biopsies was taken 24 hrs after 2X MED UVB exposure from buttock 140 skin that had been (iii) previously unexposed and (iv) exposed to the 6 week simulated sunlight. All samples were formalin-fixed and paraffin-embedded prior to histological analysis. Immunostaining 3μm skin sections were prepared for immunohistological analyses including a solubilising step utilising 0.5% Triton X-100. After washing in tris-buffered saline (TBS), hydrogen peroxide (0.3% diluted in phosphate-buffered solution, PBS) was added to inhibit endogenous peroxidase before repeated washing. Primary antibody was added (neutrophil elastase NP57, Dako, Glostrup, Denmark; dilution 1:100) in primary antibody diluent (Diagnostic Biosystems, California, USA). For control, primary antibody was not added to one slide for each staining cycle. ImmPress reagent (Vector Laboratories, California, USA) was added before repeated rinsing. Slides were developed with NovaRed (Vector Laboratories, California, USA), rinsed and counterstained with Mayer's haematoxylin. Sections were dehydrated, and permanently mounted using DPX (Sigma-Aldrich Inc, St. Louis, USA). Quantification of staining and epidermal thickness measurements Images were captured and scanned using Pannoramic 250 Flash II, 3DHisTech Ltd. (Budapest, Hungary) scanner. Counts were made of the number of positively staining nuclei per high power field (hpf; original magnification X20), in triplicate per section, with three hpf taken per section (9 hpf per slide). Cellular epidermal (from stratum basale to distal stratum granulosum) and stratum corneum thicknesses were measured using the Pannoramic Viewer, nine per slide. Epidermal and dermal areas per hpf were measured using Image J 1.48 (National 141 Institutes of Health, Bethesda, MD). The assessor (SJF) was blinded to sample identity for counting and measurements. Statistical analyses Data analyses, specifically paired and unpaired t-tests, and linear regressions, were performed using SPSS statistical software (version 21.0.0; IBM SPSS Statistics, Portsmouth, UK) and GraphPad Prism (version 6; GraphPad Software, La Jolla, CA, USA). Neutrophil counts per epidermal area were logarithmically transformed to stabilise their variance. Results were considered statistically significant if p<0.05. 142 RESULTS Eight participants were female, two male. Median age was 47 (range 30-59)yr, BMI 25 (range 22-35)kg/m2 and MED 30 (22-54)mJ/cm2 of erythemally-weighted UVR. A simulated summer’s low-level sunlight exposures significantly darkened the skin and thickened the stratum corneum The repeated 1.3 SED exposures increased skin pigmentation, L* reducing 5.5%, from mean 69.37 (SD 2.8) at baseline to 65.52 (2.33) at course-end (p<0.001; Fig 1A). Similarly, ITA decreased from 52 (5.7)° at baseline to 41 (6.4)° post-course (p<0.0001; Fig 1B). Mean viable epidermal thickness did not differ significantly between unexposed (39.1 [6.6]µm) and simulated sunlight-exposed skin (43.7 [8.1]µm; p>0.05), whilst the stratum corneum thickened significantly from 29.3 [9.6] to 41.5 [12.7]µm; p<0.05, Fig 2A). The stratum corneum thickening did not correlate with darkening (change in L* or ITA; R2=0.06, R2=0.24 respectively, p>0.05 for both). A pro-inflammatory (2X MED) UVB-challenge significantly thickened the viable epidermis of previously unexposed and exposed skin In previously unexposed skin, 2X MED UVB induced significant viable epidermal thickening, from 39.1 [6.6] to 48.1 [7.0]µm at 24h post-challenge (p<0.05, Fig 2B). In the simulated summer sunlight-exposed skin post-course, the 2X MED UVB increased the viable epidermal thickness from 43.7 [8.1] to 48.6 [6.4]µm (p=0.05). Thickening appeared due to epidermal oedema (spongiosis) not acanthosis, as keratinocyte spacing rather than number increased (Fig 2C, D). 143 In contrast to the impact on viable epidermis, the acute 2X MED UVB-challenge did not provoke measurable change to stratum corneum thickness, in either previously unexposed skin or skin following the simulated sunlight course (p>0.05 for both; Fig 2A). Degree of epidermal thickening in previously unexposed skin was significantly, negatively correlated with erythema induced by the 2X MED challenge (a*; R2=0.54; p<0.05). Simulated summer sunlight exposures significantly reduced erythemal responses to acute 2X MED UVB challenge 2X MED induced significantly greater erythema in previously unexposed skin compared with the simulated summer exposed skin: a* readings were 15% higher (9.9 [3.7] versus 8.5 [4.2]) and similarly EI values were 18% higher (80.8 [31.2] versus 66.3 [32.1] (p<0.05 for both within group comparisons; Fig 3A, B). Simulated summer’s sunlight exposures significantly reduced neutrophil infiltration response to acute 2X MED UVB challenge Dermal neutrophils were absent both in the unexposed buttock skin and that exposed to 6 weeks’ simulated sunlight course (Fig 4A, B). However, following 2X MED UVB-challenge, the induced dermal neutrophil infiltration was 71% lower in the exposed than previously unexposed skin (21.4 [24.1] versus 73.5 [72.5]/mm2; p<0.05; Fig 4C, D). In previously unexposed skin, the 2X MED UVB-induced dermal neutrophil infiltration was positively correlated with baseline pallor (ITA, R2=0.40) and with 144 erythemal response (R2=0.70, 0.72 for a*, EI respectively; p<0.05 for all); no significant associations were seen post exposure course. DISCUSSION UVR-induced tanning is recognised as hazardous because although it provides some photoprotection, DNA damage occurs during production of melanin.3 This study tested the potential photoprotective effect of low-level sunlight exposure, as following a simulated UK summer, against inflammatory challenge induced by higher UVR dose (2X MED UVB) in easy-burning white Caucasians, and found significant reduction in dermal neutrophil infiltration, as well as in erythema. To our knowledge, our study is novel in its assessment of the protection induced by repeated UVR exposures of human skin against histological in addition to clinical aspects of UVR-induced inflammation, and its performance under conditions mimicking real-life exposures. These repeated low-level exposures to radiation close to solar UVR reaching the earth’s surface in summer (95% UVA, 5% UVB) are equivalent to ~15 mins UK (53.5°N) June midday sunlight, on most days of the week. This is similar to guidance on summer sunlight exposure for white skin people, and highlights the impact of following such guidance.42 Typically, the more frequently assessed clinical erythema response to UVR is assumed to reflect the leukocytic response, while this work indicates a proportionately larger effect on neutrophil numbers. Photoprotection against sunburn erythema following a simulated summer’s sunlight exposures was quantified as 15-18% reduction, using two measures (a* and EI). The repeated low-dose UVR exposures themselves did not induce a cutaneous inflammatory response, contrasting with data from mouse models showing multiple low-dose 145 (sub-erythemal) UVR may cause low-level neutrophil infiltration.43 Rijken et al (2005) found that after single SSR challenge, neutrophils were detectable in human skin only after a dose ≥1 MED; but the effect of multiple challenge on neutrophil infiltration appears previously unreported in humans.44 In mouse models,43 the increased neutrophils are thought to contribute to photoageing through releasing proteolytic enzymes, particularly neutrophil elastase and matrix metalloproteinases, that can damage elastin fibres and collagen networks, contributing to solar elastosis. Whilst the role of neutrophil infiltration during the sunburn response is not fully elucidated, these cells have several pro-inflammatory and immunomodulatory activities,45 including release of reactive oxygen species, and the generation of chemotactic signals that attract further leucocytes to the site. 25,27,41 They also release Th2-associated cytokines including interleukins-4 and -10, which are immunosuppressive, and determine whether macrophages adopt a pro- or anti-inflammatory role. Moreover, they phagocytose damaged keratinocyte components, repairing UVR-induced injury.23,26 Thus neutrophils regulate host responses to erythemal dose UVR injury, and consequently, while our demonstrated reduction in neutrophil infiltration by photoadaptation reflects overall reduced UVR-induced inflammation, this may also be associated with loss of beneficial, pro-resolution effects. We quantified two key mechanisms that confer photoprotection, i.e. skin darkening and epidermal thickening. In our volunteers, who were carefully phototyped as skin type II, with report of poor tanning ability, the low-level exposures nevertheless induced a small but statistically significant skin-darkening, with L* decrease of 5.5%. This is in keeping with a report of 8% L* decrease in five 146 ‘lighter-skinned’ Caucasians following 3-weeks SSR exposures.16 We similarly quantified ITA as a measure of skin darkening, which takes account of b* (yellow- green) spectrophotometer measurements, and thus has been considered a more accurate indicator of skin colour.41 The simulated summer decreased ITA values by 11°, with individuals moving from the ‘very light’ to ‘light/intermediate’ category; although of small magnitude, correlations between ITA and sunburn cell-induction have been demonstrated ex vivo.41 An apparent plateau occurred at ~4 weeks, consistent with previously reported findings in darker, skin type V, people.46 Our data on epidermal thickness indicate that repeated low-level UVR exposures significantly thicken stratum corneum but have minimal impact on the viable epidermis. This is consistent with a study15 finding that 28 doses of 0.5 MED SSR (applied X5 weekly) to ‘untanned’ back skin of individuals of skin type I-III, produced a 22% thickening in stratum corneum without change of viable epidermal thickness; our greater (42%) stratum corneum thickening possibly reflects skin type and test site differences. In our homogeneous light skin subjects, stratum corneum thickening did not correlate with skin darkening. Interestingly, a UVR conduction study in ex vivo skin showed stratum corneum in phototypes I-III to filter >50% of incident radiation, whilst the filtering averaged only 20-30% in black skin samples.46 Adaptations of the stratum corneum may hold a greater role in increasing photoprotection in lighter skin types, partially compensating for the lesser skin darkening seen during a simulated summer’s exposures in people of light versus dark skin types.47, 48 Increased keratinocyte turnover is thought to be mostly responsible for stratum corneum thickening post-UVR with epidermal mitosis stimulated by increased proliferating cell nuclear antigen (PCNA) activity.49 Additional 147 alterations in stratum corneum lipid content might contribute, as increased serine palmitoyl transferase acitivity and associated increases in epidermal barrier lipids including ceramides, cholesterol and fatty acids have been detected following UVR exposure of murine skin.50 Our findings indicate that epidermal oedema, i.e. spongiosis, is an important contributor to epidermal “thickening” at 24 hrs following acute UVB. The spongiosis may have reduced spectrophotometer signal detection, accounting for the negative correlation of epidermal thickening with erythema. Human studies are scarce, but a feasibility study using high-frequency ultrasound imaging in four volunteers found epidermal thickening to be maximal 48 hrs post- single higher dose (2-3 MED) UVR (40% UVB; 60% UVA) to healthy back skin. However, this did not provide evidence that thickening was solely attributable to hyperplasia rather than spongiosis.51 Detailed time course studies in mouse skin at 3 to 72 hrs post- UVR showed PCNA expression initially confined to the basal layer, spreading upwards for 48 hrs post-irradiation.49 This was accompanied by a gradual increase in epidermal hyperplasia, with focal hyperplasia occurring as early as 6 hrs post- UVR, before becoming more uniform, and maximal 48 to 72 hrs after irradiation. Similarly, whilst metabolic changes occur within 24 hrs post-irradiation in mouse skin,13 it may take 24 to 72 hrs to detect measurable changes in thickness. In summary, our detailed non-invasive and invasive assessment shows that low-dose simulated sunlight exposures, as may be gained with repeated brief summer-time exposure, induces measurable skin pigmentation accompanied by stratum corneum thickening, in easy-burning (skin type II) individuals. These changes are accompanied by pronounced protection on higher dose UVR-induced neutrophil infiltration, alongside a slight, though statistically significant, protection against the accompanying erythema. While erythema is the most frequently 148 utilised UVR-inflammatory endpoint, this work highlights its proxy nature for histological inflammation. Since the erythemal component of sunburn, like tanning, is believed to be initiated by UVR-induced DNA damage, little protection is anticipated against the longer term UVR effect of skin cancer. Acknowledgements The authors acknowledge the assistance of Dr S Pilkington in optimisation of immunohistochemical protocol, MR Rashid for assistance with immunohistochemical staining and Dr R Kift and Professor A Webb for spectrophotometry measurements. 149 References 1. Eller MS, Ostrom K, Gilchrest BA. DNA damage enhances melanogenesis. Proc Natl Acad Sci U S A 1996; 93: 1087-92. 2. Schwarz T & Schwarz A. Molecular mechanisms of ultraviolet radiation-induced immunosuppression. Eur J Cell Biol 2011; 90: 560-4. 3. Kvam E & Tyrrell RM. The role of melanin in the induction of oxidative DNA base damage by ultraviolet A irradiation of DNA or melanoma cells. J. Invest. Dermatol., 1999; 113: 209–213. 4. Rosen CF, Seki Y, Farinelli W, et al. A comparison of the melanocyte response to narrow band UVA and UVB exposure in vivo. J Invest Dermatol. 1987;88:774-9. 5. Duval C, Régnier M, Schmidt R. Distinct melanogenic response of human melanocytes in mono-culture, in co-culture with keratinocytes and in reconstructed epidermis, to UV exposure. Pigment Cell Res. 2001;14:348-55. 6. Gniadecka M, Wulf HC, Mortensen NN et al. Photoprotection in vitiligo and normal skin. A quantitative assessment of the role of stratum corneum, viable epidermis and pigmentation. Acta Derm. Venereol. 1996;76: 429–432. 7. Brenner M and Hearing VJ. The protective role of melanin against UV damage in human skin. Photochem Photobiol. 2008; 84: 539–549. 8. Wolber R, Schlenz K, Wakamatsu K, et al. Pigmentation effects of solar-simulated radiation as compared with UVA and UVB radiation. Pigment Cell Melanoma Res. 2008;21:487-91. 9. Miller SA, Coelho SG, Zmudzka BZ, et al. Dynamics of pigmentation induction by repeated ultraviolet exposures: dose, dose interval and ultraviolet spectrum dependence. Br J Dermatol. 2008;159:921-30. 10. Bruls WA, Slaper H, van der Leun JC et al. Transmission of human epidermis and stratum corneum as a function of thickness in the ultraviolet and visible 150 wavelengths. Photochem. Photobiol.1984; 40: 485–494. 11. Miyamura Y, Coelho SG, Schlenz K, et al. The deceptive nature of UVA tanning versus the modest protective effects of UVB tanning on human skin. Pigment Cell Melanoma Res. 2011;24:136-47. 12. Pearse AD, Gaskell SA, Marks R. Epidermal changes in human skin following irradiation with either UVB or UVA. J Invest Dermatol. 1987;88:83-7. 13. Mouchet N, Adamski H, Bouvet R, et al. In vivo identification of solar radiation- responsive gene network: role of the p38 stress-dependent kinase. PLoS ONE. 2010;5:e10776. 14. Epstein JH, Fukuyama K, Fye K. Effects of ultraviolet radiation on the mitotic cycle and DNA, RNA and protein synthesis in mammalian epidermis in vivo. Photochem Photobiol. 1970;12(1):57-65. 15. Lavker RM, Gerberick GF, Veres D et al. Cumulative effects from repeated exposures to suberythemal doses of UVB and UVA in human skin. J Am Acad Dermatol. 1995;32:53-62. 16. de Winter S, Vink AA, Roza L, et al. Solar-simulated skin adaptation and its effect on subsequent UV-induced epidermal DNA damage. J Invest Dermatol. 2001;117:678-82. 17. Willis I, Cylus L. UVA erythema in skin: is it a sunburn? J Invest Dermatol. 1977;68:128-9. 18. Strickland I, Rhodes LE, Flanagan BF et al. TNF-alpha and IL-8 are upregulated in the epidermis of normal human skin after UVB exposure: correlation with neutrophil accumulation and E-selectin expression. J Invest Dermatol 1997; 108: 763-8. 19. Rhodes LE, Belgi G, Parslew R, et al. Ultraviolet-B-induced erythema is mediated by nitric oxide and prostaglandin E2 in combination. J Invest Dermatol 151 2001; 117: 880-5. 20. Rhodes LE, Gledhill K, Masoodi M, et al. The sunburn response in human skin is characterized by sequential eicosanoid profiles that may mediate its early and late phases. FASEB J. 2009;23:3947-56. 21. Andersen PH, Abrams K, Bjerring P et al. A time-correlation study of ultraviolet B-induced erythema measured by reflectance spectroscopy and laser Doppler flowmetry. Photodermatol Photoimmunol Photomed 1991; 8: 123-8. 22. Norris PG, Barker JN, Allen MH, et al. Adhesion molecule expression in polymorphic light eruption. J Invest Dermatol. 1992;99:504-8. 23. Teunissen MB, Piskin G, Di Nuzzo S, et al. Ultraviolet B radiation induces a transient appearance of IL-4+ neutrophils, which support the development of Th2 responses. J Immunol. 2002;168:3732-9. 24. Kang K, Gilliam AC, Chen G, et al. In human skin, UVB initiates early induction of IL-10 over IL-12 preferentially in the expanding dermal monocytic/macrophagic population. J Invest Dermatol. 1998;111:31-8. 25. Hawk JL, Murphy GM & Holden CA. The presence of neutrophils in human cutaneous ultraviolet-B inflammation. Br J Dermatol. 1988; 118: 27–30. 26. Shapiro SD. Neutrophil elastase: Path clearer, pathogen killer, or just pathologic? Am J Respir Cell Mol Biol 2002; 26: 266–268. 27. Piskin G, Bos JD & Teunissen MB. Neutrophils infiltrating ultraviolet B- irradiated normal human skin display high IL-10 expression. Arch. Dermatol. Res 2005; 296: 339–342. 28. Lee PL, van Weelden H & Bruijnzeel PLB. Neutrophil infiltration in normal human skin after exposure to different ultraviolet radiation sources. Photochem. Photobiol. 2008; 84:1528–34. 29. de Gruijl FR. Photocarcinogenesis: UVA vs. UVB radiation. Skin Pharmacology 152 and Applied Skin Physiology, 2002: 15: 316–320. 30. Pfeifer GP & Besaratinia A. UV wavelength-dependent DNA damage and human non-melanoma and melanoma skin cancer. Photochem Photobiol 2008; 11: 90. 31. Jung EG, Furtwängler M, Klostermann G et al. Light-induced skin cancer and prolonged uv-erythema. Arch. Dermatol. Res. 1980; 267: 33–36. 32. Aubin F, Zultak M, Blanc D, et al. Reaction to UV-induced erythema in young patients with basal cell carcinoma. Photodermatol. 1989; 6: 118–123. 33. Holmes TCA & Holmes RMM. Clearance of UVB-induced erythema in patients with non-melanoma skin cancer. Photodermatol. Photoimmunol. Photomed 1997; 13:189–192. 34. Rhodes LE, Webb AR, Fraser HI et al. Recommended summer sunlight exposure levels can produce sufficient (≥20 ng ml-1) but not the proposed optimal (≥32 ng ml-1) 25(OH)D levels at UK latitudes. J Invest Dermatol. 2010;130:1411-8. 35. Webb AR, Engelsen O. Calculated ultraviolet exposure levels for a healthy vitamin D status. Photochem Photobiol. 2006;82:1697-703. 36. Diffey BL, Jansen CT, Urbach F et al. The standard erythema dose: a new photobiological concept. Photodermatol Photoimmunol Photomed 1997;13:64–6. 37. Taylor DK, Anstey AV, Coleman AJ et al. Guidelines for dosimetry and calibration in ultraviolet radiation therapy: a report of a British Photodermatology Group workshop. Br J Dermatol 2002; 146:755–63. 38. Webb AR, Kift R, Berry JL et al. The vitamin D debate: translating controlled experiments into reality for human sun exposure times. Photochem Photobiol 2011; 87:741–5. 39. Robertson AR. The CIE 1976 color difference formulas. Color Res Appl. 1977; 2: 7-11. 40. Chardon A, Cretois I, Hourseau C. Skin colour typology and suntanning 153 pathways. Int J Cosmet Sci. 1991; 13:191-208. 41. Del Bino S, Sok J, Bessac E, et al. Relationship between skin response to ultraviolet exposure and skin color type. Pigment Cell Res. 2006; 19:606-14. 42. NRPB (2002). Health Effects from Ultraviolet Radiation: Report of an Advisory Group on Non-Ionising Radiation. Documents of the NRPB 13: 207-17. 43. Takeuchi H, Gomi T, Shishido M, et al. Neutrophil elastase contributes to extracellular matrix damage induced by chronic low-dose UV irradiation in a hairless mouse photoaging model. J Dermatol Sci. 2010;60:151-8. 44. Rijken F, Kiekens RC, Bruijnzeel PL. Skin-infiltrating neutrophils following exposure to solar-simulated radiation could play an important role in photoageing of human skin. Br J Dermatol. 2005;152:321-8. 45. Nathan C. Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol. 2006;6:173-82. 46.Kaidbey KH, Agin PP, Sayre RM, et al. Photoprotection by melanin--a comparison of black and Caucasian skin. J Am Acad Dermatol. 1979;1:249-60. 47. Farrar MD, Kift R, Felton SJ, et al. Recommended summer sunlight exposure amounts fail to produce sufficient vitamin D status in UK adults of South Asian origin. Am J Clin Nutr. 2011;94:1219-24. 48. Felton SJ, Cooke MS, Kift R, et al. Concurrent beneficial (vitamin D production) and hazardous (cutaneous DNA damage) impact of repeated low-level summer sunlight exposures. Br J Dermatol. 2016; doi 10.1111/bjd.14863. 49. Ouhtit A, Muller HK, Davis DW, Ullrich SE, Mcconkey D, Ananthaswamy HN. Temporal events in skin injury and the early adaptive responses in ultraviolet- irradiated mouse skin. Am J Pathol. 2000;156(1):201-7. 50. Holleran WM, Uchida Y, Halkier-sorensen L, et al. Structural and biochemical basis for the UVB-induced alterations in epidermal barrier function. 154 Photodermatol Photoimmunol Photomed. 1997;13(4):117-28. 51. Lopez H, Beer JZ, Miller SA, Zmudzka BZ. Ultrasound measurements of skin thickness after UV exposure: a feasibility study. J Photochem Photobiol B, Biol. 2004;73:123-32. 155 FIGURE LEGENDS Figure 1: Weekly skin colour measurements taken from buttock skin during the simulated summer’s sunlight exposure ITA represents skin colour on the L*-b* plane, based on spectrophotometer readings taken from buttock skin. The vertical axis is the skin lightness (L*) and the horizontal axis the yellow-blue component (b*). (a) L* and (b) ITA show significant darkening over the simulated summer sunlight exposures, from 69.37 (2.8) to 65.52 (2.33) and from 52 (5.7)° at baseline to 41 (6.4)°at course-end for the 10 volunteers (p<0.001; p<0.0001). ITA represents skin colour on the L*-b* plane, based on spectrophotometer readings taken from buttock skin. The vertical axis is the skin lightness (L*) and the horizontal axis the yellow-blue component (b*). ITA values were higher at baseline (open circles) representing ‘very light’ skin colour and were significantly decreased (closed circles) by the simulated summer‘s UVR exposures to ‘light’ or ‘intermediate’ colour. Short horizontal bars in (a) denote the mean. * p<0.0001. Figure 2: Effect of UVR exposure on epidermal thickness. (a) The 6-week course of simulated summer sunlight exposures significantly thickened the stratum corneum, from mean 29.3 (9.59) to 41.5 (12.7)µm (p<0.05). Acute 2X UVB MED exposure did not provoke measurable change to stratum corneum thickness. (b) Simulated summer sunlight exposures apparently increased mean viable epidermal thickness from 39.1 (6.6) to 43.7 (8.1)µm (p>0.05). Acute 2X MED UVB exposure apparently increased viable epidermal thickness from 43.7 (8.1) to 48.6 (6.4)µm at 24 hr post-challenge (p=0.05). Markers represent mean readings from each participant as measured by Pannoramic viewer (n=8 for photoprotected skin due to lack of epidermis in 2 156 individuals, n=10 for the remainder). Horizontal bars denote the overall mean. Thickening of the stratum corneum and the viable epidermis (black arrows) is seen between photoprotected skin (c) and that following the simulated summer sunlight (d). Original magnification x20. * p<0.05. Figure 3: Cutaneous erythema following acute 2X MED UVB-challenge Erythema induced by 2X MED challenge from all participants (n=10) was greater in previously photoprotected skin, as demonstrated by (a) a* from L*a*b* scores (Konica Minolta spectrophotometer) and (b) erythema index (EI; Diastron reflectance instrument). To obtain background-corrected data, induced erythema was calculated by subtracting a* or EI readings taken on proximal, untreated buttock skin from those taken on 2X MED UVB-challenged buttock skin. Horizontal bars denote mean values. * p<0.05 Figure 4: Neutrophil infiltration under varying conditions of UVR exposure A-D: Neutrophil elastase NP57 staining (Dako, Denmark; black arrows), original magnification x20 in (a) Unexposed skin (b) Immediately following the last 1.3 SED exposure of the 6-weeks’ simulated summer sunlight exposures (c) 24h after 2X MED UVB exposure to skin that was photoprotected during the simulated summer’s sunlight, (d) 24h after 2X MED UVB exposure to skin that was previously exposed to the simulated summer’s sunlight exposures (e) Neutrophil counts per mm2 dermis. Neutrophils were absent in unexposed skin and skin that had been exposed to the simulated summer’s sunlight. 24h following acute 2X MED UVB- challenge, mean counts were significantly higher in previously unexposed skin, at 73.5 (72.5)/mm2 than in skin that had received the simulated summer’s sunlight (21.4 [24.1]/mm2; p<0.05). n=8 for counts in photoprotected skin due to lack of 157 epidermis/dermis in 2 individuals, n=10 for the remainder. Horizontal bars denote the mean. * p<0.05. 158 Figure 1 a b 159 Figure 2 a b c d 160 Figure 3 a b 161 Figure 4 e 162 CHAPTER 5: Manuscript 3 163 Serum endocannabinoids and N-acyl ethanolamines and the influence of simulated solar UVR exposure in humans in vivo Short title: Impact of UVR on human serum endocannabinoids Sarah J. Felton,1 Alexandra C. Kendall,2 Abdalla F.M. Almaedani,2 Paula Urquhart,2 Ann R Webb,3 R. Kift,3 Andy Vail,4 Anna Nicolaou,2 Lesley E. Rhodes1 1. Division of Musculoskeletal and Dermatological Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester Academic Health Science Centre, Salford Royal NHS Foundation Trust, Manchester, UK. 2. Division of Pharmacy and Optometry, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, UK. 3. School of Earth Atmospheric and Environmental Sciences, Faculty of Science and Engineering, The University of Manchester, Manchester, UK 4. Division of Population Health, Health Services Research & Primary Care, School of Health Sciences, University of Manchester, Manchester Academic Health Science Centre, Salford Royal NHS Foundation Trust, Manchester, UK. Word count: 3119 Tables: 3 Figures: 4 Key words: endocannabinoids, fatty acids, skin, sunlight, ultraviolet radiation Correspondence to: Prof L E Rhodes, Photobiology Unit, Dermatology Centre, University of Manchester, Salford Royal Hospital, Manchester, M6 8HD, UK Tel 00 44 161 1150, Email: [email protected] 164 Abstract Solar ultraviolet radiation (UVR) exposure of human skin has beneficial and harmful effects on health, including impact on immune function, inflammation and reportedly mood, but these are not fully elucidated. Since the endocannabinoid system is implicated in many activities including mood alteration, our objective was to (i) determine and quantify circulating levels of a wide range of endocannabinoid and N-acyl ethanolamine (NAE) species (ii) evaluate whether these are modulated by cutaneous UVR exposures, as attained through repeated low level summer sunlight exposure. Wearing goggles to prevent eye exposure, 16 healthy volunteers (23-59y; 10 light skin, phototype II, and 6 dark skin, phototype V) received the same UVR exposures (1.3 SED, 95% UVA/5% UVB) thrice weekly for 6 weeks, whilst casually dressed to expose ~35% skin surface area. Blood samples were taken at baseline, days 1, 3 and 5 of week one, then at weekly intervals, and analysed by LC-MS/MS. Eleven endocannabinoids and NAEs were detected and quantified at baseline, with N-palmitoyl ethanolamine the most abundant (30% of total). Levels did not vary according to phototype (p>0.05), except for the NAE docosapentaenoyl ethanolamide, which was higher in phototype II than V (p=0.0002). Level of the endocannabinoid, 2-AG, was elevated during the UVR exposure course (p<0.05 vs baseline for all subjects; p<0.01 for each phototype group), with maximum levels reached by week 2-3, while NAE species did not significantly alter. These findings suggest differential involvement of the cutaneous endocannabinoid system in low dose solar UVR responses in humans. 165 Abbreviations 2-AG 2-Arachidonoyl glycerol AEA N-Arachidonoyl ethanolamine (anandamide) CB Cannabinoid receptor DGLEA N-Dihomo-γ-linolenoyl ethanolamine DHEA N-Docosahexaenoyl ethanolamine DPEA N-Docosapentaenoyl ethanolamine EPEA N-Eicosapentaenoyl ethanolamine LC-MS/MS Liquid chromatography coupled to tandem mass spectrometry LEA N-Linoleoyl ethanolamine MED Minimal erythemal dose MEA N-Myristoyl ethanolamine NAE N-Acyl ethanolamine OEA N-Oleoyl ethanolamine PEA N-Palmitoyl ethanolamine STEA N-Stearoyl ethanolamine UVR Ultraviolet radiation 166 INTRODUCTION Solar ultraviolet radiation (UVR) exposure of the skin has a range of beneficial but also harmful effects on health, with vitamin D synthesis, sunburn, skin cancer induction, photosensitivity and photoageing being well documented,1 while further impacts including production of other hormones and modulation of immune and inflammatory status are less well elucidated. UVR is pro- inflammatory and immunomodulatory, reducing cell-mediated immunity while augmenting innate responses, and in predisposed individuals activates the Herpes simplex virus. It is also observed that sunlight exposure causes a ‘feel-good factor’ or euphoria, which could be mediated by UVR.2-4 Mood enhancement is observed in indoor tanning, where skin is exposed to the UVR component of sunlight alone; many individuals continue to self-expose despite knowledge of the adverse consequences, leading to the term ‘tanorexia’ or addictive-like tanning behaviour.2,3,5 Whilst this phenomenon was previously suspected to be attributable to circulating endorphins, akin to mood enhancement after intense exercise,6 β-endorphin is unable to cross the blood-brain barrier7 and cannot itself have central effects when produced peripherally. Moreover, investigations for a role of endorphins in tanorexia proved inconsistent.8-10 Recently, the endocannabinoid system has been implicated in ‘runner’s high,’ with increased circulating levels of anandamide (AEA), which can cross the blood-brain barrier, detected after intense aerobic exercise.11-13 Recent studies evidence the extensive cutaneous profile of lipid mediators in human skin, encompassing the endocannabinoids, NAE, sphingolipids and eicosanoid families.14-18 Some of these lipid species are known to be modulated by UVR and to play a role in photobiological effects in humans19, 20 whilst the potential involvement of the endocannabinoids and their congeners in UVR-induced effects, 167 awaits further exploration. The main endocannabinoids anandamide (AEA) and 2- arachidonoyl glycerol (2-AG), and a range of N-acyl ethanolamine (NAE) species, derive from membrane lipids (Fig 1).21-23 AEA, 2-AG and some NAE species are physiological ligands for the G-protein coupled cannabinoid (CB) receptors, originally identified as the target for biologically active components of the cannabis plant.24-26 They are active in neurotransmission in the central and peripheral nervous systems, including reduction in pain perception via CB1 and transient receptor potential vanilloid-1 receptors (TRV-1),27 and show anti- inflammatory/immune-modulatory effects via peripheral CB2 receptors, including 28 reduced IFN-δ and increased IL-10 secretion. Although CB1 receptors were traditionally described in the central nervous system and CB2 peripherally as in immune cells, it has become evident their distribution is more variable and widespread throughout organ systems, including skin18,29-30 which has been shown 14 to possess a functional endocannabinoid system. CB1 and CB2 receptors are expressed by keratinocytes and melanocytes, and also identified in sebocytes and hair follicles. Recent evidence also suggests that the endocannabinoid system helps skin maintain homeostasis and respond to UVR challenge, with CB1/CB2-deficient mice experiencing increased allergic contact dermatitis31 and cutaneous carcinogenesis.32 Despite increased interest in roles of endocannabinoids in human physiology and disease, including mood, information on individual mediators and their responses to cutaneous UVR exposure is sparse. Skin is a large organ that may substantially contribute to circulating endocannabinoids and NAEs; this may have consequences for mood, immune, inflammatory and other functions. In this study, we used a UVR protocol (including UVR emission, dose and skin site) mimicking a summer’s repeated low-level, sunlight exposures, to examine 168 potential influence on circulating endocannabinoids and NAEs in daily life, with particular interest in AEA and 2-AG. Detection and quantification of a wide range of circulating species, was by LC-MS/MS. Different phototypes were included as melanisation may affect UVR responses. Our aims were to assess the range and quantity of endocannabinoids and related NAE in human sera and to examine their responses to multiple low-level UVR exposures, as could be experienced incidentally in summer. 169 MATERIALS AND METHODS Study subjects and design Healthy volunteers were recruited (January 2010). Exclusion criteria were pregnancy, breastfeeding, taking photoactive medication or supplements that contained vitamin D, a history of skin cancer or a photosensitivity disorder, and use of a sunbed or sunbathing in the 3 months prior to or during the study. Body mass index (in kg/m2) was calculated as weight/height2. Ethical approval was obtained from the North Manchester Research Ethics Committee (reference 09/H1014/73). Informed consent was obtained and the study adhered to the principles of the Declaration of Helsinki. Participants proceeded through the study process as outlined in the protocol overview (Fig 2). Participants were white Caucasians of Fitzpatrick33 sun-reactive skin type II (i.e. usually burns, sometimes tans) or of South Asian ethnicity with skin type V (brown skin). Minimal erythemal dose (MED) assessment The MED, defined as the lowest dose of UVR that produced a visually discernable erythema at 24 hours, was assessed in each subject prior to the exposure course, as a precaution. A geometric series of 10 doses (7–80mJ/cm2 for phototype II; 26.6–271mJ/cm2 for phototype V) of erythemally weighted UVR was applied over 2 horizontal rows of buttock skin with a Waldmann UV 236B unit containing Waldmann CF-L 36W/UV6 lamps (peak emission: 313nm; range: 290–400nm; Waldmann GmbH, Villinge-Schwenningen, Germany). Simulated summer sunlight UVR exposures Volunteers were given a six-week course of UVR exposures 3x weekly (Monday, 170 Wednesday and Friday at approximately the same time of day), concordant with the length of the summer school holiday period when the population is most exposed to sunlight, as previously described.34 They wore opaque UVR-blocking eye protection goggles (4-eyez, Scottsdale, AZ, USA), and standardised T-shirts and knee-length shorts to expose approximately 35% skin surface area. A Philips HB588 whole body horizontal irradiation cabinet (Eindhoven, The Netherlands) fitted with Arimed B (Cosmedico GmbH, Stuttgart, Germany) and Cleo Natural (Philips, Eindhoven, The Netherlands) fluorescent tubes provided an UVR emission close to summer sunlight (95% UVA: 320–400nm, 5% UVB: 290–320nm), which was characterised and monitored by spectroradiometry, as described.34 The course of simulated solar UVR was given in wintertime, with a low dose UVR exposure of 1.3 standard erythemal dose35 at every visit. The time to deliver this dose was 6.5 minutes; a constant UVR dose was maintained throughout the study by adjusting for decrease in irradiance by increasing delivery time. Using radiative- transfer modeling to translate this to real-life exposures, this equates to ~13-17 minutes of unshaded sunlight exposure on a clear June midday in Manchester, UK (53.5N) 6x weekly, which takes account of (i) ventral and dorsal surfaces are not irradiated simultaneously in sunlight and (ii) postures may range from the horizontal to the vertical randomly orientated to the sun.36 Endocannabinoid and NAE analysis Blood samples were taken pre-UVR exposures on Monday, Wednesday and Friday of the first week of irradiation at the same time of day to within 60 minutes on each occasion, to look for any acute changes in levels, and each subsequent Monday until course-end (i.e. 3 days after last irradiation of the week) to identify any cumulative effects, and serum was stored at −20°C until study completion. 171 Samples were defrosted on ice and 3 ml of ice-cold 2:1 (v/v) chloroform/methanol added. Anandamide-d8 (20ng/sample) and 2-arachidonoyl glycerol-d8 (40ng/sample) (Cayman Chemicals, Ann Arbor, MI, USA) were added as internal standards. Samples were mixed and incubated on ice for 30min. 500µl of water was added to each sample before centrifugation (5000rpm, 4°C, 5min). The organic phase was dried under a steam of nitrogen and the lipid extract reconstituted in 100µl HPLC-grade ethanol, and stored at -20°C awaiting LC-MS/MS analysis. LC-MS/MS was performed on a UPLC pump (Acquity, Waters) coupled to an electrospray ionisation triple quadrupole mass spectrometer (TQ-S, Waters). Analytes were separated on a C18 column (Acquity UPLC ® BEH Phenyl C18, 1.7µm, 21 x 5mm; Waters) using a gradient of solvent A (water:acetic acid; 99.98:0.02; v/v) and solvent B (acetonitrile:acetic acid; 99.98:0.02; v/v) as follows: 22-28 % B (0-3 min), 28-55% B (3-3-1min), 55-80% B (3.1-11min), 80% B (11- 12.5 min), 80-22% B (12.5-12.51min) and 22% B (12.51-15min), at a flow rate of 0.6ml/min. The instrument was operated in the positive ionisation mode and, for all compounds, the MS/MS settings were as follows: capillary voltage 1800V, source temperature 100°C, desolvation temperature 400°C, dwell time 0.025s. Mass LynxTM V 4.1 was used as operating software to control the instrument and acquire data. Calibration lines using commercially available standards (Caymen Chemicals, USA) were generated to cover a range of 1-20pg/µl, which showed a linear response and samples were analysed within this range prior to normalisation against volume. The limit of detection for each compound was <0.16pg on the column. Outcome measures 172 Primary outcome measures were baseline levels of serum endocannabinoids and NAE, and their changes during the simulated summer UVR exposures of the skin. Comparisons were additionally made between white Caucasian and south Asian individuals. Statistical analyses Data analyses, specifically paired and unpaired t-tests, linear regression and repeated measures ANOVAs with Greenhouse-Geisser corrections and Bonferroni post-hoc tests, were performed using SPSS statistical software (version 21.0.0; SPSS Inc., Chicago, IL, USA) and GraphPad Prism (version 6; GraphPad Software, La Jolla, CA, USA). Serum concentrations were logarithmically transformed to make them normally distributed. Results were considered statistically significant if p<0.05. 173 RESULTS Volunteer characteristics Of the 18 recruited subjects, two of the eight South Asians dropped out early for personal reasons unrelated to the study; their results were not analysed. Table 1 displays baseline characteristics of participants. The presence of AEA, 2-AG and NAE in human serum at baseline AEA, 2-AG and nine NAE species were detected and quantified in human serum. These comprised myristoyl-ethanolamine (MEA), N-palmitoyl ethanolamine (PEA), N-linoleoyl ethanolamine (LEA), N-oleoyl ethanolamine (OEA), N-stearoyl ethanolamine (STEA), N-eicosapentaenoyl ethanolamine (EPEA), N-dihomo-γ- linolenoyl ethanolamine (DGLEA), N-docosahexaenoyl ethanolamine (DHEA) and N-docosapentaenoyl ethanolamine (DPEA). Endocannabinoid and NAE species’ concentrations varied widely at baseline. Prior to UVR exposure, median serum AEA concentration for all subjects was 318.6 (range 62.5 to 636.0)pg/ml and 2-AG was 1018.0 (312.6 to 5025.0)pg/ml. PEA was the most abundant NAE quantified (median 2824.0 [range 2282.6 to 4506.7]pg/ml) followed by LEA and OEA (median values around 1000pg/ml), then STEA and DHEA, (median values around 600pg/ml), while EPEA and DGLEA were undetectable in some individuals or at values <100pg/ml when present (Table 2). Figure 2 displays baseline levels of these eleven compounds. When participants were analysed according to their skin type (II or V), baseline serum endocannabinoid and NAE levels were not statistically different between 174 the two groups, apart from DPEA, which was higher in the phototype II cohort at 73.6pg/ml (61.1 to 103.3pg/ml) than the phototype V cohort (median 39.8 [30.4 to 60.2]pg/ml); p=0.0002 (Tables 2, 3). Changes in serum endocannabinoids and NAE over the first week of UVR- exposures Serum samples collected prior to cutaneous UVR exposures on Monday, Wednesday and Friday during the first week of the study showed variation in 2- AG, the median value for all subjects apparently increasing from 1018.0 [312.6 to 5025.0]pg/ml at baseline to 1713.0 [637.6 to 9039.3pg/ml] following two irradiations although this did not reach statistical significance (repeated measures ANOVA, p=0.067; Fig 3A). No changes in serum levels of AEA or NAE species were detected over week one for all participants combined (p>0.05). Similarly, when participants were analysed according to their skin type, levels did not vary significantly between the two groups, (p>0.05 for all). Changes in serum endocannabinoids and NAE species over the six weeks’ repeated UVR exposures Serum 2-AG concentration for all subjects increased significantly over the six weeks of simulated summer sunlight exposures (one-way repeated measures ANOVA, p<0.05). Levels reached a peak around week 4 with a median value of 1704.0 pg/ml (range 300.1 to 4850.6pg/ml) before returning towards baseline (median 1157. 7 [275.1 to 2283.9]pg/ml, Fig 3B). No relationship was seen between either baseline AEA or change in serum AEA concentration over the six weeks’ simulated summer sunlight and body mass index (data not shown). The 175 remaining NAE were unaffected by the repeated, low-level UVR exposures (p>0.05 for all). Analysis of the endocannabinoids and NAE species over the six-weeks’ irradiation for both skin type groups showed only 2-AG to vary significantly, reaching a maximum of 1609.4 (range 587.6 to 4246.3)pg/ml at week 3 in phototype II and 2257.3 (range 319.8 to 4850.6)pg/ml at week 4 in phototype V (two-way repeated measures ANOVA, p<0.01; Table 3; Fig 4). 176 DISCUSSION This study makes a novel examination of healthy human in vivo endocannabinoid and NAE responses to cutaneous UVR exposures that simulate incidental summer sunlight exposures, in people of light and dark skin types. The protocol, with UVR emission close to summer sunlight (95% UVA, 5% UVB), subjects wearing informal clothing (T-shirt and shorts) to expose only commonly -exposed skin sites, and brief as opposed to prolonged times, reflects the exposures occurring in everyday life rather than deliberate sunbathing. The UVR doses were equivalent to ~15 minutes June midday exposure (53.5°N), gained on most days of the week.36 Circulating level of 2-AG, a NAE, significantly increased during the UVR exposure course, thus implicating an in vivo role for UVR modulation of the skin endocannabinoid system even at these low doses. In view of the reported activities of 2-AG, health implications may include mood alteration and wider aspects such as UVR-induced inflammation and immunomodulation. At baseline we detected the endocannabinoids (AEA and 2-AG) and nine NAE species (MEA, PEA, LEA, OEA, STEA, EPEA, DGLEA, DHEA and DPEA) in human sera (Tables 3A, 3B). All major organ tissues may be contributing to these levels, including brain, liver and skin. Studies examining circulating levels in healthy humans are scarce. We found that prior to UVR exposure, median serum levels for most species were similar to recently reported values for healthy subjects,37-39 including for AEA and 2-AG, while PEA was the most abundant NAE quantified. LEA, OEA and DHEA showed similar levels, while EPEA, DGLEA and DPEA had the lowest serum concentrations. However, STEA showed a median concentration of 697.1pg/ml, contrasting with ~6000ng/ml reported for “healthy controls” by Pavon et al.38 The reason for the difference is unknown, although in the Pavon et al study it is not clear if those taking regular medications affecting 177 fatty acid metabolism were excluded, and 17% subjects had received psychiatric treatment.38 Interestingly, our ethnically different subject groups showed a DPEA level that was significantly higher in white Caucasians than South Asians. This may be attributable to the higher omega-3 fatty acid (as found in fish oil) dietary intake observed in white Caucasians than south Asians,40,41 as consumption of the omega- 3 fatty acid eicosapentaenoic acid (EPA) leads to increased DPEA. We discovered that circulating concentration of 2-AG was significantly raised during the course of UVR treatments (p<0.05), with the highest levels overall (median 1704pg/ml) achieved after 3 weeks, and no significant changes observed in other fatty acids after these low-level simulated summer exposures. A lack of further increase in 2-AG levels after 3-4 weeks of 3x weekly exposures could imply saturation of endocannabinoid biosynthesizing enzymes, depletion of their precursors, saturation of CB receptors and/or photoadaptation. Despite the sub-erythemal doses being fixed, i.e. 1.3 SED rather than individually MED-related, the same response was seen in both phototype II and V subjects, i.e. it occurred regardless of skin pigmentation. The differential increase in 2-AG and not other species may relate to their different biosynthetic pathways (Fig 1).42 Since UVR influences lipolytic enzymes including phospholipase C (PLC),43 modifications could include increased release of diacylglycerol (DAG) from membrane phospholipids,44 resulting in increased availability of DAG as 2-AG substrate. Indeed UVR exposure to keratinocyte cultures has been shown to increase endogenous DAG production.43 Additionally, the catabolising enzymes FAAH and MAG lipase, found in many tissues, may reduce the concentration of metabolites produced post low-dose UVR, limiting the detection particularly of metabolites present at lower concentration than 2-AG.45 178 Potentially, further UVR effects on endocannabinoids/NAEs might be observed with an increasing UVR-dosing schedule (more hazardous to skin), as can be found with deliberate sunbathing, indoor tanning or phototherapy regimes. A recent exploratory study by our group examined cutaneous endocannabinoid and NAE levels in skin biopsies taken 24 hours after UVR exposure to a localised area of the buttock, and found no alteration.46 However, that study employed only a single exposure of 2xMED of principally UVB (275-380, peak 305nm) implying that repeated UVR exposures may be necessary for endocannabinoid and NAE responses. In-keeping with this hypothesis, Magina et al detected changes in plasma endocannabinoids after six weeks of whole-body narrowband UVB (311nm) therapy.47 However, in contrast to our results, Magina et al report a decrease in AEA with 2-AG remaining unchanged. Potential reasons for these differences include their escalating UVR-dose (from 0.3 to 2 J/cm2), the very different UVR emission employed, and their study population being psoriasis patients. Levels of endocannabinoids may be influenced by skin conditions including psoriasis and cutaneous itching, in addition to comorbidities of diabetes and hypertension48-50 that are prevalent in psoriasis,52 thus confounding observations compared with healthy volunteer studies. Implications of our study may include involvement of sunlight in mood control via the endocannabinoid system. Support for endocannabinoid activity in mood control includes studies in rats where depressive models had reduced AEA 53 levels and differential changes in CB1 receptor binding density in the brain; moreover, activation of the endocannabinoid system had anti-depressive effects 54 mediated through CB1 receptors. Human studies demonstrated reduced serum 2- AG and AEA levels in patients with untreated depression compared to controls.55-56 The endocannabinoids are hypothesised to activate CB1 receptors on brain GABA- 179 ergic neurons, thereby increasing dopamine release in central reward centres,57 13 while the neutrophin brain-derived neurotrophic factor, and peripheral CB1 and 58 CB2 receptor activation may be involved. Responses of the skin endocannabinoid system might also contribute to mediation of UVR-induced skin inflammation, possibly mediated via alterations in arachidonic acid and prostaglandin levels,42 and immunomodulation, including of cell-mediated immunity.59 In mouse studies, genetic deletion or pharmacologic blockade of keratinocyte CB1 and CB2 receptors enhances allergic contact dermatitis,31 potentially mediated through endocannabinoid regulation of monocyte chemotactic protein 2 (MCP- 2)/chemokine ligand 8 (CCL8) expression.60 Strengths of the study include examination of a range of serum endocannabinoids and NAEs in healthy human volunteers in vivo, and assessment of their responses to carefully performed low-level simulated summer solar UVR exposures, with UVA/UVB emission close to midday sunlight, and exposures whilst wearing casual clothing, as it cannot be assumed that responses of normally unexposed skin are the same as routinely exposed sites. Protective goggles were worn throughout exposures, eliminating possible impact on endocannabinoid or NAE levels of transmission through the eyes. Invasive (skin biopsy) assessment to quantify endocannabinoids/NAEs directly in the skin following multiple UVR exposures, and assessment of health outcome measures, were not performed but would be appropriate for future studies. In addition to incidental exposures seen in everyday life in summertime, the impact of deliberate sunbathing on endocannabinoids/NAEs in healthy individuals could be insightful to explore. In summary, repeated low-dose simulated sunlight exposure, as may be gained incidentally in summer-time, is associated with activation of the endocannabinoid system with elevation in serum 2-AG. This may contribute to 180 health effects of UVR exposure of human skin, including influence on mood, inflammation and immunity, and warrants further study. Acknowledgements The authors acknowledge the assistance of Marie Durkin in volunteer recruitment. 181 References 1. NRPB (2002). Health Effects from Ultraviolet Radiation: Report of an Advisory Group on Non-Ionising Radiation. Documents of the NRPB 13: 207-17. 2. Robinson JK, Kim J, Rosenbaum S, Ortiz S. Indoor tanning knowledge, attitudes, and behavior among young adults from 1988-2007. Arch Dermatol 2008;144:484-8. 3. Harrington CR, Beswick TC, Leitenberger J, et al. Addictive-like behaviours to ultraviolet light among frequent indoor tanners. Clin Exp Dermatol 2011;36:33-8. 4. Schneider S, Diehl K, Bock C, et al. Sunbed use, user characteristics, and motivations for tanning: results from the German population-based SUN-Study 2012. JAMA Dermatol 149:43-9. 5. Felton S, Adinoff B, Jeon-slaughter H, Jacobe H. The significant health threat from tanning bed use as a self-treatment for psoriasis. J Am Acad Dermatol. 2016;74(5):1015-7. 6. Carr DB, Bullen BA, Skinar GS, et al. Physical conditioning facilitates the exercise-induced secretion of beta-endorphin and beta-lipotropin in women. N Engl J Med 1981; 305: 560-3. 7. Houghten RA, Swann RW, Li CH. beta-Endorphin: stability, clearance behavior, and entry into the central nervous system after intravenous injection of the tritiated peptide in rats and rabbits. Proc Natl Acad Sci USA. 1980;77:4588-91. 8. Wintzen M, Ostijn DM, Polderman MC, et al. Total body exposure to ultraviolet radiation does not influence plasma levels of immunoreactive beta- endorphin in man. Photodermatol Photoimmunol Photomed 2001;17:256-60. 9. Gambichler T, Bader A, Vojvodic, M et al. Plasma levels of opioid peptides after sunbed exposures. Br J Dermatol 2002;147:1207-11. 182 10. Fell GL, Robinson KC, Mao J, Woolf CJ, Fisher DE. Skin β-endorphin mediates addiction to UV light. Cell. 2014;157:1527-34. 11. Sparling PB, Giuffrida A, Piomelli D, Rosskopf L, Dietrich A. Exercise activates the endocannabinoid system. Neuroreport. 2003;14:2209-11. 12. Raichlen DA, Foster AD, Gerdeman GL, Seillier A, Giuffrida A. Wired to run: exercise-induced endocannabinoid signaling in humans and cursorial mammals with implications for the 'runner's high'. J Exp Biol. 2012;215(Pt 8):1331-6. 13. Heyman E, Gamelin FX, Goekint M et al. Intense exercise increases circulating endocannabinoid and BDNF levels in humans-possible implications for reward and depression. Psychoneuroendocrinology. 2012;37:844-51. 14. Bíró T, Tóth B, Haskó G, Paus R, Pacher P. The endocannabinoid system of the skin in health and disease: novel perspectives and therapeutic opportunities. Trends Pharmacol Sci 2009;30:411-420. 15. van Smeden J, Hoppel L, van der Heijden R et al. LC/MS analysis of stratum corneum lipids: ceramide profiling and discovery. J Lipid Res 2011; 52: 1211–21. 16. Nicolaou A. Eicosanoids in skin inflammation. Prost Leuk Essen Fatty Acids 2013;88: 131-138. 17. Breiden B, Sandhoff K. The role of sphingolipid metabolism in cutaneous permeability barrier formation. Biochim Biophys Acta 2014; 1841:441–52. 18. Kendall AC, Pilkington SM, Massey, KA Sassano G, Rhodes LE, A Nicolaou. Distribution of bioactive lipid mediators in human skin. J Invest Dermatol. 2015;135:1510-20. 19. Pilkington SM, Rhodes LE, Al-Aasswad NM et al. Impact of EPA ingestion on COX- and LOX-mediated eicosanoid synthesis in skin with and without a pro- 183 inflammatory UVR challenge − report of a randomised controlled study in humans. Mol Nutr Food Res 2014; 58: 580-90. 20. Kendall AC, Nicolaou A. Bioactive lipid mediators in skin inflammation and immunity. Prog Lipid Res 2013; 52:141–64. 21. Mechoulam R, Fride E, Di Marzo V. Endocannabinoids. Eur J Pharmacol. 1998;359:1-18. 22. Mechoulam R, Hanus L. A historical overview of chemical research on cannabinoids. Chem Phys Lipids. 2000;108:1-13. 23. Bisogno T. Endogenous cannabinoids: structure and metabolism. J Neuroendocrinol. 2008;20 Suppl 1:1-9. 24. Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature. 1990;346:561-4. 25. Munro S, Thomas KL, Abu-shaar M. Molecular characterization of a peripheral receptor for cannabinoids. Nature. 1993;365:61-65. 26. Howlett AC. The cannabinoid receptors. Prostaglandins Other Lipid Mediat 2002:68-69:619-631. 27. Guindon J, Hohmann AG. The endocannabinoid system and pain. CNS Neurol Disord Drug Targets. 2009;8:403-21. 28. Boorman E, Zajkowska Z, Ahmed R, Pariante CM, Zunszain PA. Crosstalk between endocannabinoid and immune systems: a potential dysregulation in depression? Psychopharmacology 2016;233:1591-604. 29. Hansen HS, Artmann A. Endocannabinoids and nutrition. J Neuroendocrinol 2008;20:94-9. 184 30. Ghafouri N, Ghafouri B, Turkina MV, et al. High levels of N- palmitoylethanolamide and N-stearoylethanolamide in microdialysate from myalgic trapezius muscle in women. PLoS ONE 2011;6:e27257. 31. Karsak M, Gaffal E, Date R, et al. Attenuation of allergic contact dermatitis through the endocannabinoid system. Science. 2007;316:1494-7. 32. Zheng D, Bode AM, Zhao Q, Cho YY, Ma WY, Dong Z. The cannabinoid receptors are required for ultraviolet-induced inflammation and skin cancer development. Cancer Res. 2008;68:3992-8. 33. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988;124:869-71. 34. Rhodes LE, Webb AR, Fraser HI et al. Recommended summer sunlight exposure levels can produce sufficient (> or =20 ng ml(-1)) but not the proposed optimal (> or =32 ng ml(-1)) 25(OH)D levels at UK latitudes. J Invest Dermatol. 2010;130:1411-8. 35. Diffey BL, Jansén CT, Urbach F, Wulf HC. The standard erythema dose: a new photobiological concept. Photodermatol Photoimmunol Photomed. 1997;13:64-6. 36. Webb AR, Kift R, Berry JL, Rhodes LE. The vitamin D debate: translating controlled experiments into reality for human sun exposure times. Photochem Photobiol 2011;87:741–5. 37. Feuerecker M, Hauer D, Gresset T et al. Effect of an acute consumption of a moderate amount of ethanol on plasma endocannabinoid levels in humans. Alcohol Alcohol. 2012;47:226-32. 38. Pavón FJ, Araos P, Pastor A, et al. Evaluation of plasma-free endocannabinoids and their congeners in abstinent cocaine addicts seeking outpatient treatment: impact of psychiatric co-morbidity. Addict Biol. 185 2013;18:955-69. 39. Jones PJ, Lin L, Gillingham LG, Yang H, Omar JM. Modulation of plasma N- acylethanolamine levels and physiological parameters by dietary fatty acid composition in humans. J Lipid Res. 2014;55:2655-64. 40. Lovegrove JA, Lovegrove SS, et al. Moderate fish-oil supplementation reverses low-platelet, long-chain n-3 polyunsaturated fatty acid status and reduced plasma triacylglyceral concentration in British Indo-Asians. Am J Clin Nutr. 2004;79:974-82. 41. Manav M, Su J, Hughes K, Lee HP, Ong CN. Omega-3 fatty acids and selenium as coronary heart disease risk modifying factors in Asian Indian and Chinese males. Nutrition. 2004;20:967-73. 42. Turcotte C, Chouinard F, Lefebvre JS, Flamand N. Regulation of inflammation by cannabinoids, the endocannabinoids 2-arachidonoyl-glycerol and arachidonoyl-ethanolamide, and their metabolites. J Leukoc Biol. 2015;97(6):1049- 70. 43. Carsberg CJ, Ohanian J, Friedmann PS. Ultraviolet radiation stimulates a biphasic pattern of 1,2-diacylglycerol formation in cultured human melanocytes and keratinocytes by activation of phospholipases C and D. Biochem J. 1995;305(Pt 2):471-7. 44. Hall H, Williams EJ, Moore SE, Walsh FS, Prochiantz A, Doherty P. Inhibition of FGF-stimulated phosphatidylinositol hydrolysis and neurite outgrowth by a cell-membrane permeable phosphopeptide. Curr Biol. 1996;6(5):580-7. 45. P Urquhart, A Nicolaou, DF Woodward. Endocannabinoids and their oxygenation by cyclo-oxygenases, lipoxygenses and other oxygenase. Biochim Biophys Acta 2015;1851:366-376. 186 46. Kendall AC, Pilkington SM, Sassano G, Rhodes LE, Nicolaou A. N-Acyl ethanolamide and eicosanoid involvement in irritant dermatitis. Br J Dermatol. 2016;175(1):163-71. 47. S Magina, MA Vieira-Coelho, E Moura, MP Serrão, F Piscitelli, D Moura, V DiMarzo. Effect of narrowband ultraviolet B treatment on endocannabinoid plasma levels in psoriasis patients. Br J Dermatol. 2014; 171:198-201. 48. JD Wilkinson, EM Williamson. Cannabinoids inhibit human keratinocyte proliferation through a non-CB1/CB2 mechanism and have a potential therapeutic value in the treatment of psoriasis. J Dermatol Sci. 2007;45:87-92. 49. Sugawara K, Biro T, Tsuruta D, et al. Endocannabinoids limit excessive mast cell maturation and activation in human skin. J Allergy Clin Immunol 2012;129:726-738. 50. Montecucco F, Di Marzo V. At the heart of the matter: the endocannabinoid system in cardiovascular function and dysfunction. Trends Pharmacol Sci. 2012;33:331-40. 51. Horváth B, Mukhopadhyay P, Haskó G, Pacher P. The endocannabinoid system and plant-derived cannabinoids in diabetes and diabetic complications. Am J Pathol. 2012;180:432-42. 52. Sommer DM, Jenisch S, Suchan M, Christophers E, Weichenthal M. Increased prevalence of the metabolic syndrome in patients with moderate to severe psoriasis. Arch Dermatol Res. 2006;298:321-8. 53. Hill MN, Carrier EJ, McLaughlin RJ, Morrish AC, Meier SE, Hillard CJ, Gorzalka BB. Regional alterations in the endocannabinoid system in an animal model of depression: effects of concurrent antidepressant treatment. J Neurochem. 2008;106:2322–2336. 187 54. Adamczyk P, Gołda A, McCreary AC, Filip M, Przegalinski E. Activation of endocannabinoid transmission induces antidepressant-like effects in rats. J Physiol Pharmacol. 2008;59:217–228. 55. Hill MN, Miller GE, Ho WS, Gorzalka BB, Hillard CJ. Serum endocannabinoid content is altered in females with depressive disorders: a preliminary report. Pharmacopsychiatry. 2008;41:48-53. 56. Hill MB, Miller GE, Carrier EJ, Gorzalka BB, Hillard CJ. Circulating endocannabinoids and N-acyl ethanolamines are differentially regulated in major depression and following exposure to social stress. Psychonueroendocrinology 2009; 34:1257-1262. 57. Lupica CR, Riegel AC. Endocannabinoid release from midbrain dopamine neurons: a potential substrate for cannabinoid receptor antagonist treatment of addiction. Neuropharmacology. 2005;48:1105-16. 58. Fuss J, Steinle J, Bindila L, Auer MK, Kircherr H, Lutz B, Gass P. A runner's high depends on cannabinoid receptors in mice. Proc Natl Acad Sci USA. 2015;112:13105-8. 59. Schwarz T, Schwarz A. Molecular mechanisms of ultraviolet radiation- induced immunosuppression. Eur J Cell Biol 2011;90:560-4. 60. Taub DD, Proost P, Murphy WJ, M Anver M, DL Longo DL, J van Damme J, Oppenheim JJ. Monocyte chemotactic protein-1 (MCP-1), -2, and -3 are chemotactic for human T lymphocytes. J Clin Invest. 1995;95:1370-6. 188 Table 1: Subject demographics Ethnicity White Caucasian South Asian Phototype II V Participants (n) 10 6 Sex (n): Male 2 4 Female 8 2 Age (years) Median Range Median Range 47 30-59 42 23-51 BMI (kg/m2) 25 22-35 25 24-31 MED (mJ/cm2) 30 22-54 125 104-271 189 3890.3) ND 24.9 46.2 161.4 417.8 556.7 417.7 734.8 1985.7 2413.0 1007.1 Friday (15.1 to 40.8) (35.3 to 64.5) (106.4 to 196.5) (313.3 to 555.1) (134.8 to (526.9 to 1112.0) (238.4 to 1614.4) (327.7 to 1037.4) (1109.9 to 3194.2) (2188.8 to 3231.6) e V) p 73.3) ND 20.9 58.6 187.1 288.1 886.9 540.8 893.3 1940.3 3022.9 1048.8 Wednesday (11.1 to 25.4) (41.8 to (27.8 to 376.4) (396.2 to 575.4) (442.8 to 5568.1) (151.9 to 1042.3) (795.6 to 1777.2) (596.3 to 1148.4) (605.1 to 1239.1) (1949.3 to 4015.6) median pg/ml (range) Friday. outh Asians (Phototy S for white Caucasians (n=10) and ND 19.3 39.8 285.6 315.4 740.0 599.4 605.1 1086.1 2695.0 1006.5 Monday (14.8 to 29.7) (30.4 to 60.2) (62.5 to 583.7) (118.0 to 642.8) (577.1 to 5025.0) (550.4 to 1591.5) (362.2 to 1283.2) (350.2 to 1288.9) (410.4 to 1066.2) (2314.4 to 4112.3) exposure* - Wednesday UVR 0.6) e of Monday, ND 25.0 72.4 299.0 326.4 880.2 800.7 524.6 1279.8 3259.0 1478.8 Friday on (12.5 to 25.3) (99.7 to 653.4) (25.0 to 104.2) (165.0 to 900.2) (637.6 to 9039.3) (546.4 to 1650.7) (450.1 to 3272.8) (247.4 to 275 (172.9 to 1095.0) (2087.5 to 9376.9) II) e exposures UVR to 832.9) ND 23.2 82.6 264.9 305.9 818.1 693.9 489.9 .0 to 1197.8 2941.9 1564.7 Wednesday (11.1 to 87.5) (25.0 to 95.6) 571 ( (156.7 to 862.7) (317.0 to 979.7) (212.5 to 2112.9) (418.1 to 1837.9) (475.1 to 2659.0) (315.5 to 1500.3) (2141.0 to 6663.8) prior median pg/ml (range) White Caucasians (Phototyp sampled were ND 25.0 73.6 343.1 860.3 318.4 845.8 650.1 662.8 3054.0 1288.7 Monday (21.0 to 50.0) (61.1 to 103.3) (106.7 to 636.0) (198.8 to 825.2) (312.6 to 2283.2) (550.1 to 1269.3) (562.6 to 2802.3) (271.8 to 1425.3) (144.1 to 1123.8) (2282.6 to 4506.7) *Bloods erum endocannabinoid and NAE levels during week on S detected. AG - OES LEA PEA AEA MEA 2 STEA EPEA DPEA DHEA DGLEA not Table 2: and South Asians (n=6). ND 190 - UVR 4655.5) 7 to 93.5) ND 21.2 72.0 496.4 366.2 928.3 769.9 764.9 1375.7 3040.2 1402.1 (14.5 to 27.3) (50.0 (145.3 to 766.1) (226.6 to 737.9) (259.3 to 886.6) (275.1 to 2283.9) (490.5 to 1345.7) (525.1 to 4173.6) (324.8 to 1445.7) (2279.9 to 1787.9) 6 ND 27.4 73.6 354.1 364.7 887.2 649.7 561.9 1452.5 2916.4 1370.6 (16.8 to 50.0) (37.2 to 125.0) (179.0 to 624.7) (172.3 to 962.7) (250.1 to 3213.8) (475.1 to (825.2 to 2170.3) (242.6 to 1262.8) (259.3 to 1037.4) (2045.1 to 4988.5) 5326.1) 5 ND 27.0 87.9 258.6 348.4 686.5 763.6 1476.5 3301.6 1046.2 1637.8 (14.4 to 38.3) (78.5 to 901.7) (36.4 to 114.3) (350.1 to 3453.2) (148.0 to 2557.6) (354.6 to 1750.4) (454.6 to 3198.0) (399.8 to 1447.9) (230.5 to 1210.3) (1845.8 to and during the six weeks of simulated summer k 0) 4151.5) Week 4 (wee ND 25.0 66.9 325.4 246.8 797.3 656.4 561.9 1269.6 3201.4 1121.0 (13.4 to 42.4) median pg/ml (range) (50.0 to 122.6) (10.6 to 5549.0) (300.1 to 3382.1) (164.9 to 2425.5) (555.5 to 1213.2) (587.6 to 2975.7) (274.3 to 1137.7) (201.7 to 1095.0) (1710.4 to 4963.5) 2696.7) 3 ND 25.0 78.1 354.7 246.5 797.5 651.7 648.4 1609.4 2811.6 1529.0 (14.7 to 37.5) (69.2 to 530.7) (40.8 to 106.6) (165.0 to 900.2) (587.6 to 4246.3) (409.2 to 1700.3) (672.9 to (300.2 to 1412.8) (201.7 to 1066.2) (2350.2 to ) 2 ND 25.4 77.1 209.3 414.3 825.0 626.5 691.6 1182.7 2914.9 1191.3 (12.5 to 40.8) (39.1 to 112.5) (103.0 to 728.5) (403.4 to 896.3) (325.1 to 1853.2) (125.0 to 2360.1) (422.2 to 1487.8) (650.1 to 2518.5) (334.6 to 1675.3) (1920.6 to 4963.5) white Caucasians (n=10 1 ND to 1425.3) 25.0 73.6 343.1 860.3 318.4 845.8 650.1 662.8 3054.0 1288.7 for (21.0 to 50.0) * (61.1 to 103.3) (106.7 to 636.0) (198.8 to 825.2) (312.6 to 2283.2) (550.1 to 1269.3) (562.6 to 2802.3) (271.8 (144.1 to 1123.8) (2282.6 to 4506.7) : Serum endocannabinoid and NAE levels at baseline A A AG - LEA PEA AEA OE MEA 2 STEA EPEA DPEA DHEA DGLEA Table 3 exposures ND not detected. *Bloods were sampled prior to UVR exposures on Monday; exposures were performed Monday, Wednesday and Friday. 191 7 ND 14.2 42.0 259.3 278.1 955.8 565.5 502.4 547.5 1099.2 2553.1 (8.2 to 25.9) (29.5 to 65.8) (71.8 to 631.2) (137.9 to 501.7) (432.0 to 734.7) (288.2 to 993.0) (592.8 to 2151.6) (709.1 to 1333.3) (422.0 to 1187.7) (1922.0 to 2891.2) Friday. and 6 ND 22.3 61.4 268.4 402.4 635.7 469.7 859.7 1357.4 3020.5 1120.3 50.1 to 76.1) (16.0 to 26.9) ( (150.7 to 523.9) (340.6 to 994.2) (201.8 to1144.1) (840.8 to 3297.0) (135.3 to 1069.5) (654.8 to 1597.1) (235.8 to 1815.4) (1942.4 to 3522.6) Wednesday 3626.6) 2.7 Monday, 2 5 ND 25.6 60.7 313.8 515.0 548.9 613.9 806.8 2224.7 29 1120.1 (13.8 to 34.6) (33.0 to 122.0) (271.8 to 418.5) (237.4 to 609.9) (434.8 to 761.8) (876.7 to 6389.6) (853.0 to 1686.6) (448.4 to 1778.7) (547.5 to 2253.9) 2722.8 to (n=6) ( and during the six weeks of simulated summer performed were eek 0) 6.0 3 4 ND 18.7 47.5 Week 148.4 218.2 547.1 508.0 446.6 2257.3 26 1002.4 (w (6.7 to 327.4) (10.4 to 21.9) (26.5 to 56.2) (85.4 to 556.1) (280.6 to 621.5) (288.2 to 749.2) exposures (319.8 to 4850.6) (548.9 to 1325.1) (148.1 to 1377.7) 1866.8 to 3043.9) for South Asians ( median pg/ml (range) for * Monday; on 4026.4) 13166.6) 1570.6) 3 ND 27.7 50.0 280.0 565.8 460.9 558.6 1797.4 2906.8 1114.0 1095.0 (12.2 to 37.7) (29.7 to 83.0) exposures (114.7 to 446.5) (81.9 to 1769.8) (465.2 to 604.9) (691.1 to (380.6 to 1732.8) (253.0 to 1251.4) - (2068.9 to (1402.2 to exposures UVR UVR to 3115.5) 2 ND prior 15.0 50.5 191.3 370.5 759.9 555.6 447.6 540.4 1268.6 2466.9 (12.1 to 32.8) (33.0 to 55.0) (57.8 to 467.3) (115.4 to 951.3) (348.7 to 645.7) (317.0 to 835.7) (959.3 to 4631.1) (533.9 to 1569.8) (332.5 to 1223.9) (1897.2 to sampled were 1283.2) 1 ND 19.3 39.8 285.6 315.4 740.0 599.4 605.1 1086.1 2695.0 1006.5 (14.8 to 29.7) (30.4 to 60.2) *Bloods (62.5 to 583.7) (118.0 to 642.8) (577.1 to 5025.0) (550.4 to 1591.5) (362.2 to (350.2 to 1288.9) (410.4 to 1066.2) (2314.4 to 4112.3) Serum endocannabinoid and NAE levels at baseline B. detected. A AG - LEA PEA AEA OE MEA 2 STEA EPEA DPEA DHEA DGLEA not Table 3 ND 192 Figure Legends Figure 1: Schematic of endocannabinoid and NAE metabolism. The same set of enzymes catalyse NAPE to NAE and NArPE to AEA, and their further metabolism to fatty acid and AA, respectively, while 2-AG is synthesised from DAG by DAG lipase and is also catalysed by MAG lipase. DAG= diacyl glycerol; FAAH= fatty acid amide hydrolase; MAG= monoacyl glycerol; NAPE= N-acyl phosphatidyl ethanolamine; NArPE = N-arachidonyl phosphatidyl ethanolamine; PLA2= phospholipase A2; PLC= phospholipase C; PLD= phospholipase D. Figure 2: Flow chart demonstrating an individual’s progression through the study protocol. Figure 3: Serum endocannabinoid and NAE levels at baseline: A. Data for all participants (n=16). Data shown are median, interquartile and full range. B. A representative UPLC-MS/MS chromatogram. Figure 4: Serum 2-AG levels following UVR exposures. A. For all individuals (n=16) during week 1 of UVR-exposures, blood sampled Monday, Wednesday and Friday (no statistically significant change). B. For all individuals and C. for phototype II (n=10; black) and phototype V (n=6; grey) separately, during the six weeks’ simulated summer UVR-exposures with weekly samples taken, showing an increase from baseline compared with levels over the UVR course (p<0.05 for all 193 subjects; p<0.01 for each phototype separately; ANOVA). Data are logged to achieve normality, and expressed as median, interquartile and full range. 194 Figure 1 195 Figure 2 196 Figure 3A 197 Figure 3B 198 Figure 4A Figure 4B Figure 4C 199 CHAPTER 6: Manuscript 4 Published as a research letter (See appendix 4): Felton S, Adinoff B, Jeon-slaughter H, Jacobe H. The significant health threat from tanning bed use as a self-treatment for psoriasis. J Am Acad Dermatol. 2016;74(5):1015-7. 200 The significant health threat from tanning bed use as a self-treatment for psoriasis Sarah Felton,1 Bryon Adinoff,2, 3 Haekjung Jeon-Slaughter,3 Heidi Jacobe4 1 Dermatology Research Fellow Institute of Inflammation and Repair Manchester Academic Health Science Centre University of Manchester, Manchester, UK 2 Professor Department of Psychiatry University of Texas Southwestern Medical Center 5323 Harry Hines Blvd., Dallas, TX, USA 75390-9070 3 Assistant Professor Department of Psychiatry University of Texas Southwestern Medical Center 5323 Harry Hines Blvd., Dallas, TX, USA 75390-9070 4 Associate Professor Department of Dermatology University of Texas Southwestern Medical Center 5323 Harry Hines Blvd., Dallas, TX, USA 75390-9069 Corresponding Author: Heidi Jacobe, MD, MSCS Associate Professor University of Texas Southwestern Medical Center 5323 Harry Hines Blvd. Dallas, TX 75390-9069 Ph: 214-648-3324 Fx: 214-648-5559 Email: [email protected] Word Count: 2424 Table Count: 6 201 Figure Count: 0 Abstract: 198 Capsule Summary: 50 Funding Sources: This study was supported, in part, by the National Institute on Arthritis and Musculoskeletal and Skin Diseases AR063018. Conflict of Interest: The authors report no conflicts of interest. IRB: This study was qualified as exempt by the Institutional Review Board at the University of Texas Southwestern Medical Center at Dallas, USA. 202 CAPSULE SUMMARY: • Psoriasis improves with sunlight exposure. • A third of psoriasis sufferers had used tanning beds. 11% fulfilled diagnostic criteria for addictive behavior, particularly if they had started tanning to treat their psoriasis. • Physicians should be aware of the possibility of tanning bed usage and addictive tanning behaviors when assessing psoriasis patients. 203 ABSTRACT: Background: Little is known about the prevalence of commercial tanning bed usage in psoriasis sufferers. Objective: To quantify tanning bed usage in psoriasis sufferers and whether their tanning behaviors result in an addictive-like tanning disorder. Methods: Cross-sectional study using an online questionnaire sent to all members of National Psoriasis Foundation (NPF) in USA (February-April 2014). Results: 1932 respondents were analyzed overall. Of the 1832 participants who answered queries regarding tanning bed usage, 34% had used tanning beds at least once and 28% were current users. Individuals who had ever-tanned were younger and female and had been diagnosed with psoriasis at a younger age (p<0.0001 for all). 63% commenced tanning to treat their psoriasis; they were more likely to have received phototherapy in the medical setting and to have experienced more severe psoriasis (p<0.0001 for both). 11% of current tanners fulfilled diagnostic criteria consistent with addictive-like tanning behaviors. This was more common in those who commenced tanning specifically as a treatment for their psoriasis (p<0.02). Limitations: Survey population was limited to members of NPF who had Internet access and opened the e-newsletter mailings (c. 14,000 individuals). Conclusions: Tanning bed usage among psoriasis sufferers is widespread and linked with tanning addiction. Key words: Psoriasis, tanning bed, ultraviolet, phototherapy, addiction 204 INTRODUCTION Psoriasis is a chronic papulosquamous disease that is prevalent in many countries, including the United States. It improves with sunlight exposure in the majority of sufferers1 so a prescribed course of medical phototherapy (usually narrow-band ultraviolet [UV]-B at 311nm) is often a successful treatment for widespread cutaneous plaques.2 Given these improvements, it is likely that some sufferers choose to use UV, particularly commercial tanning beds, outside the medical setting to self-treat their psoriasis. Efficacy of tanning beds for psoriasis treatment has been demonstrated,3 with some physicians endorsing their usage to select groups of patients.4 Although small prevalence studies have suggested that around a third to one half of patients with psoriasis have at some point used commercial tanning beds,5,6 little is known about the tanning behaviors of psoriasis sufferers. Despite the documented association between UV exposure and skin cancer7,8 and public health campaigns designed to heighten awareness of the dangers of tanning,9 recreational tanning behavior in the general public continues unabated.10,11 Motivations for tanning include increased attractiveness, relaxation and mood enhancement.12-14 These latter factors may be considered positive reinforcements to drive further tanning. Addictive behavioral patterns of UV self- administration, similar to those of other substance-related disorders, have been reported in a number of studies.14-17 Surveys of frequent sunbathers and users of tanning beds suggest the prevalence of abnormal behavioral problems in these populations to be over 50%.14,18 These so-called “tanorexics” endorse symptoms such as feelings of anxiety or unattractiveness if their tan is not maintained, an inability to stop tanning and continued tanning despite adverse consequences (e.g. premature ageing, skin cancer). Neurobiological studies support the rewarding 205 potential of UV, as increased activation of striatal brain areas during UV administration implies that UV may have centrally active effects that drive excessive tanning,14 perhaps driven by cutaneous β-endorphin production stimulated by UV exposure.19 Whilst tanning bed usage by the general population is reportedly increasing, previous studies of tanning bed users have focused on those individuals who tan for recreational purposes but typically exclude participants with skin disorders such as psoriasis and other conditions amenable to UV therapy.10,11 In order to better understand the relationship between psoriasis and the potential for pathological tanning behaviors, we conducted a cross-sectional survey of members of the National Psoriasis Foundation (NPF). Questions were asked to evaluate a) whether tanning commenced as a self-treatment for psoriasis, b) their frequencies of and motivations for tanning, and c) whether their indoor tanning behavior resulted in problems consistent with an addictive disorder. These results were analyzed in relation to participant demographics and indicators of psoriasis severity. METHODS Study design and Subjects Survey participants were identified by their NPF membership. An invitation was incorporated into February 2014’s e-newsletter: “Please participate in a short, anonymous online survey to gather information about the duration, severity and impact of your psoriasis, and previous treatments you have tried. This will provide valuable data to direct future treatments offered by dermatology departments.” The NPF reports that approximately 14,000 members open these emails. The invitation included a link to the online survey (supplementary data), with 206 reminder invitations sent one and two months later. To prevent duplication from the same individual, Internet Protocol (IP) addresses were recorded. Respondents were asked to confirm they had received a physician’s diagnosis of psoriasis to be eligible. The Institutional Review Board of University of Texas Southwestern Medical Center at Dallas, USA, qualified the study protocol as exempt. Between February-April 2014, 2077 responses were received. 43 were excluded as their diagnosis was not physician-confirmed. 48 responses were from duplicate IP addresses and 54 individuals did not continue beyond the introductory question “Have you been diagnosed with psoriasis by a doctor?” Therefore, 1932 respondents were included. All provided responses were analyzed. Psoriasis Assessment Introductory questions included basic demographic assessment: age, gender and health insurance status. Questions regarding psoriasis encompassed age at diagnosis, severity-indicators (including previous hospitalization or erythroderma) and previous/currents treatments, specifically topical, systemic, biologic and/or photo- therapies. The previously validated self-assessed Simplified Psoriasis Index (SPI) was used to calculate current psoriasis severity.20 In brief, participants were asked to indicate its degree of interference with daily activities on an ascending scale of 0-10, and to score the overall state of their psoriasis that day, in addition to the extent/severity of their psoriasis on ten named body areas. An overall score of <10 was classified as mild, 10-20 moderate and >20 as severe psoriasis.21 207 Assessment of Tanning Participants were asked if they had ever used UV light outside of the medical setting, such as tanning beds. If affirmative, the survey queried whether tanning had commenced to treat their psoriasis and, if so, what factors had contributed to this self-treatment (e.g. difficulty accessing/cost of care, perceived inefficacy of physician-provided therapy). Other reasons could be described in a free-text area. All tanners were asked their age at commencement and total lifetime number of tanning-sessions. Those who endorsed using a tanning bed within the past 12 months (herein defined as current users) were asked how often they typically used a tanning bed; those who had tanned within the past 5 years were asked about any seasonal variations. In addition, five possible reasons for tanning were presented in a randomly generated order and participants were asked to rank their perceived importance of these reasons for tanning. Those who had tanned within the past 5 years were asked specific questions about tanning behaviors consistent with an addictive disorder, as described in Hillhouse et al (2012)20 and adapted from the Diagnostic Statistical Manual (DSM IV) for Substance Use Disorder.14 To reflect the changes instituted in DSM V, craving was included. To limit the time required to complete the questionnaire, only 7 of the 11 criteria were asked. Positive responses were summed to calculate severity of an addictive disorder, i.e. 2-3 mild, 4-5 moderate, and >5 as severe.22 Participants were also asked if they considered their tanning to be a problem. Those who had ever-tanned were questioned whether they have been diagnosed with any form of skin cancer and, if so, whether they continued to tan following this diagnosis. 208 Statistical Analyses Statistical analyses, specifically Chi-squared and t-tests, were performed using SPSS statistical software (IBM Statistics, Version 21.0. Armonk, NY, USA). Results were considered significant if p<0.05. 209 RESULTS Mean age was 53 (standard deviation [SD] 14) years and mean age at diagnosis of psoriasis 31 [18] years. Two thirds were female. 89% had health insurance, with most (62%) provided by private carrier, 24% Medicare and 3% Medicaid (Table 1). Tanning bed Use 95% of the 1932 participants completed questions regarding tanning bed use (n=1830). 34% of these 1830 respondents had used tanning beds (n=617). Respondents who had tanned were significantly younger, had been diagnosed with psoriasis at a younger age, and were more likely to be female compared to non- tanners (p<0.0001 for all; Table 2). 71% of the 617 tanning bed users were female (n=437) and their mean age at first use of indoor tanning bed was 30 [13] years. 16% started using a tanning bed when aged under 18 years (n=98). Commencement of tanning to treat psoriasis versus tanning for alternate reasons 62% of the 617 participants who had ever-tanned stated that they started tanning in order to treat their psoriasis (n=383). These individuals had been diagnosed at a younger age (p<0.0001), were significantly more likely to have been treated with phototherapy in the medical setting (p<0.0001) and to have experienced more severe psoriasis, as evidenced by their higher incidences of hospitalization and previous erythroderma (p<0.01; p<0.0001) than those tanned for alternate reasons. Insurance status did not differ between those who tanned for psoriasis and those who tanned for other reasons (p>0.05; Table 3). 210 The most commonly stated reasons for tanning as a self-treatment for psoriasis were the inconvenience (30%; n=145) and the expense (30%; n=139) of attending the physician’s office, in addition to the inefficacy of prescribed treatments (30%; n=141). In a minority of cases, tanning commenced upon a physician’s recommendation, due to a lack of available medical phototherapy or for insurance reasons (Table 4). Frequencies of tanning 31% of the 617 participants who have ever-tanned have tanned in excess of 50 times (n=189), of which 15% have tanned over 100 times (n=93). 99% of 170 current users answered questions regarding their tanning frequency, with 55% tanning at least once a week (n=93). For those who have tanned within the past 5 years (n= 322), tanning in the winter months was the most common seasonal pattern at 37% (n=119), but almost a quarter tanned year-round (n=73; Table 5). In the free text area, other reasons stated for variable tanning throughout the year (n=25) included during psoriasis flares, before vacations and when able to afford it. Motivations for tanning 312 current and 295 former tanners ranked their motivations for using a tanning bed. The pattern of responses were the same for the two groups: the most highly ranked incentive was to “prevent sunburn,” followed by “relax, reduce stress or improve mood” and “look or feel good” (equal rank). To “prevent psoriasis from coming back” and lastly “treat or improve existing psoriasis” were the least important reasons, even for the majority of those (55%) who had endorsed tanning-commencement to treat their psoriasis. 211 Tanning and Skin cancer 97% of the 617 individuals who had ever-tanned answered questions regarding skin cancer (n=599). 12% had received a diagnosis of any skin cancer and, of these, 26% (n=19) continued to tan following this diagnosis. Problematic Tanning Behavior 11% of 322 individuals who had tanned within the past 5 years met modified criteria for addictive-like tanning behaviors (n=34; Table 6), of which 79% (n=27) were classified as mild, 18% (n=6) moderate and 3% (n=1) severe. Although 14% of these 322 individuals acknowledged that tanning was a problem for them (n=45), 85% (n=29) of those who met modified criteria for addictive-like tanning behavior did not consider their tanning to be problematic. Individuals who commenced tanning to treat psoriasis were significantly more likely to show features compatible with a tanning addiction than those who commenced for alternate reasons (p<0.05). DISCUSSION This large cross-sectional study examined the frequency of commercial tanning bed use and tanning behaviors in psoriasis sufferers. We found that in our study population of almost 2000 participants, a third of sufferers had used tanning beds and one-fifth were current users. Of these, over 10% acknowledged addictive- like tanning behavior. Usage of tanning beds in our psoriasis population is consistent with two previous studies that identified between a third and half of patients as having ever used non-prescription tanning equipment for their psoriasis.5,6 However, those studies had much smaller cohorts of patients, and were limited to those patients attending university dermatology clinics.5,6 In 212 contrast, our study collected data from a larger group of sufferers throughout the United States and was inclusive of a wide range of ages and severities. In the general population, young females have been shown to be most likely to use tanning beds,23-26 and this was also the case for our psoriasis sufferers. One fifth of our participants specifically commenced tanning to treat their psoriasis. This included those who had received successful phototherapy treatment in a medical setting and subsequently went on to seek out tanning beds themselves. Similarly, in the knowledge that sunlight improves their psoriasis, others “self- medicated” with tanning beds due to difficulties in attending/receiving courses of prescribed phototherapy. Whilst some may consider this to be a cost-effective and beneficial treatment for psoriasis, we would caution against this judgment because we identified that even when individuals initially sought out tanning beds for psoriasis treatment, over half continued tanning for non-medicinal purposes, particularly the prevention of sunburn or for mood enhancement. Few studies have addressed the prevalence of problematic tanning behavior in healthy individuals who use tanning beds for recreational purposes, and their results have varied widely. Mosher et al (2010)27 sampled 421 college students whilst Heckman et al (2013)28 questioned 306 female university students similarly aged 18-25 years. These identified 39% and 5% of participants respectively, as fulfilling criteria for problematic tanning, but are limited by their relatively small sample sizes and strong selection bias. Consistent with other substance use disorders, in our psoriasis population we identified younger age at onset of tanning to be significantly associated with development of addictive behavior.14 Additional notable characteristics of our “addicted” cohort include young age at psoriasis diagnosis, history of severe disease with erythroderma or hospitalization, and previous (prescribed) phototherapy treatment, the latter 213 possibly being analogous to the development of opioid addiction following controlled prescription of these medications for an initial therapeutic purpose.29 When such characteristics are present, the physician should be particularly vigilant as to the possibility of tanning bed use. Limitations of our study include that the survey population was limited to those psoriasis sufferers who were members of NPF, and additionally to those with Internet access who open the e-newsletter mailings. Thus, it is not possible to calculate our exact response rate and, similarly, our cohort may not represent the psoriasis population as a whole. However, the NPF has open membership to any psoriasis-sufferer nationwide. As this is a cross-sectional study design relying on patients’ self-reports of psoriasis and tanning histories, it is subject to an element of recall bias. However, these limitations are at least partially negated by the large sample size, and the absence of any information suggesting that the survey queried tanning behaviors until respondents had completed over a third of the survey. In order to shorten the survey and thus increase our completion rate we limited the number of questions, asking only 7 of the 11 modified DSM IV criteria. In doing so, this would lead to a type II error and so we may therefore be underestimating the prevalence of addictive-like tanning behaviors. Due to the lack of a control group we are unable to confirm whether psoriasis sufferers have an increased risk of developing tanning disorders in comparison to the general population. However, our findings report a significant health threat for psoriasis sufferers, as associations between tanning exposures and both non-melanoma and melanoma risk have been confirmed.10,11 Furthermore, many psoriasis patients already have a heightened risk of skin cancer due to immunosuppressant medications such as ciclosporin, methotrexate and tumour necrosis factor-alpha inhibitors.30-33 214 These findings should alert physicians that a significant number of psoriasis sufferers use tanning beds and, of these, some appear to have a tendency towards addictive-like tanning behaviors. Physicians may wish to directly question their psoriasis patients about any tanning bed use, counsel them on the risks of such practice whilst explaining other treatment options, and offer regular skin examinations. Acknowledgements The authors would like to thank the National Psoriasis Foundation for their generous assistance in disseminating the survey to its members. 215 REFERENCES: 1. Baden HP. The treatment of psoriasis. N Engl J Med. 1963;269:907-9. 2. Coven TR, Burack LH, Gilleaudeau R, Keogh M, Ozawa M, Krueger JG. Narrowband UV-B produces superior clinical and histopathological resolution of moderate-to-severe psoriasis in patients compared with broadband UV-B. Arch Dermatol. 1997;133:1514-22. 3. Su J, Pearce DJ, Feldman SR. The role of commercial tanning beds and ultraviolet A light in the treatment of psoriasis. J Dermatolog Treat. 2005;16:324-6. 4. Fleischer AB, Clark AR, Rapp SR, Reboussin DM, Feldman SR. Commercial tanning bed treatment is an effective psoriasis treatment: results from an uncontrolled clinical trial. J Invest Dermatol. 1997;109(2):170-4. 5. Turner RJ, Farr PM, Walshaw D. Many patients with psoriasis use sunbeds. BMJ. 1998;317(7155):412. 6. Fleischer AB, Feldman SR, Rapp SR, Reboussin DM, Exum ML, Clark AR. Alternative therapies commonly used within a population of patients with psoriasis. Cutis. 1996;58(3):216-20. 7. Coblentz WW. Experimental production of cancer of the skin by ultraviolet radiation; its implications in the use of sunlamps. J Am Med Assoc. 1948;136:1040-3. 8. Health Effects from Ultraviolet Radiation: Report of an Advisory Group on Non-Ionising Radiation. Documents of the NRPB. 2002;13:207-17. 9. Sunsmart. UV, the Sun and Skin Cancer, Retrieved from www.sunsmart.org.uk, updated 21 Feb 2014. 2014; 216 10. Wehner MR, Chren MM, Nameth D, Choudhry A, Gaskins M, Nead KT, et al. International prevalence of indoor tanning: a systematic review and meta- analysis. JAMA Dermatol. 2014;150:390-400. 11. Colantonio S, Bracken MB,Beecker J. The association of indoor tanning and melanoma in adults: systematic review and meta-analysis. J Am Acad Dermatol. 2014;70:847-57 e1-18. 12. Schneider S, Diehl K, Bock C, Schluter M, Breitbart EW, Volkmer B, et al. Tanning bed use, user characteristics, and motivations for tanning: results from the German population-based SUN-Study 2012. JAMA Dermatol. 149:43-9 13. Robinson JK, Kim J, Rosenbaum S,Ortiz S. Indoor tanning knowledge, attitudes, and behavior among young adults from 1988-2007. Arch Dermatol. 2008;144:484-8. 14. Harrington CR, Beswick TC, Leitenberger J, Minhajuddin A, Jacobe HT,Adinoff B. Addiction-like behaviors to ultraviolet light among frequent indoor tanners. Clin Exp Dermatol. 2011;36:33-8. 15. Mayfield D, McLeod G,Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry. 1974;131:1121-3. 16. Nolan BV, Taylor SL, Liguori A,Feldman SR. Tanning as an addictive behavior: a literature review. Photodermatol Photoimmunol Photomed. 2009;25:12-9. 17. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association. 2000;4th ed. 18. Warthan MM, Uchida T,Wagner RF, Jr. UV light tanning as a type of substance-related disorder. Arch Dermatol. 2005;141:963-6. 217 19. Fell GL, Robinson KC, Mao J, Woolf CJ,Fisher DE. Skin beta-Endorphin Mediates Addiction to UV Light. Cell. 2014;157:1527-34. 20. Chularojanamontri L, Griffiths CE,Chalmers RJ. The Simplified Psoriasis Index (SPI): a practical tool for assessing psoriasis. J Invest Dermatol. 133:1956-62. 21. Chularojanamontri L, Griffiths CE, Chalmers RJ. Responsiveness to change and interpretability of the simplified psoriasis index. J Invest Dermatol. 2014;134:351-8. 22. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association 2013, 5th ed. 23. Heckman CJ, Coups EJ,Manne SL. Prevalence and correlates of indoor tanning among US adults. J Am Acad Dermatol. 2008;58:769-80. 24. Boyle R, O'Hagan AH, Donnelly D, Donnelly C, Gordon S, McElwee G, et al. Trends in reported sun bed use, sunburn, and sun care knowledge and attitudes in a U.K. region: results of a survey of the Northern Ireland population. Br J Dermatol. 2010;163:1269-75. 25. Schneider S, Diehl K, Bock C, Schluter M, Breitbart EW, Volkmer B, et al. Tanning bed use, user characteristics, and motivations for tanning: results from the German population-based SUN-Study. JAMA Dermatol. 2012;149:43-9. 26. Guy GP, Berkowitz Z, Tai E, Holman DM, Everett jones S, Richardson LC. Indoor tanning among high school students in the United States, 2009 and 2011. JAMA Dermatol. 2014;150:501-11. 27. Mosher CE,Danoff-Burg S. Addiction to indoor tanning: relation to anxiety, depression, and substance use. Arch Dermatol. 2010;146:412-7. 218 28. Heckman CJ, Darlow S, Kloss JD, et al. Measurement of tanning dependence. J Eur Acad Dermatol Venereol. 2014;28:1179-85. 29. Dhalla IA, Persaud N,Juurlink DN. Facing up to the prescription opioid crisis. BMJ. 2011;343:d5142. 30. Marcil I,Stern RS. Squamous-cell cancer of the skin in patients given PUVA and ciclosporin: nested cohort crossover study. Lancet. 2001;358:1042-5. 31. Paul CF, Ho VC, McGeown C, Christophers E, Schmidtmann B, Guillaume JC, et al. Risk of malignancies in psoriasis patients treated with cyclosporine: a 5 y cohort study. J Invest Dermatol. 2003;120:211-6. 32. Pierard-Franchimont C, Pierard GE,Quatresooz P. Focus on skin cancer association and progression under TNF antagonist therapy. Expert Opin Biol Ther. 2011;11:1215-22. 33. Pouplard C, Brenaut E, Horreau C, Barnetche T, Misery L, Richard MA, et al. Risk of cancer in psoriasis: a systematic review and meta-analysis of epidemiological studies. J Eur Acad Dermatol Venereol. 2013; 27 Suppl 3:36-46. 219 Table 1: Demographics and Clinical Characteristics Mean current age, years (SD) 53 (14) Mean age at diagnosis psoriasis, years (SD) 31 (18) Sex % (n) Male 37% (709) Female 63% (1,223) Insurance Status % (n) No insurance 11% (218) Private Carrier 62% (1,195) Medicaid 3% (54) Medicare 24% (462) Mean current psychological impact of psoriasis (SD) 4 (3) Self-Assessed Simplified Psoriasis Index % (n) Mild 71% (1,309) Moderate 21% (383) Severe 8% (146) Psoriasis Characteristics % (n) Erythroderma 38% (719) Hospitalised 7% (141) Topical Treatment 93% (1,763) Oral Systemic 50% (944) Biologic therapy 46% (860) Phototherapy 51% (962) 220 Table 2: Characteristics of tanners and non-tanners Ever-tanned Never-tanned n=617 n=1213 Age Mean current age, years (SD) * 50 (13) 55 (14) Mean age at diagnosis, years (SD) * 25 (15) 34 (18) Mean age at tanning commencement (SD) 30 (13) N/A Sex Female, % (n) * 71% (437) 61% (739) Male, % (n) 29% (180) 39% (474) Insurance status No insurance, % (n) 12% (72) 11% (135) Private carrier, % (n) 68% (422) 59% (717) Medicaid, % (n) 2% (14) 3% (36) Medicare, % (n) 18% (109) 27% (325) * p<0.0001 221 Table 3: Characteristics of those who commenced tanning bed usage as a treatment for psoriasis Initiated tanning Initiated for psoriasis tanning for n=383 alternate reasons n=234 Age Mean age (SD) yrs 50 (14) 49 (13) Mean age at diagnosis, (SD) yrs *** 22 (13) 29 (16) Sex Female, % (n) ** 66% (256) 79% (182) Male, % (n) ** 34% (127) 21% (52) Insurance status No insurance, % (n) 13% (48) 10% (24) Private Carrier, % (n) 66% (251) % (169) Medicaid, % (n) 3% (10) 2% (4) Medicare, % (n) 19% (72) 16% (37) Psoriasis severity/treatment Previous erythroderma, % (n) *** 55% (213) 34% (78) Hospital admission, % (n) * 11% (43) 5% (12) Systemic medication, % (n) 56% (216) 48% (111) Biologic therapy, % (n) 52% (201) 47% (109) Previous phototherapy, % (n) *** 66% (255) 53% (122) * p≤0.01 ** p≤0.001 *** p≤0.0001 222 Table 4: Reasons for tanning as a self-treatment for psoriasis from 383 respondents* Reason n Inconvenient to go to doctor’s office for ultraviolet light treatment 145 Other treatments prescribed by doctor not working 141 Too expensive to go doctor’s office for ultraviolet light treatment 139 Waiting list at doctor’s office for prescribed ultraviolet light 14 treatment too long Recommended by doctor 13 Phototherapy not available at doctor’s office 9 No insurance/not covered by insurance/high deductible 8 * Responses sum to over 383 as respondents were asked to identify all reasons that may apply 223 Table 5: Tanning Characteristics of tanning bed users Number of # of sessions % (n) Lifetime Tanning Sessions < 10 25% (155) (n=617) 10 - 24 26% (160) 25 - 49 18% (113) 50 - 99 16% (96) 100 - 499 11% (67) > 500 4% (26) Frequency of Frequency Tanning* < once weekly 45% (77) (n=170) once weekly 19% (32) > once weekly 36% (61) Seasonal Season Variation** (n=322) Winter/Fall 37% (119) Year-round 23% (73) Summer 15% (48) Other# 25% (82) * in those who have tanned within the last 12 months ** in those who have tanned within the last 5 years # Other reasons included before vacation or a special event 224 Table 6: Distribution of responses adapted from the Diagnostic Statistical Manual (DSM-V) criteria for Substance Use Disorder endorsed by current tanners (n=301)* Survey question Number of affirmative responses Have you ever considered indoor tanning to be a problem for 45 you? Are there times your indoor tanning is more important to you 8 than most other things in your life such as friends, family, school or job? Has your indoor tanning caused you any problems with other 3 people such as family members, friends, romantic partners or people at work or school? Did you find that when you started using indoor tanning you 30 ended up using it much more than you were planning to? Have you tried to cut down or stop indoor tanning but been 24 unsuccessful? Have you found that you needed a lot more indoor tanning 20 sessions in order to get the feeling you wanted (for example, feel good, reduce stress, feel relaxed, etc) than when you first started indoor tanning? When you haven’t tanned for a while, do you have a strong 36 desire to tan? Have you been diagnosed with a skin cancer and continued to 15 use a tanning bed after this diagnosis? * Responses sum to less than 301 as respondents confirmed only applicable criteria 225 Supplementary Material Psoriasis Severity and Treatment You are invited to participate in a short online survey to help us better understand psoriasis and potential barriers to its treatment. It should take no more than 5-10 minutes of your time. Your participation is completely anonymous. Thank you for your help. 1. Have you been diagnosed with psoriasis by a doctor? Yes No We are sorry but you don't qualify for this survey. Thank you for your time! 2. Are you male or female? Male Female 3. How old are you? years 4. Do you have currently have health insurance? Yes- Medicare Yes- Medicaid Yes- Private carrier No 5. How old were you when you were first diagnosed with psoriasis? years 6. About your psoriasis: Yes No I have had very inflamed psoriasis of ALL my skin I have been admitted to hospital for my psoriasis I have been treated with topical medication prescribed by a physician for my psoriasis e.g. dovonex (calcipotriene), clobetasol I have been treated with oral tablet treatment prescribed by a physician for my psoriasis e.g. methotrexate, cyclosporine, soriatane (acitretin) I have been treated with a ‘biological’ drug by injection or drip prescribed by a physician for my psoriasis e.g. Enbrel (etanercept), Humira (adalimumab), Stelara (ustekinumab) I have had at least one course of ultraviolet light treatment (NBUVB, UVB, psoralen and UVA) IN A DOCTOR’S OFFICE 7. Which of these choices best described the overall state of your psoriasis TODAY? Your score should reflect the average of ALL your psoriasis, not just the worst areas. Clear or just slight redness/scaling Mild redness/scaling with no more than slight thickening Definite redness, scaling or thickening Moderately severe with obvious redness, scaling or thickening Very red and inflamed; very scaly or thick Intensely inflamed skin with or without pustules (pus spots) 226 Psoriasis Severity and Treatment 8. For each of these ten body areas, please click one choice which best describes your psoriasis TODAY: Clear or so minor it does not Obvious but still leaving plenty of Widespread and involving most of bother me normal skin this area Scalp and hairline Face, neck and ears Arms and armpits Hands, fingers and finger nails Chest and stomach Back and shoulders Groin/genitals Buttocks and thighs Knees, lower legs and ankles Feet, toes and toe nails 9. How much is your psoriasis affecting you in your day-to-day activities? Guide: 0 = Not at all 5 = Quite a lot (half the time) 10 = Very much (I could not imagine it bothering me more) 0 1 2 3 4 5 6 7 8 9 10 10. Have you ever used ultraviolet light outside of the medical setting, like a tanning salon? Yes No 11. When you used a tanning bed for the first time, was this to treat your psoriasis? Yes No 12. Did any of these factors contribute to you starting to use a tanning bed? The other treatments prescribed to me by my doctor weren’t working I found it inconvenient to go to my doctor’s office for ultraviolet light treatment It was too expensive to go my doctor’s office for ultraviolet light treatment I had already had a prescribed course of ultraviolet light treatment from my doctor The waiting list at the doctor’s office for prescribed ultraviolet light treatment was too long Other (please specify) 227 Psoriasis Severity and Treatment 13. How old were you when you first used an indoor tanning bed? years 14. In total, since using a tanning bed for the first time, how many times have you used a tanning bed? 15. When was the last time you used a tanning bed? 1-30 days ago 1-12 months ago 1-5 years ago More than 5 years ago/no longer using 16. How often do you typically use an indoor tanning bed? 17. Do you typically use an indoor tanning bed: In the spring/summer In the Fall/winter Year-round Other (please specify) 18. People use a tanning bed for a variety of reasons. How important are these reasons to you? Please rank from 1 = least important, to 5 = most important. To look or feel good To relax, reduce stress or improve my mood To prevent sunburn To prevent my psoriasis from coming back To treat or improve my existing psoriasis 19. Do you tan for any other reasons? No Yes- please explain 228 Psoriasis Severity and Treatment 20. With regards to your tanning: Yes No Have you ever considered indoor tanning to be a problem for you? Are there times your indoor tanning is more important to you than most other things in your life such as friends, family, school or job? Has your indoor tanning caused you any problems with other people such as family members, friends, romantic partners or people at work or school? Did you find that when you started using indoor tanning you ended up using it much more than you were planning to? Have you tried to cut down or stop indoor tanning but been unsuccessful? Have you found that you needed a lot more indoor tanning sessions in order to get the feeling you wanted (for example, feel good, reduce stress, feel relaxed, etc) than when you first started indoor tanning? When you haven’t tanned for a while, do you have a strong desire to tan? 21. Have you been diagnosed with a skin cancer? Yes No 22. Have you continued to use a tanning bed after receiving this diagnosis? Yes No 23. If you received a diagnosis of skin cancer, do you think you would continue to use a tanning bed? Yes No Not Sure 24. People use a tanning bed for a variety of reasons. How important were these reasons to you? Please rank from 1 = least important, to 5 = most important. To prevent my psoriasis from coming back To treat or improve my existing psoriasis To look or feel good To prevent sunburn To relax, reduce stress or improve my mood 229 Psoriasis Severity and Treatment 25. Did you tan for any other reasons? No Yes (please specify) 26. Have you been diagnosed with a skin cancer? Yes No 27. Did you continue to use a tanning bed after this diagnosis? Yes No - I then stopped tanning N/A - I had already stopped tanning Thank you for your time. We hope that your responses will help guide psoriasis treatments in the future. 230 CHAPTER 7: CONCLUSIONS The overall aim of the work presented in this thesis was to evaluate indicators of beneficial and hazardous effects of cumulative low-level UVR exposures, simulating a summer’s sunlight exposures, on human skin. Furthermore, selected consequences were compared between white Caucasians of phototype II, and South Asians of phototype V. Published manuscript 1 examined the principal risk- benefit of summer sunlight exposure in terms of cutaneous DNA damage and vitamin D formation, while manuscript 2 investigated the photoprotective effects against sunburn in white Caucasians, that are conferred by a summer’s sunlight exposure. Manuscript 3 studied the response of serum endocannabinoids to a simulated summer’s sunlight exposures, in view of their potential involvement in the ‘feel good factor’ associated with sunlight exposure. Finally, manuscript 4 assessed the addictive behavioural response to similar low dose UVR-exposures in sufferers of the inflammatory skin condition psoriasis, a disease for which many individuals are prescribed UVR-treatment. 7.1 The principal risk and benefit of summer sunlight exposure The present work demonstrated that low-level UVR exposures, as could be experienced during a UK summer, readily induced keratinocyte DNA damage, in the form of CPD-positive nuclei, in white Caucasian skin and, to a much lesser extent, in South Asian skin in vivo. Furthermore, their induction was significantly positively associated with skin pallor. In contrast with previous studies that employed higher UVR doses (Chouinard et al., 2001, Sheehan et al., 2002), it was reassuring from a human health perspective that no evidence was found for the repeated low-level exposures over six weeks causing any accumulated DNA damage over that induced by one 1.3 SED dose, supporting that damage was 231 effectively repaired between exposures. However in phototype II volunteers, a substantial proportion of DNA damage remained at 24 hours post-UVR, thereby providing the potential for mutagenesis after each DNA-damaging event, and consequently future cutaneous carcinogenesis. Urinary CPD remained below the limit of detection following both single and repeated (x18) low-level UVR exposures, and oxidatively damaged DNA did not increase from baseline levels, for either skin type. Given that CPD were confirmed in skin sections, and they were detected after high dose UVR exposures in a sunbathing holiday at lower latitude (Petersen et al., 2014), while previous work using whole-body UVA exposure resulted in increased urinary 8-oxodG (Cooke et al., 2001), the current results suggest that from these low-levels of UVR exposure the damage was relatively small and/or the number of cells affected was insufficient to generate a signal in urine. These findings regarding urinary biomarkers of DNA damage were also consistent with the observation that cumulative DNA damage was not found in skin after the multiple low dose UVR exposures. The phototype V participants had notably greater increase in skin darkening than phototype II during the UVR course. Since facultative pigmentation includes involvement of higher levels of the epidermis (Alaluf et al., 2002), this can limit penetration of UVB to the chromophore 7-DHC, and consequently the initiation of vitamin D synthesis. Most phototype II participants reached the 25(OH)D concentration ≥20 ng/ml reflecting vitamin D sufficiency (WHO 2003), consistent with a larger sample of white Caucasians studied under the same exposure conditions (Rhodes et al., 2010). In comparison, over half of the South Asian 232 volunteers attained serum 25(OH)D >10 ng/ml, i.e. the cut-off for vitamin D deficiency (WHO 2003), but none reached ≥20 ng/ml, in-line with previous work (Farrar et al., 2011). This is relevant to public health, as in its recent report of July 2016, SACN confirmed that all UK subjects should have a circulating 25(OH)D of ≥10 ng/ml (SACN 2016), while USA/Canada and Europe cite ≥20 ng/ml as the desired target (IOM 2011; European Food Safety Authority 2016). In the current study, a positive association was seen between cutaneous CPD-positive nuclei and serum 25(OH)D gain, but this did not reach statistical significance, possibly reflecting low study numbers. 7.2 UVR-induced photoprotection from summer sunlight exposure In easy-burning white Caucasians (phototype II), the six weeks of repeated low- level simulated sunlight exposures, as following a UK summer, provided photoprotection against sunburn challenge: There was significant reduction in both cutaneous erythema and dermal neutrophil infiltration induced by higher- dose UVR challenge, i.e. 2X MED UVB, with the impact on neutrophils more marked than on clinical erythema. This was novel as photoprotection against subsequent sunburn conferred by low-level UVR given according to a protocol mimicking summer sunlight exposures has not previously been reported, nor the impact on the inflammatory infiltrate, as clinical erythema alone is the usually utilised marker of this acute inflammatory reaction (del Bino et al., 2006). Key mechanisms that may confer photoprotection, namely skin darkening and epidermal thickening (viable epidermis and stratum corneum) were assessed. These volunteers, despite being of phototype II, who reported tanning poorly, showed a small but significant skin-darkening induced by the simulated summer, 233 with measurable reductions in both L* and ITA. Histologically, the repeated low- level UVR (5% UVB, 95% UVA) exposures significantly thickened the stratum corneum with minimal impact on viable (cellular) epidermis, in keeping with a previous study utilising repeated SSR (Lavker et al., 1995). The higher-dose (2X MED) UVB-challenge itself, significantly thickened the viable epidermis, although this appeared to be explained by the evident spongiosis that is an integral part of this acute inflammatory response (Paterson, 2014). Whilst the role of neutrophil infiltration during the sunburn response is not fully understood, these cells have a plethora of pro-inflammatory and immunomodulatory functions, including the generation of chemotactic signals to attract further leukocytes, and release of Th2-associated cytokines including IL-4 and IL-10 (Hawk et al., 1988; Piskin et al., 2005) which are immunosuppressive, and determine if surrounding macrophages adopt a pro- or anti-inflammatory role. Moreover, they phagocytose damaged keratinocyte components, repairing UVR- induced injury (Shapiro et al., 2002; Teunissen et al., 2002). By virtue of their release of proteolytic enzymes including MMP and neutrophil elastase, which damage collagen and elastin fibres, they have also been implicated in the photoageing process (Takeuchi et al., 2010). Thus, although reduction in neutrophil infiltration by photoadaptation is assumed to be beneficial, reflecting reduction of the acute inflammatory response induced by a higher UVR (sunburn) dose, it may also be associated with loss of some beneficial effects. Despite this evidence of photoadaptation following the simulated summer’s sunlight exposures, protection against skin cancer is not assumed since the erythemal component of sunburn is believed to be initiated by UVR-induced 234 DNA damage, as is tanning itself (Eller et al., 1996; de Gruijl, 2002; Pfeifer and Besaratinia 2008). 7.3 Circulating endocannabinoids and their responses to UVR The simulated summer’s sunlight exposures were associated with activation of the endocannabinoid system, with significant elevation in serum 2-AG. Given that the endocannabinoid system is thought to be involved in control of mood status (Adamczyk et al., 2008; Hill et al., 2008; Hill et al., 2009) and, furthermore, that endocannabinoids have been implicated in the “runner’s high” mood elevation experienced after intense aerobic exercise, with increases in circulating AEA occurring in healthy human volunteers (Sparling et al., 2003; Raichlen et al., 2012; Heyman et al., 2012), it is possible that they are, therefore, also involved in the ‘feel good factor’ associated with UVR-exposure. As 2-AG peaked around week 2-3 of the simulated summer, before returning towards baseline, this may to some degree reflect adaptation to the continued low- dose UVR exposures. In view of their increasingly recognised range of activities, endocannabinoid activation is also implicated in the maintenance of homeostasis in the face of external insults such as UVR-induced inflammation, through modulation of arachidonic acid and prostaglandin levels, in addition to immunomodulation, particularly of cell-mediated immunity (Karsak et al., 2007; Zheng et al., 2008; Turcotte et al., 2015). It is, however, unclear whether on balance endocannabinoid responses would increase or reduce the cutaneous inflammation caused by UVR. 235 No differences were identified in endocannabinoid or fatty acid NAE expression between the two phototype groups following the UVR exposures, despite receiving the same absolute UVR doses. Thus one may propose that sufficient molecules absorbing the UVR, i.e. chromophores, for the skin fatty acids influenced, are present at a superficial level in the skin with respect to the melanin present. While there is one previous report examining serum endocannabinoids following UVR exposures (Magina et al., 2014), which showed decreased serum AEA, this differed significantly from the current study, as it involved whole body narrowband UVB phototherapy, escalating dosages, and was performed in psoriasis patients of unspecified phototype. Thus, the current work examining impact of a simulated summer, and people of light and dark skin, has many novel features. 7.4 The addictive behavioural component to tanning The large cross-sectional study of ~2000 psoriasis sufferers examined their frequency of commercial sunbed use and tanning behaviours. In comparison with the simulated summer sunlight utilised in manuscript 3, which comprises 95% UVA, 5% UVB, and with medical phototherapy cabinets of which the majority are narrowband UVB, sunbeds use bulbs that emit UVA and UVB in varying proportions, but again with UVA as the predominant wavelength, often at least 97% UVA with ~3% UVB (Nilsen et al., 2016). The survey in manuscript 4 showed that a third of patients had used sunbeds and one-fifth were current users. In the general population, young females are most likely to use sunbeds (Heckman et al., 2008; Boyle et al., 2010; Schneider et al., 2012), and this was similarly the case for the psoriasis sufferers. One fifth of participants had specifically commenced tanning in order to treat their psoriasis, including those who had received 236 successful phototherapy treatment in a medical setting and subsequently went on to seek out sunbeds themselves. The survey also addressed the prevalence of addictive-like tanning behaviour in psoriatic individuals and found this to be over 10%. Younger age at onset of tanning, younger age at psoriasis diagnosis, a history of severe disease, and previous (prescribed) phototherapy treatment were significantly associated with subsequent development of addictive behaviour. Even when individuals initially declared seeking out sunbeds for psoriasis treatment, over half continued tanning for non-medicinal purposes, particularly for prevention of sunburn or for mood enhancement, further implicating the ‘feel good factor’ reportedly associated with sunlight (Robinson et al., 2008; Harrington et al., 2011; Schneider et al., 2012). 7.5 Synthesis of thesis findings The work presented demonstrates that a simulated summer’s sunlight exposure results in higher levels of keratinocyte DNA damage in skin type II than V, usually accompanied by attainment of vitamin D sufficiency in the former and on achieving the 25(OH)D cut-off for deficiency in the latter. Although this DNA damage does not accumulate over the course, indicating repair keeps pace with damage, a significant proportion is unrepaired at 24 hours post single exposure in skin type II subjects, leaving potential for mutagenesis. The simulated summer’s sunlight provided a degree of photoprotection through stratum corneum thickening, as explored in skin type II subjects, and facultative pigmentation, which was significantly greater in skin type V than II. The photoprotection was demonstrated against an acute UVB challenge, and showed a greater impact on neutrophil 237 infiltration than clinical erythema. It may also have protected against serum 2-AG rise, which was modest. Overall, a simulated summer’s sunlight showed evidence of providing both more beneficial and more harmful effects to individuals of phototype II than V, indicating that guidance on safe sunlight exposure should be better tailored for different phototypes. Thus, exposures could be more restricted for sensitive white skinned individuals at northerly latitudes who attained vitamin D deficiency at the potential risk of skin cancer, whereas increased exposures could be recommended for darker skin types, who experienced negligible DNA damage but remained vitamin D insufficient and consequently at risk of musculoskeletal disease. This could be a useful addition to dietary vitamin D supplementation as advised by SACN (SACN, 2016); moreover supplements may not be popular as this guidance has been provided for darker skin people since 1991, with little heed (DH 1991). Results of this thesis also confirmed that individuals often seek UVR-exposure to “look or feel good”, and that UVR exposure appears to drive further UVR-seeking behaviour, possibly attributable to rise in serum endocannabinoids, specifically 2- AG. Whereas opioids have been popularly suspected to be involved in “tanorrhexia”, this thesis findings suggest the possibility that endocannabinoids might be involved, and this is supported by the knowledge that they would be capable of crossing the blood brain barrier (Biro et al., 2009), such as on release by the skin following UVR exposure. Sunbed addiction was found to be highly prevalent in those with the skin disease psoriasis, where it appears to be at least partially driven by prior UVR-exposures. 238 In the light of these survey results, practitioners offering phototherapy treatment to patients could exert caution and consider alternative treatment options, particularly in at-risk groups such as young females, due to the subsequent risk of skin cancer. 7.6 Strengths and limitations The UVR intervention study had the strength of being conducted in healthy human volunteers of both phototype II and V, while such directly comparative studies are scarce. Furthermore, the protocol was carefully designed to simulate casual exposure to UK summer, using repeated suberythemal exposures of radiation close to solar UVR reaching the earth’s surface (95% UVA, 5% UVB). These low level exposures (equivalent to 13-17 minutes northerly latitude (53.5°N) June midday sunlight, on most days of the week (Webb et al., 2010) were similar to national recommendations on summer sunlight exposure for people with white skin (NRPB 2002), and applied whilst casually-dressed as would be expected in the summer months, thus assisting understanding of the impact of following such guidance. Limitations include the small sample size, which may have resulted in statistical significance being missed for some outcomes, and exclusion of other phototypes. There were no control groups for the intervention, so that while inter-group comparisons could be made between phototypes II and V, other analyses involved intra-group before-after comparisons between baseline and UVR-course end, and between exposed and photoprotected skin. Strengths of the questionnaire survey included its large sample size, and inclusion of sufferers across a large geographical area (USA) regardless of their disease severity. As this was a cross-sectional study design relying on patients’ self-reports 239 of psoriasis and tanning histories, it was also subject to recall bias. However, these limitations were at least partially negated by the large sample size, and the absence of any information suggesting that the survey queried tanning behaviours until respondents had completed over a third of the survey. Due to the lack of a control group it is not possible, however, to confirm whether psoriasis sufferers have an increased risk of developing tanning disorders in comparison to the general population. 7.7 Implications for future research Building on the findings of the current thesis, future studies may explore risk- benefit outcomes in larger cohorts of healthy human volunteers. In particular, this could include representation from all phototype groups (I to VI), and employ differing doses and patterns of UVR-exposures. This may enable assessment of whether there are UVR-dosing regimens at which vitamin D benefit is gained with minimal DNA damage in light-skin people and, similarly, higher level exposures could be given to examine whether dark-skin individuals can experience higher serum 25(OH)D gain with still limited DNA damage. Potentially the effects on serum endocannabinoids could become more pronounced and sustained if an increasing UVR-dosing schedule were employed, as can be found with sunlight and indoor tanning regimes. Histological studies to quantify endocannabinoids/NAE in the skin could examine more directly the relationships between UVR and endocannabinoid production. Concurrent psychological assessment of mood during an UVR-course would enable correlation with serum endocannabinoid levels to further investigate the ‘feel good factor’ of sunlight exposure. 240 Additionally this thesis identified that phototype II individuals had greater urinary 8-oxodG and serum DPEA than phototype V at baseline, with consistent levels at all time points; these findings imply a non-UVR explanation, such as differences in metabolism and/or dietary antioxidant intake between the white Caucasian and South Asian ethnic groups, and warrant further exploration. In addition, further potential risks and benefits of a simulated summer could be examined, such as photoageing and immunomodulation. 241 BIBLIOGRAPHY Adamczyk P, Gołda A, McCreary AC et al. Activation of endocannabinoid transmission induces antidepressant-like effects in rats. J Physiol Pharmacol 2008;59:217–228. Alaluf S, Atkins D, Barrett K et al. Ethnic variation in melanin content and composition in photoexposed and photoprotected human skin. Pigment Cell Res 2002;15:112-8. Andersen PH, Abrams K, Bjerring P et al. A time-correlation study of ultraviolet B-induced erythema measured by reflectance spectroscopy and laser Doppler flowmetry. Photodermatol Photoimmunol Photomed 1991;8:123-8. Armas LA, Dowell S, Akhter M et al. Ultraviolet-B radiation increases serum 25-hydroxyvitamin D levels: the effect of UVB dose and skin color. J Am Acad Dermatol. 2007;57:588-93. Asami S, Manabe H, Miyake J et al. Cigarette smoking induces an increase in oxidative DNA damage, 8-hydroxydeoxyguanosine, in a central site of the human lung. Carcinogenesis 1997;18:1763-6. Aubin F, Zultak M, Blanc D et al. Reaction to UV-induced erythema in young patients with basal cell carcinoma. Photodermatol. 1989;6:118–123. Baden HP. The treatment of psoriasis. N Engl J Med. 1963;269:907-9. Belon PE. UVA exposure and pituitary secretion. Variations of human lipotropin concentrations (beta LPH) after UVA exposure. Photochem Photobiol 1985;42:327-9. Berdyshev EV, Schmid PC, Dong Z et al. Stress-induced generation - acylethanolamines in mouse epidermal JB6 P+ cells. Biochem J 2000;346 Pt 2:369- 74. Berry JL, Davies M, Mee AP. Vitamin D metabolism, rickets, and osteomalacia. Semin Musculoskelet Radiol 2002;6:173-82. Berry JL, Selby PL, Davies M, Martin J. Observations from the UK Supra- Regional Assay Service laboratory for the measurement of Vitamin D metabolites. J Steroid Biochem Mol Biol. 2007;103:477-9. Birch-Machin MA, Tindall M, Turner R et al. Mitochondrial DNA deletions in human skin reflect photo- rather than chronologic aging. J Invest Dermatol 1998;110:149–52. Birch-Machin MA, Swalwell H. How mitochondria record the effects of UV exposure and oxidative stress using human skin as a model tissue. Mutagenesis 2010;25:101–7. Bíró T, Tóth B, Haskó G et al. The endocannabinoid system of the skin in health and disease: novel perspectives and therapeutic opportunities. Trends Pharmacol Sci 2009;30:411-420. Bischoff-Ferrari HA, Dietrich T, Orav EJ et al. Positive association between 25-hydroxy vitamin D levels and bone mineral density: a population-based study of younger and older adults. Am J Med 2004;116:634-9. Bisogno T. Endogenous cannabinoids: structure and metabolism. J Neuroendocrinol 2008;20 Suppl 1:1-9. Björn LO. Vitamin D synthesis may be independent of skin pigmentation only with UV of short wavelength. J Invest Dermatol 2010;130:2848-50. Bogh MK, Schmedes AV, Philipsen PA et al. Vitamin D production after UVB exposure depends on baseline vitamin D and total cholesterol but not on skin pigmentation. J Invest Dermatol 2010;130:546-53. Bogh MK, Schmedes AV, Philipsen PA et al. Vitamin D production depends 242 on ultraviolet-B dose but not on dose rate: a randomized controlled trial. Exp Dermatol 2011;20:14-8. Boorman E, Zajkowska Z, Ahmed R, et al. Crosstalk between endocannabinoid and immune systems: a potential dysregulation in depression? Psychopharmacology 2016;233:1591-604. Bouwes Bavinck JN, Hardie DR, Green A et al. The risk of skin cancer in renal transplant recipients in Queensland, Australia. A follow-up study. Transplantation 1996;61:715-21. Boyle R, O'Hagan AH, Donnelly D et al. Trends in reported sun bed use, sunburn, and sun care knowledge and attitudes in a U.K. region: results of a survey of the Northern Ireland population. Br J Dermatol 2010;163:1269-75. Brash DE, Rudolph JA, Simon JA et al. A role for sunlight in skin cancer: UV- induced p53 mutations in squamous cell carcinoma. Proc Natl Acad Sci U S A 1991; 88:10124-8. Breiden B, Sandhoff K. The role of sphingolipid metabolism in cutaneous permeability barrier formation. Biochim Biophys Acta 2014;1841:441–52. Brenner M & Hearing VJ. The protective role of melanin against UV damage in human skin. Photochem Photobiol. 2008;84:539–549. British Association of Dermatology (2012). "Sun Awareness Campaign." Public Information: Skin Cancer. Retrieved from http://www.bad.org.uk/for-the- public/sun-awareness-campagin 02/11/2015. Bruls WA, Slaper H, van der Leun JC et al. Transmission of human epidermis and stratum corneum as a function of thickness in the ultraviolet and visible wavelengths. Photochem. Photobiol.1984;40:485–494. Bykov VJ, Sheehan JM, Hemminki K, Young AR. In situ repair of cyclobutane pyrimidine dimers and 6-4 photoproducts in human skin exposed to solar simulating radiation. J Invest Dermatol 1999;112:326-31. Cancer Council Australia (2014). Public Information: Skin Cancer. Retrieved from http://www.cancer.org.au/about-cancer/what-is-cancer/facts-and- figures.html 02/11/2015. Cancer Research U.K (2010). Public Information: Incidence statistics. Retrieved from http://www.cancerresearchuk.org/about-cancer/type/skin- cancer/treatment/statistics-and-outlook-for-skin-cancer 02/11/2015 Cancer Research UK (2013). "Sunsmart Campaign". Public Information. Retrieved from http://www.sunsmart.org.uk/ 02/11/2015. Carr DB, Bullen BA, Skinar GS et al. Physical conditioning facilitates the exercise-induced secretion of beta-endorphin and beta-lipotropin in women. N Engl J Med 1981;305:560-3. Carsberg CJ, Ohanian J, Friedmann PS. Ultraviolet radiation stimulates a biphasic pattern of 1,2-diacylglycerol formation in cultured human melanocytes and keratinocytes by activation of phospholipases C and D. Biochem J 1995;305 (Pt 2):471-7. Centers for Disease Control and Prevention (2010). Public Information: Skin Cancer Rates by State. Retrieved from http://www.cdc.gov/cancer/skin/statistics/state.htm 02/11/2015. Chardon A, Cretois I, Hourseau C. Skin colour typology and suntanning pathways. Int J Cosmet Sci. 1991;13:191-208. Chouinard N, Therrien JP, Mitchell DL et al. Repeated exposures of human skin equivalent to low doses of ultraviolet-B radiation lead to changes in cellular functions and accumulation of cyclobutane pyrimidine dimers. Biochem Cell Biol 2001;79:507-15. 243 Chularojanamontri L, Griffiths CE, Chalmers RJ. The Simplified Psoriasis Index (SPI): a practical tool for assessing psoriasis. J Invest Dermatol 2013; 133:1956-62. Chularojanamontri L, Griffiths CE, Chalmers RJ. Responsiveness to change and interpretability of the simplified psoriasis index. J Invest Dermatol 2014;134:351-8. CIE. Action spectrum for the production of previtamin D3 in human skin. J Bouillon, J Eisman, M Garabedian et al. International Commission on Illumination 2006; 174. Clemens TL, Adams JS, Henderson SL, Holick MF. Increased skin pigment reduces the capacity of skin to synthesise vitamin D3. Lancet 1982;1:74-6. Coblentz WW. Experimental production of cancer of the skin by ultraviolet radiation; its implications in the use of sunlamps. J Am Med Assoc 1948;136:1040- 3. Colantonio S, Bracken MB, Beecker J. The association of indoor tanning and melanoma in adults: systematic review and meta-analysis. J Am Acad Dermatol 2014;70:847-57 e1-18. Cooke MS, Evans MD, Burd RM et al. Induction and excretion of ultraviolet- induced 8-oxo-2'-deoxyguanosine and thymine dimers in vivo: implications for PUVA. J Invest Dermatol 2001;116:281-5. Cooke MS, Olinski R, Loft S. Measurement and meaning of oxidatively modified DNA lesions in urine. Cancer Epidemiol Biomarkers Prev. 2008;17:3-14. Coven TR, Burack LH, Gilleaudeau R, et al. Narrowband UV-B produces superior clinical and histopathological resolution of moderate-to-severe psoriasis in patients compared with broadband UV-B. Arch Dermatol. 1997;133:1514-22. Cripps, D. J. Natural and artificial photoprotection. J Invest Dermatol 1981; 77:154-7. Dawson-Hughes B, Mithal A, Bonjour JP et al. IOF position statement: vitamin D recommendations for older adults. Osteoporos Int 2010;21:1151-4. de Gruijl FR. Photocarcinogenesis: UVA vs. UVB radiation. Skin Pharmacology and Applied Skin Physiology, 2002:15:316–320. de Winter S, Vink AA, Roza L et al. Solar-simulated skin adaptation and its effect on subsequent UV-induced epidermal DNA damage. J Invest Dermatol 2001;117:678-82. Del bino S, Sok J, Bessac E, Bernerd F. Relationship between skin response to ultraviolet exposure and skin color type. Pigment Cell Res. 2006; 19:606-14. Del bino S, Bernerd F. Variations in skin colour and the biological consequences of ultraviolet radiation exposure. Br J Dermatol 2013;169 Suppl 3:33-40. DH (1998). Nutrition and Bone Health: with particular reference to calcium and vitamin D. London, Department of Health. no.49. DH (2010). Vitamin D: an essential nutrient for all...but who is at risk of vitamin D deficiency? Important information for healthcare professionals, Department of Health. DH (2011). Sunbeds (Regulations) Act 2010: Guidance on the implementation of the Sunbeds (Regulation) Act 2010. London, Department of Health. Dhalla IA, Persaud N, Juurlink DN. Facing up to the prescription opioid crisis. BMJ 2011; 343:d5142. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association 2000; 4th ed. text rev. 244 Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association 2013, 5th ed. Diffey BL, Farr PM, Oakley AM. Quantitative studies on UVA-induced erythema in human skin. Br J Dermatol 1987;117:57-66. Diffey BL, Jansen CT, Urbach F, Wulf HC. The standard erythema dose: a new photobiological concept. Photodermatol Photoimmunol Photomed 1997;13:64–6. DiGiovanna JJ & Kraemer KH. Shining a light on xeroderma pigmentosum. J Invest Dermatol 2012;132:785-96. Di nuzzo S, De rie MA, Van der loos CM, et al. Solar-simulated ultraviolet irradiation induces selective influx of CD4+ T lymphocytes in normal human skin. Photochem Photobiol 1996;64:988-93. Drobetsky EA, Turcotte J, Chateauneuf A et al. A role for ultraviolet A in solar mutagenesis. Proc Natl Acad Sci U S A 1995;92:2350-4. Duval C, Régnier M, Schmidt R. Distinct melanogenic response of human melanocytes in mono-culture, in co-culture with keratinocytes and in reconstructed epidermis, to UV exposure. Pigment Cell Res 2001;14:348-55. El-Domyati M, Attia S, Saleh F et al. Intrinsic aging vs. photoaging: a comparative histopathological, immunohistochemical, and ultrastructural study of skin. Exp Dermatol 2002; 11: 398-405. Eller MS, Ostrom K, Gilchrest BA et al. DNA damage enhances melanogenesis. Proc Natl Acad Sci U S A 1996; 93: 1087-92. Engel LS, Orlow I, Sima CS et al. Vitamin D receptor gene haplotypes and polymorphisms and risk of breast cancer: a nested case-control study. Cancer Epidemiol Biomarkers Prev 2012; 21: 1856-67. Epstein JH, Fukuyama K, Fye K. Effects of ultraviolet radiation on the mitotic cycle and DNA, RNA and protein synthesis in mammalian epidermis in vivo. Photochem Photobiol 1970;12:57-65. European Food Safety Authority Panel on Dietetic Products, Nutrition and Allergies. Scientific opinion on dietary reference values for vitamin D. EFSA J 2016;14: in press. European tanning. Public Information. http://www.europeantanning.com/high-pressure-tanning/ Accessed 01/07/2014. Farrar MD, Kift R, Felton SJ et al. Recommended summer sunlight exposure amounts fail to produce sufficient vitamin D status in UK adults of South Asian origin. Am J Clin Nutr 2011;94:1219-24. Farrar MD, Webb AR, Kift R et al. Efficacy of a dose range of simulated sunlight exposures in raising vitamin D status in South Asian adults: implications for targeted guidance on sun exposure. Am J Clin Nutr 2013;97:1210-6. Felder CC, Briley EM, Axelrod J et al. Anandamide, an endogenous cannabimimetic eicosanoid, binds to the cloned human cannabinoid receptor and stimulates receptor-mediated signal transduction. Proc Natl Acad Sci USA 1993;90:7656–60. Feldman SR, Liguori A, Kucenic M et al. Ultraviolet exposure is a reinforcing stimulus in frequent indoor tanners. J Am Acad Dermatol 2004;51:45-51. Fell GL, Robinson KC, Mao J et al. Skin β-Endorphin Mediates Addiction to UV Light. Cell. 2014;157:1527-34. Felton S, Navid F, Schwarz A, et al. Ultraviolet radiation-induced upregulation of antimicrobial proteins in health and disease. Photochem Photobiol Sci. 2013;12:29-36. Felton S, Adinoff B, Jeon-slaughter H, Jacobe H. The significant health threat from tanning bed use as a self-treatment for psoriasis. J Am Acad Dermatol. 245 2016;74:1015-7. Felton SJ, Cooke MS, Kift R, et al. Concurrent beneficial (vitamin D production) and hazardous (cutaneous DNA damage) impact of repeated low-level summer sunlight exposures. Br J Dermatol. 2016; doi 10.1111/bjd.14863. Feuerecker M, Hauer D, Gresset T et al. Effect of an acute consumption of a moderate amount of ethanol on plasma endocannabinoid levels in humans. Alcohol Alcohol 2012;47:226-32. Fisher GJ, Datta SC, Talwar HS et al. Molecular basis of sun-induced premature skin ageing and retinoid antagonism. Nature 1996; 379: 335-9. Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988; 124: 869-71. Fleischer AB, Feldman SR, Rapp SR, et al. Alternative therapies commonly used within a population of patients with psoriasis. Cutis. 1996;58:216-20. Fleischer AB, Jr., Clark AR, Rapp SR, et al. Commercial tanning bed treatment is an effective psoriasis treatment: results from an uncontrolled clinical trial. J Invest Dermatol 1997;109:170-4. Fleischer J, Krusekopf S, Goldammer M et al. Vitamin D receptor haplotypes protect against development of colorectal cancer. Eur J Clin Pharmacol 2007;63: 997-1005. Food Standards Agency. McCance and Widdowson's the composition of foods. 6th ed. Cambridge, United Kingdom: Royal Society of Chemistry, 2002. Franklin WA & Haseltine WA. Removal of UV light-induced pyrimidine- pyrimidone (6-4) products from Escherichia coli DNA requires the uvrA, uvrB, and urvC gene products. Proc Natl Acad Sci U S A 1984;81:3821-4. Fuss J, Steinle J, Bindila L et al. A runner's high depends on cannabinoid receptors in mice. Proc Natl Acad Sci U S A 2015;112:13105-8. Galloway AM, Liuzzi M, Paterson MC. Metabolic processing of cyclobutyl pyrimidine dimers and (6-4) photoproducts in UV-treated human cells. Evidence for distinct excision-repair pathways. J Biol Chem 1994;269:974-80. Gambichler T, Bader A, Vojvodic M et al. Plasma levels of opioid peptides after sunbed exposures. Br J Dermatol 2002;147:1207-11. Ghafouri N, Ghafouri B, Turkina, MV et al. High levels of N- palmitoylethanolamide and N-stearoylethanolamide in microdialysate from myalgic trapezius muscle in women. PLoS ONE 2011;6:e27257. Gillie O. A new government policy is needed for sunlight and vitamin D. Br J Dermatol 2006;154:1052–1061. Gläser R, Navid F, Schuller W et al. UV-B radiation induces the expression of antimicrobial peptides in human keratinocytes in vitro and in vivo. J Allergy Clin Immunol 2009;123:1117-23. Gniadecka M, Wulf HC, Mortensen NN et al. Photoprotection in vitiligo and normal skin. A quantitative assessment of the role of stratum corneum, viable epidermis and pigmentation. Acta Derm. Venereol. 1996;76: 429–432. Grant WB. Epidemiology of disease risks in relation to vitamin D insufficiency. Prog Biophys Mol Biol 2006;92:65-79. Greinert R, Boguhn O, Harder D et al. The dose dependence of cyclobutane dimer induction and repair in UVB-irradiated human keratinocytes. Photochem Photobiol 2000;72:701-8. Guindon J, Hohmann AG. The endocannabinoid system and pain. CNS Neurol Disord Drug Targets. 2009;8:403-21. Guy GP, Jr., Berkowitz Z, Tai E, et al. Indoor Tanning Among High School Students in the United States, 2009 and 2011. JAMA Dermatol 2014;150:501-11. 246 Hall H, Williams EJ, Moore SE, Walsh FS, Prochiantz A, Doherty P. Inhibition of FGF-stimulated phosphatidylinositol hydrolysis and neurite outgrowth by a cell- membrane permeable phosphopeptide. Curr Biol 1996;6:580-7. Hansen HS, Artmann A. Endocannabinoids and nutrition. J Neuroendocrinol 2008;20:94-9. Harrington CR, Beswick TC, Leitenberger J et al. Addictive-like behaviours to ultraviolet light among frequent indoor tanners. Clin Exp Dermatol 2011;36:33-8. Harrison GI, & Young AR. Ultraviolet radiation-induced erythema in human skin. Methods 2002;28:14-9. Hawk JL, Murphy GM & Holden CA. The presence of neutrophils in human cutaneous ultraviolet-B inflammation. Br J Dermatol 1988; 118: 27–30. Health Protection Agency (2009). Public Information "Sunsense". Retrieved from http://hpa.org.uk/radiation/understand/at_a_glance/sunsense/sunsense.htm 02/11/2015. Heckman CJ, Coups EJ, Manne SL. Prevalence and correlates of indoor tanning among US adults. J Am Acad Dermatol 2008;58:769-80. Heyman E, Gamelin FX, Goekint M et al. Intense exercise increases circulating endocannabinoid and BDNF levels in humans-possible implications for reward and depression. Psychoneuroendocrinology 2012;37:844-51. Hedges T, Scriven A. Sun safety: what are the health messages? J R Soc Promot Health 2008;128:164-9. Hill MN, Carrier EJ, McLaughlin RJ et al. Regional alterations in the endocannabinoid system in an animal model of depression: effects of concurrent antidepressant treatment. J Neurochem 2008;106:2322–2336. Hill MN, Miller GE, Ho WS, et al. Serum endocannabinoid content is altered in females with depressive disorders: a preliminary report. Pharmacopsychiatry 2008;41:48-53. Hill MB, Miller GE, Carrier EJ et al. Circulating endocannabinoids and N-acyl ethanolamines are differentially regulated in major depression and following exposure to social stress. Psychonueroendocrinology 2009;34:1257-1262. Holleran WM, Uchida Y, Halkier-sorensen L et al. Structural and biochemical basis for the UVB-induced alterations in epidermal barrier function. Photodermatol Photoimmunol Photomed. 1997;13:117-28. Holmes TCA & Holmes RMM. Clearance of UVB-induced erythema in patients with non-melanoma skin cancer. Photodermatol. Photoimmunol. Photomed 1997; 13:189–192. Hollis BW. Circulating 25-hydroxyvitamin D levels indicative of vitamin D sufficiency: implications for establishing a new effective dietary intake recommendation for vitamin D. J Nutr 2005;135:317-22. Houghten RA, Swann RW, Li CH. beta-Endorphin: stability, clearance behavior, and entry into the central nervous system after intravenous injection of the tritiated peptide in rats and rabbits. Proc Natl Acad Sci USA. 1980;77:4588-91. Horváth B, Mukhopadhyay P, Haskó G, Pacher P. The endocannabinoid system and plant-derived cannabinoids in diabetes and diabetic complications. Am J Pathol 2012;180:432-42. Howlett AC. The cannabinoid receptors. Prostaglandins Other Lipid Mediat 2002:68-69:619-631. Hsu GW, Ober M, Carrel T et al. Error-prone replication of oxidatively damaged DNA by a high-fidelity DNA polymerase. Nature 2004;431: 217-21. Hudson L, Bowman A, Rashdan E, Birch-machin MA. Mitochondrial damage 247 and ageing using skin as a model organ. Maturitas. 2016; May 7. pii: S0378- 5122(16)30100-1. doi: 10.1016/j.maturitas.2016.04.021. [Epub ahead of print] IARC. International Agency for Research on Cancer monographs on the evaluation of carcinogenic risks to humans. Solar and ultraviolet radiation. IARC Monogr Eval Carcinog Risks Hum 1992;55:1-316. IARC. International Agency for Research on Cancer (2008). Vitamin D and Cancer. IARC Working Group Reports, 2008. Eds World Health Organization. IARC. International Agency for Research on Cancer (2009). Monographs on the evaluation of carcinogenic risks to humans. Public Information: Sunbeds and UV radiation. Retrieved from http://www.iarc.fr/en/mediacentre/iarcnews/2009/sunbeds_uvradiation.php 02/11/2015 Ibbotson SH, Bilsland D, Cox NH et al. An update and guidance on narrowband ultraviolet B phototherapy: a British Photodermatology Group Workshop Report. Br J Dermatol 2004;151:283-97. Institute of Medicine. Dietary reference intakes for calcium, phosphorus magnesium, vitamin D and fluoride. I. o. Medicine. 1997. Washington, DC, The National Academies Press. Institute of Medicine 2011. Dietary reference intakes for calcium and vitamin D. Washington, DC: The National Academies Press. Jones PJ, Lin L, Gillingham LG et al. Modulation of plasma N-acylethanolamine levels and physiological parameters by dietary fatty acid composition in humans. J Lipid Res 2014;55:2655-64. Jung EG, Furtwängler M, Klostermann G et al. Light-induced skin cancer and prolonged uv-erythema. Arch. Dermatol. Res. 1980; 267: 33–36. Kaidbey KH, Agin PP, Sayre RM et al. Photoprotection by melanin--a comparison of black and Caucasian skin. J Am Acad Dermatol 1979;1:249-60. Kamiya H, Murata-Kamiya N, Suzuki M et al. 8-Hydroxyguanine (7,8-dihydro- 8-oxoguanine) in hot spots of the c-Ha-ras gene: effects of sequence contexts on mutation spectra. Carcinogenesis 1995;16:883-9. Kang K, Gilliam AC, Chen G et al. In human skin, UVB initiates early induction of IL-10 over IL-12 preferentially in the expanding dermal monocytic/macrophagic population. J Invest Dermatol 1998;111:31-8. Kappes UP, Luo D, Potter M et al. Short- and long-wave UV light (UVB and UVA) induce similar mutations in human skin cells. J Invest Dermatol 2006;126:667-75. Karsak M, Gaffal E, Date R et al. Attenuation of allergic contact dermatitis through the endocannabinoid system. Science 2007;316:1494-7. Kasai H, Hayami H, Saito H et al. Detection and identification of mutagens and carcinogens as their adducts with guanosine derivatives. Nucleic Acids Res 1984;12:2127-36. Kaur M, Liguori A, Kucenic M et al. Induction of withdrawal-like symptoms in a small randomized, controlled trial of opioid blockade in frequent tanners. J Am Acad Dermatol 2006;54:709-11. Kendall AC, Nicolaou A. Bioactive lipid mediators in skin inflammation and immunity. Prog Lipid Res 2013; 52:141–64. Kendall AC, Pilkington SM, Massey KA et al. Distribution of bioactive lipid mediators in human skin. J Invest Dermatol 2015;135:1510-20. Kendall AC, Pilkington SM, Sassano G, Rhodes LE, Nicolaou A. N-Acyl ethanolamide and eicosanoid involvement in irritant dermatitis. Br J Dermatol 2016;175:163-71. 248 Kligman AM. Early destructive effect of sunlight on human skin. JAMA 1969; 210:2377-80. Kress S, Sutter C, Strickland PT et al. Carcinogen-specific mutational pattern in the p53 gene in ultraviolet B radiation-induced squamous cell carcinomas of mouse skin. Cancer Res 1992;52:6400-3. Kripke ML, Cox PA, Alas AG et al. Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice. Proc Natl Acad Sci U S A 1992; 89:7516-20. Kupczyk P, Reich A, Szepietowski JC. Cannabinoid system in the skin – a possible target for future therapies in dermatology. Exp Dermatol 2009;18:669–79. Kvam E, Tyrrell RM. Induction of oxidative DNA base damage in human skin cells by UV and near visible radiation. Carcinogenesis 1997;18:2379-84. Lam PM, Mistry V, Marczylo TH et al. Rapid measurement of 8-oxo-7,8- dihydro-2'-deoxyguanosine in human biological matrices using ultra-high- performance liquid chromatography-tandem mass spectrometry. Free Radic Biol Med 2012;52:2057-63. Lavker RM & Kligman AM. Chronic heliodermatitis: a morphologic evaluation of chronic actinic dermal damage with emphasis on the role of mast cells. J Invest Dermatol 1988;90:325-30. Le Curieux F, Hemminki K. Cyclobutane thymidine dimers are present in human urine following sun exposure: Quantitation using 32P-postlabeling and High-performance liquid chromatography. J Invest Dermatol 2001; 117: 263-268. Lee PL, van Weelden H & Bruijnzeel PLB. Neutrophil infiltration in normal human skin after exposure to different ultraviolet radiation sources. Photochem. Photobiol. 2008; 84:1528–34. Leiter U & Garbe C. Epidemiology of melanoma and nonmelanoma skin cancer-the role of sunlight. Adv Exp Med Biol 2008;624:89-103. Levins PC, Carr DB, Fisher JE et al. Plasma beta-endorphin and beta- lipoprotein response to ultraviolet radiation. Lancet 1983;2:166. Liljendahl TS, Kotova N, Segerback D. Quantification of ultraviolet radiation- induced DNA damage in the urine of Swedish adults and children following exposure to sunlight. Biomarkers 2012;17:634-41. Liljendahl TS, Blomqvist A, Andersson EM et al. Urinary levels of thymine dimer as a biomarker of exposure to ultraviolet radiation in humans during outdoor activities in the summer. Mutagenesis 2013;28:249-56. Liu F-T, Yang NC. Photochemistry of cytosine derivatives. 1. Photochemistry of thymidylyl-(3’5’)-deoxycytidine. Biochemistry 1978;17:4865-4876. Lo CW, Paris PW, Holick MF. Indian and Pakistani immigrants have the same capacity as Caucasians to produce vitamin D in response to ultraviolet irradiation. Am J Clin Nutr 1986;44:683-5. Lopez H, Beer JZ, Miller SA, Zmudzka BZ. Ultrasound measurements of skin thickness after UV exposure: a feasibility study. J Photochem Photobiol B, Biol. 2004;73:123-32. Lovegrove, JA Lovegrove SS, Lesauvage SV. Moderate fish-oil supplementation reverses low-platelet, long-chain n-3 polyunsaturated fatty acid status and reduced plasma triacylglyceral concentration in British Indo-Asians. Am J Clin Nutr 2004;79:974-82. Lupica CR, Riegel AC. Endocannabinoid release from midbrain dopamine neurons: a potential substrate for cannabinoid receptor antagonist treatment of addiction. Neuropharmacology 2005;48:1105-16. Lurie SJ, Gawinski B, Pierce D, Rousseau SJ. Seasonal affective disorder. Am 249 Fam Physician 2006;74:1521-4. MacLaughlin JA, Anderson RR, Holick MF. Spectral character of sunlight modulates photosynthesis of previtamin D3 and its photoisomers in human skin. Science 1982; 216:1001-3. MacLaughlin J & Holick MF. Aging decreases the capacity of human skin to produce vitamin D3. J Clin Invest 1985;76:1536-8. Madronich S, McKenzie RL, Bjorn LO, Caldwell MM. Changes in biologically active ultraviolet radiation reaching the Earth's surface. J Photochem Photobiol 1998;46:5-19. Magina S, Vieira-coelho M, Moura E et al. Effect of narrowband ultraviolet B treatment on endocannabinoid plasma levels in psoriasis patients. Br J Dermatol 2014;171:198-201. Manav M, Su J, Hughes K et al. Omega-3 fatty acids and selenium as coronary heart disease risk modifying factors in Asian Indian and Chinese males. Nutrition 2004;20:967-73. Marcil I, Stern RS. Squamous-cell cancer of the skin in patients given PUVA and ciclosporin: nested cohort crossover study. Lancet 2001;358:1042-5. Matsuda LA, Lolait SJ, Brownstein MJ et al. Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 1990;346:561-4. May E, Asadullah K, Zugel U. Immunoregulation through 1,25- dihydroxyvitamin D3 and its analogs. Curr Drug Targets Inflamm Allergy 2004;3:377-93. Mayfield D, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry 1974;131:1121-3. Mechoulam R, Fride E, Di Marzo V. Endocannabinoids. Eur J Pharmacol 1998;359:1-18. Mechoulam R, Hanus L. A historical overview of chemical research on cannabinoids. Chem Phys Lipids 2000;108:1-13. Mckenzie R, Scragg R, Liley B et al. Serum 25-hydroxyvitamin-D responses to multiple UV exposures from solaria: inferences for exposure to sunlight. Photochem Photobiol Sci 2012;11:1174-85. Mellanby E. An experimental investigation on rickets. Lancet 1919; I:407. Miller SA, Coelho SG, Zmudzka BZ et al. Dynamics of pigmentation induction by repeated ultraviolet exposures: dose, dose interval and ultraviolet spectrum dependence. Br J Dermatol 2008;159:921-30. Mitchell DL. The induction and repair of lesions produced by the photolysis of (6-4) photoproducts in normal and UV-hypersensitive human cells. Mutat Res 1988;194:227-37. Mitchell DL, Greinert R, de Gruijl FR et al. Effects of chronic low-dose ultraviolet B radiation on DNA damage and repair in mouse skin. Cancer Res 1999;59:2875-84. Miyamura Y, Coelho SG, Schlenz K et al. The deceptive nature of UVA tanning versus the modest protective effects of UVB tanning on human skin. Pigment Cell Melanoma Res 2011;24:136-47. Moan J, Porojnicu AC, Dahlback A, Setlow RB. Addressing the health benefits and risks, involving vitamin D or skin cancer, of increased sun exposure. Proc Natl Acad Sci U S A 2008;105:668-73. Moller P, Cooke MS, Collins A et al. Harmonising measurements of 8-oxo-7,8- dihydro-2'-deoxyguanosine in cellular DNA and urine. Free Radic Res 2012;46:541- 53. Montecucco F, Di Marzo F Montecucco, V Di Marzo. At the heart of the matter: 250 the endocannabinoid system in cardiovascular function and dysfunction. Trends Pharmacol Sci. 2012;33:331-40. Mori T, Nakane M, Hattori T et al. Simultaneous establishment of monoclonal antibodies specific for either cyclobutane pyrimidine dimer or (6-4) photoproduct from the same mouse immunized with ultraviolet-irradiated DNA. Photochem Photobiol 1991;54:225-32. Mosher CE & Danoff-Burg S. Addiction to indoor tanning: relation to anxiety, depression, and substance use. Arch Dermatol 2010;146:412-7. Mouchet N, Adamski H, Bouvet R et al. In vivo identification of solar radiation-responsive gene network: role of the p38 stress-dependent kinase. PLoS ONE. 2010;5:e10776. Mouret S, Baudouin C, Charveron M et al. Cyclobutane pyrimidine dimers are predominant DNA lesions in whole human skin exposed to UVA radiation. Proc Natl Acad Sci U S A 2006;103:13765-70. Munro S, Thomas KL, Abu-shaar M. Molecular characterization of a peripheral receptor for cannabinoids. Nature 1993;365:61-65. Nathan C. Neutrophils and immunity: challenges and opportunities. Nat Rev Immunol 2006;6:173-82. Neenan A, Lea CS, Lesesky EB. Reasons for tanning bed use: a survey of community college students in North Carolina. N C Med J 2012;73:89-92. Nicolaou A. Eicosanoids in skin inflammation. Prost Leuk Essen Fatty Acids 2013;88:131-138. Nilsen LT, Hannevik M, Veierød MB. Ultraviolet exposure from indoor tanning devices: a systematic review. Br J Dermatol 2016;174:730-40. Nolan BV, Taylor SL, Liguori A et al. Tanning as an addictive behavior: a literature review. Photodermatol Photoimmunol Photomed 2009;25:12-9. Norris PG, Barker JN, Allen MH et al. Adhesion molecule expression in polymorphic light eruption. J Invest Dermatol 1992;99:504-8. NRPB. Health Effects from Ultraviolet Radiation: Report of an Advisory Group on Non-Ionising Radiation. Documents of the National Radiological Protection Board 2002;13:207-17. Ouhtit A, Muller HK, Davis DW et al. Temporal events in skin injury and the early adaptive responses in ultraviolet-irradiated mouse skin. Am J Pathol. 2000;156:201-7. Ozkan B, Hatun S, Bereket A. Vitamin D intoxication. Turk J Pediatr 2012;54:93-8. Paradisi A, Pasquariello N, Barcaroli D, Maccarrone M. Anandamide regulates keratinocyte differentiation by inducing DNA methylation in a CB1 receptor- dependent manner. J Biol Chem 2008;283:6005-12. Parrish KA, Jaenicke KF, Anderson R. Erythema and melanogenesis action spectra of normal human skin. Photochem Photobiol 1982;36:187-91. Patterson JW (2014). The spongiotic reaction pattern. In: Weedon’s skin pathology Churchill Livingstone. Chapter 5:pp104. Paul CF, Ho VC, McGeown C et al. Risk of malignancies in psoriasis patients treated with cyclosporine: a 5 y cohort study. J Invest Dermatol 2003;120:211-6. Pavón FJ, Araos P, Pastor A et al. Evaluation of plasma-free endocannabinoids and their congeners in abstinent cocaine addicts seeking outpatient treatment: impact of psychiatric co-morbidity. Addict Biol 2013;18:955-69. Pearse AD, Gaskell SA, Marks R. Epidermal changes in human skin following irradiation with either UVB or UVA. J Invest Dermatol 1987;88:83-7. Pelle E, Huang X, Mammone T et al. Ultraviolet-B-induced oxidative DNA base 251 damage in primary normal human epidermal keratinocytes and inhibition by a hydroxyl radical scavenger. J Invest Dermatol 2003;121:177-83. Pertwee RG. The pharmacology of cannabinoid receptors and their ligands: an overview. Int J Obes 2006;(Lond) 30: S13-8. Petersen B, Wulf HC, Triguero-mas M, et al. Sun and Ski Holidays Improve Vitamin D Status, but Are Associated with High Levels of DNA Damage. J Invest Dermatol. 2014;134:2806-13. Pfeifer GP & Besaratinia A. UV wavelength-dependent DNA damage and human non-melanoma and melanoma skin cancer. Photochem Photobiol 2008;11:90. Pierard-Franchimont C, Pierard GE, Quatresooz P. Focus on skin cancer association and progression under TNF antagonist therapy. Expert Opin Biol Ther 2011;11:1215-22. Piskin G, Bos JD & Teunissen MB. Neutrophils infiltrating ultraviolet B- irradiated normal human skin display high IL-10 expression. Arch. Dermatol. Res 2005;296:339–342. Podmore ID, Cooke MS, Herbert KE et al. Quantitative Determination of Cyclobutane Thymine Dimers in DNA by Stable Isotope-Dilution Mass Spectrometry. Photochem Photobiol 1996;64:310-315. Pouplard C, Brenaut E, Horreau C et al. Risk of cancer in psoriasis: a systematic review and meta-analysis of epidemiological studies. J Eur Acad Dermatol Venereol 2013;27 Suppl 3:36-46. Public Health England. Vitamin D advice. https://www.gov.uk/government/news/phe-publishes-new-advice-on- vitamin-d. Accessed 24th August 2016 Pucci M, Pasquariello N, Battista N et al. Endocannabinoids stimulate human melanogenesis via type-1 cannabinoid receptor. J Biol Chem 2012;287:15466-78. Raichlen DA, Foster AD, Gerdeman GL et al Wired to run: exercise-induced endocannabinoid signaling in humans and cursorial mammals with implications for the 'runner's high'. J Exp Biol 2012;215(Pt 8):1331-6. Raichlen DA, Foster AD, Seillier A et al. Exercise-induced endocannabinoid signaling is modulated by intensity. Eur J Appl Physiol 2013;113:869-75. Reichrath J, Reichrath S. Hope and challenge: the importance of ultraviolet (UV) radiation for cutaneous vitamin D synthesis and skin cancer. Scand J Clin Lab Invest Suppl 2012;243:112-9. Rhodes LE, Belgi G, Parslew R et al. Ultraviolet-B-induced erythema is mediated by nitric oxide and prostaglandin E2 in combination. J Invest Dermatol 2001;117:880-5. Rhodes LE, Gledhill K, Masoodi M et al. The sunburn response in human skin is characterized by sequential eicosanoid profiles that may mediate its early and late phases. FASEB J 2009;23:3947-56. Rhodes LE, Webb AR, Fraser HI et al. Recommended summer sunlight exposure levels can produce sufficient (≥20 ng ml-1) but not the proposed optimal (≥32 ng ml-1) 25(OH)D levels at UK latitudes. J Invest Dermatol 2010;130:1411-8. Rigel DS, Cockerell CJ, Carucci J, Wharton J (2008). Neoplasms of the skin. In: Dermatology. (Bolognia JL, Rapini RP Eds), Mosby, Elsevier. 2:1641-1660. Rijken F, Kiekens RC, Bruijnzeel PL. Skin-infiltrating neutrophils following exposure to solar-simulated radiation could play an important role in photoageing of human skin. Br J Dermatol 2005;152:321-8. Rivas JM & Ullrich SE. Systemic suppression of delayed-type hypersensitivity by supernatants from UV-irradiated keratinocytes. An essential 252 role for keratinocyte-derived IL-10. J Immunol 1992;149:3865-71. Rizwan M, Rodriguez-blanco I, Harbottle A et al. Tomato paste rich in lycopene protects against cutaneous photodamage in humans in vivo: a randomized controlled trial. Br J Dermatol 2011;164:154-62. Robertson AR. The CIE 1976 color difference formulas. Color Res Appl 1977;2:7-11. Robinson JK, Kim J, Rosenbaum S, Ortiz S. Indoor tanning knowledge, attitudes, and behavior among young adults from 1988-2007. Arch Dermatol 2008;144:484-8. Rosen CF, Seki Y, Farinelli W, et al. A comparison of the melanocyte response to narrow band UVA and UVB exposure in vivo. J Invest Dermatol. 1987;88:774-9. Scientific Advisory Committee on Nutrition, Vitamin D and Health, 2016. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file /537616/SACN_Vitamin_D_and_Health_report.pdf Accessed 22nd August 2016. Salehin D, Haugk C, Thill M et al. Vitamin D receptor expression in patients with vulvar cancer. Anticancer Res 2012;32:283-9. Schneider S, Diehl K, Bock C et al. Sunbed use, user characteristics, and motivations for tanning: results from the German population-based SUN-Study 2012. JAMA Dermatol 149:43-9. Schwarz T. Regulatory T cells induced by ultraviolet radiation. Int Arch Allergy Immunol 2005;137:187-93. Schwarz T. 25 years of UV-induced immunosuppression mediated by T cells-from disregarded T suppressor cells to highly respected regulatory T cells. Photochem Photobiol 2008;84:10-8. Schwarz T & Schwarz A. Molecular mechanisms of ultraviolet radiation- induced immunosuppression. Eur J Cell Biol 2011;90:560-4. Setlow RB. The wavelengths in sunlight effective in producing skin cancer: a theoretical analysis. Proc Natl Acad Sci USA 1974;71:3363–6. Shapiro SD. Neutrophil elastase: Path clearer, pathogen killer, or just pathologic? Am J Respir Cell Mol Biol 2002;26:266–268. Sheehan JM, Cragg N, Chadwick CA et al. Repeated ultraviolet exposure affords the same protection against DNA photodamage and erythema in human skin types II and IV but is associated with faster DNA repair in skin type IV. J Invest Dermatol 2002; 118:825-9. Shibutani S, Gentles R, Johnson F, Grollman AP. Isolation and characterization of oligodeoxynucleotides containing dG-N2-AAF and oxidation products of dG-C8-AF. Carcinogenesis 1991; 12:813-8. Simon JC, Cruz PD Jr., Bergstresser PR, Tigelaar RE. Low dose ultraviolet B- irradiated Langerhans cells preferentially activate CD4+ cells of the T helper 2 subset. J Immunol 1990;145:2087-91. Sinclair CA, Makin JK, Tang A, Brozek I, Rock V. The role of public health advocacy in achieving an outright ban on commercial tanning beds in Australia. Am J Public Health 2014;104:e7-9. Skin Cancer Foundation (2013). “Skin Cancer Information”. Public Information Retrieved 12/02/13, 2012, from http://www.skincancer.org Skowronski RJ, Peehl DM, Feldman D. Vitamin D and prostate cancer: 1,25 dihydroxyvitamin D3 receptors and actions in human prostate cancer cell lines. Endocrinology 1993;132:1952-60. Sommer DM, Jenisch S, Suchan M et al. Increased prevalence of the 253 metabolic syndrome in patients with moderate to severe psoriasis. Arch Dermatol Res 2006;298:321-8. PB Sparling, A Giuffrida, D Piomelli, L Rosskopf, A Dietrich. Exercise activates the endocannabinoid system. Neuroreport 2003;14:2209-11. Springbett P, Buglass S, Young AR. Photoprotection and vitamin D status. J Photochem Photobiol B. 2010;101:160-8. Strickland I, Rhodes LE, Flanagan BF, Friedmann PS. TNF-alpha and IL-8 are upregulated in the epidermis of normal human skin after UVB exposure: correlation with neutrophil accumulation and E-selectin expression. J Invest Dermatol 1997;108:763-8. Su J, Pearce DJ, Feldman SR. The role of commercial tanning beds and ultraviolet A light in the treatment of psoriasis. J Dermatolog Treat 2005;16:324-6. Sugawara K, Biro T, Tsuruta D, et al. Endocannabinoids limit excessive mast cell maturation and activation in human skin. J Allergy Clin Immunol 2012;129:726- 738. Sunsmart. Cancer Research UK. 2014. Public Information: UV, “Sun, UV and Cancer”, Retrieved from http://www.sunsmart.org.uk 02/11/2015. Sutherland BM, Cimino JS, Delihas N et al. Ultraviolet light-induced transformation of human cells to anchorage-independent growth. Cancer Res 1980;40:1934-9. Takeuchi H, Gomi T, Shishido M et al. Neutrophil elastase contributes to extracellular matrix damage induced by chronic low-dose UV irradiation in a hairless mouse photoaging model. J Dermatol Sci 2010;60:151-8. Taub DD, Proost P, Murphy WJ et al. Monocyte chemotactic protein-1 (MCP- 1), -2, and -3 are chemotactic for human T lymphocytes. J Clin Invest 1995;95:1370- 6. Taylor DK, Anstey AV, Coleman AJ et al. Guidelines for dosimetry and calibration in ultraviolet radiation therapy: a report of a British Photodermatology Group workshop. Br J Dermatol 2002;146:755–63. Teunissen MB, Piskin G, Di Nuzzo S et al. Ultraviolet B radiation induces a transient appearance of IL-4+ neutrophils, which support the development of Th2 responses. J Immunol 2002;168:3732-9. Tewari A, Sarkany RP, Young AR. UVA1 induces cyclobutane pyrimidine dimers but not 6-4 photoproducts in human skin in vivo. J Invest Dermatol 2012;132:394-400. Thong HY, Jee SJ, Sun CC, Boissy RE. The patterns of melanosome distribution in keratinocytes of human skin as one determining factor of skin colour. Br J Dermatol 2003;149:498-505. Thorne J & Campbell MJ. The vitamin D receptor in cancer. Proc Nutr Soc 2008; 67:115-27. Tian XQ, Chen TC, Matsuoka LY et al. Kinetic and thermodynamic studies of the conversion of previtamin D3 to vitamin D3 in human skin. J Biol Chem 1993;268:14888-92. Tian XQ & Holick MF. A liposomal model that mimics the cutaneous production of vitamin D3. Studies of the mechanism of the membrane-enhanced thermal isomerization of previtamin D3 to vitamin D3. J Biol Chem 1999;274:4174- 9. Turcotte C, Chouinard F, Lefebvre JS, Flamand N. Regulation of inflammation by cannabinoids, the endocannabinoids 2-arachidonoyl-glycerol and arachidonoyl-ethanolamide, and their metabolites. J Leukoc Biol. 2015;97:1049-70. Turner RJ, Farr PM, Walshaw D. Many patients with psoriasis use sunbeds. 254 BMJ 1998;317(7155):412. Urquhart P, Nicolaou A, Woodward DF. Endocannabinoids and their oxygenation by cyclo-oxygenases, lipoxygenses and other oxygenase. Biochim Biophys Acta 2015;1851:366-376. van Etten & Mathieu C. Immunoregulation by 1,25-dihydroxyvitamin D3: basic concepts. J Steroid Biochem Mol Biol 2005;97:93-101. van Smeden J, Hoppel L, van der Heijden R et al. LC/MS analysis of stratum corneum lipids: ceramide profiling and discovery. J Lipid Res 2011; 52: 1211–21. Vieth R. What is the optimal vitamin D status for health? Prog Biophys Mol Biol 2006;92:26-32. Walsh LJ, Trinchieri G, Waldorf HA et al. Human dermal mast cells contain and release tumor necrosis factor alpha, which induces endothelial leukocyte adhesion molecule 1. Proc Natl Acad Sci U S A 1991;88:4220-4. Wang SY. Photochemical reactions in frozen solutions. Nature 1961; 190:690-694. Wang J, Ueda N. Biology of endocannabinoid synthesis system. Prostaglandins Other Lipid Mediat 2009; 89:112–9. Warthan M, Uchida MT, Wagner RF Jr. UV light tanning as a type of substance-related disorder. Arch Dermatol 2005;141:963-6. Webb AR & Holick MF. The role of sunlight in the cutaneous production of vitamin D3. Annu Rev Nutr 1988;8:375-99. Webb AR, DeCosta BR, Holick MF. Sunlight regulates the cutaneous production of vitamin D3 by causing its photodegradation. J Clin Endocrinol Metab 1989;68:882-7. Webb AR, Engelesen O. Calculated ultraviolet exposure levels for a healthy vitamin D status. Photochem Photobiol 2006;82:1697–703. Webb AR, Kift R, Berry JL, Rhodes LE. The vitamin D debate: translating controlled experiments into reality for human sun exposure times. Photochem Photobiol 2011; 87:741–5. Wehner MR, Chren MM, Nameth D et al. International prevalence of indoor tanning: a systematic review and meta-analysis. JAMA Dermatol 2014;150:390- 400. Wilkins CH, Sheline YI, Roe CM, Birge SJ, Morris JC. Vitamin D deficiency is associated with low mood and worse cognitive performance in older adults. Am J Geriatr Psychiatry 2006;14:1032-40. Wilkinson KD, Williamson EM. Cannabinoids inhibit human keratinocyte proliferation through a non-CB1/CB2 mechanism and have a potential therapeutic value in the treatment of psoriasis. J Dermatol Sci 2007;45:87-92. Willis I, Cylus L. UVA erythema in skin: is it a sunburn? J Invest Dermatol 1977;68:128-9. Wintzen M, Ostijn DM, Polderman MC et al. Total body exposure to ultraviolet radiation does not influence plasma levels of immunoreactive beta- endorphin in man. Photodermatol Photoimmunol Photomed 2001;17:256-60. Wolber R, Schlenz K, Wakamatsu K et al. Pigmentation effects of solar- simulated radiation as compared with UVA and UVB radiation. Pigment Cell Melanoma Res 2008;21:487-91. World Health Organization. Vitamin and mineral requirements in human nutrition, 2nd edition 2004. World Health Organization, Food and Agricultural Organization of the United Nations. Eds World Health Organization. World Health Organization (2012). "Ultraviolet Radiation and the INTERSUN programme." Retrieved from 255 http://www.who.int/uv/intersunprogramme/en/ 02/11/2015 Wikonkal NM & Brash DE. Ultraviolet radiation induced signature mutations in photocarcinogenesis. J Investig Dermatol Symp Proc 1999;4:6-10. Wilkins CH, Sheline YI, Roe CM et al. Vitamin D deficiency is associated with low mood and worse cognitive performance in older adults. Am J Geriatr Psychiatry 2006;14:1032-40. Willis I, Cylus L. UVA erythema in skin: is it a sunburn? J Invest Dermatol. 1977;68:128-9. Wintzen M, Ostijn DM, Polderman MC et al. Total body exposure to ultraviolet radiation does not influence plasma levels of immunoreactive beta- endorphin in man. Photodermatol Photoimmunol Photomed 2001;17:256-60. Wolber R, Schlenz K, Wakamatsu K et al. Pigmentation effects of solar- simulated radiation as compared with UVA and UVB radiation. Pigment Cell Melanoma Res. 2008;21:487-91. Yoon JH, Lee CS, O’Connor TR et al. The DNA damage spectrum produced by simulated sunlight. J Mol Biol 2000;299:681-93. Young AR, Chadwick CA, Harrison GI et al. The in situ repair kinetics of epidermal thymine dimers and 6-4 photoproducts in human skin types I and II. J Invest Dermatol 1996;106:1307–13. Young AR, Chadwick CA, Harrison GI et al. The similarity of action spectra for thymine dimers in human epidermis and erythema suggests that DNA is the chromophore for erythema. J Invest Dermatol 1998;111:982-8. Zheng D, Bode AM, Zhao Q, et al. The cannabinoid receptors are required for ultraviolet-induced inflammation and skin cancer development. Cancer Res 2008; 68:3992–8. Ziegelberger G, Repacholi M, McKinlay A. International commission on non- ionizing radiation protection. Prog Biophys Mol Biol 2006;92:1–3. Ziegler A, Jonason AS, Leffell DJ et al. Sunburn and p53 in the onset of skin cancer. Nature 1994;372:773-6. Zittermann A. Vitamin D in preventive medicine: are we ignoring the evidence? Br J Nutr 2003;89:552-72. 256 Appendix 1: Case Report Form Subject No ………….. Risk versus benefits of simulated sunlight exposure CASE REPORT FORM For clinical research Subject Number ……………. Subject Code ………………… Investigator: Researcher: LREC ref no: 09/H1014/73 257 Subject No ……. Visit 1 Date: DOB: ……….. Male/Female Height Weight BMI PMH/PSH/Allergies: History of Photosensitivity/Rashes: AutoAb (ANA, ENA) test needed? Medications (systemic/topical): Skin type: Eye/hair colour: Information sheet read and understood: Informed consent given/consent form signed: GP letter sent: 258 Subject No …………. Date: Minimal Erythemal Dose (MED) test (UV6 lamp): Warm up period of lamp: Exposure time of lamp: Irradiance of lamp at start of exposure: Irradiance of lamp at end of exposure: Skin carefully marked up: Participant asked to keep area dry: MED to be assessed 24 hrs later at next appointment: Sign ………………………………… Date ………………………………… 259 Subject No………….. Week 0, Visit 2 Date ………………… 1. Visual MED test result (the lowest UV dose at which erythema is definitely visible) MED = ……….. mJ/cm2 2. Are they able to bring their own suitable T-shirt? 3. Have shorts been cut to size? 4. Explain intervention study patient information sheet. 5. Give diet diary Sign …………………………. Date …………………………. 260 Subject No ………….. Week 1 Day Date Bloods Other Time Erythema given /other and actions taken Monday Vit D Urine Biochem Tanning PTH Give Endo Diet diary Tuesday Urine Wednesday Endo Urine Thursday Urine Friday Endo Urine Tanning Week 2 Day Date Bloods Other Time Erythema given /other and actions taken Monday Vit D Collect Endo Diet diary Wednesday Friday Urine Tanning 261 Week 3 Day Date Bloods Other Time Erythema/other and actions given taken Monday Vit D Endo Wednesday Friday Tanning Urine Week 4 Day Date Bloods Other Time Erythema/other and actions given taken Monday Vit D Endo Wednesday Friday Tanning Urine Week 5 Day Date Bloods Other Time Erythema/other given and actions taken Monday Vit D Endo Wednesday Friday Tanning Urine 262 Week 6 Day Date Bloods Other Time Erythema/other given and actions taken Monday Vit D Give Diet diary Endo Wednesday Friday Vit D Urine PTH Collect diet diary Tanning Uncover new area of buttock skin for one UV exposure Prepare phototype II for 2X MED 263 Subject No …………… Date …………….. (Week 6) 1. Collect diet diary 2. Take bloods 3. Collect urine 4. Assess tanning 5. Uncover new area of (previously photoprotected) buttock skin to receive one UV exposure pre-biopsy. 6. Skin biopsies x 3 (One exposure, cumulative exposure; control- no exposure): Biopsy consent form signed? 7. Phototype II only: a. What was their MED? ………… b. What is their 2X MED time? ………….. c. Expose two areas to 2X MED: Cumulative site …….. Photoprotected site ……. 8. All: Arrange time for next-day biopsies i.e. exactly 24 hours after UV exposure, and final endocannabinoid sampling Date …………….. 1. All: Skin biopsy x 1 (cumulative exposure- 24hrs later): confirm consent. Blood sampling Phototype II: Skin biopsy from 2X MED sites: Cumulative and photoprotected 2. Thank you and discuss reimbursement and suture removal Sign ………………… 264 Appendix 2: Diet Diary Subject No:……………….. Date:……………….. While we are monitoring your urine we would also like to know how much vitamin D you are getting from your diet. To help us assess this please complete this food diary for the days when you are completing the urine study. Please tick where appropriate. 1. Do you take supplements that contain vitamin D? (E.g. multivitamin, fish oils) Yes No You should not be taking supplements before or during the study as discussed during your Sunburn threshold assessment. Please do not start taking supplements during the study. 2. If the answer above is yes, please give us details of the supplement, and how much vitamin D it contains (if you know) ………………………………………………………………………………………………… 3. Are you a vegetarian? Yes No 4. Are you a vegan? Yes No 5. What spread do you usually use on bread / vegetables? None Butter (E.g. Country life) …………………………. Other spread …………………………. (E.g. Utterly Butterly / Flora Light) Are they Vitamin D fortified? Yes No Please give details ………………………………………………………… 6. What type of milk do you drink or use? Whole milk (full fat) Semi-skimmed Skimmed Vitamin D fortified (Please give brand) ……………………………………………………………. 265 Here is an example of how to fill in your diet diary: • Within the table overleaf fill in as much information as possible. Please fill in the amount of certain foods you have eaten each day. You can fill it in by meal (sometimes easiest to remember) or simply as a total for the whole day. • Please estimate portions, use common measures to help you (e.g. teaspoon (tsp), tablespoon (tbsp), cup, small bowl, large bowl, you can estimate how much of a full pack or tin you ate e.g. half can tinned tuna or 92.5g use other dimensions e.g. steak = 2” x 4” x 1” (or 5cm x 10 cm x 2.5 cm). You can use any units that you are familiar with (lb, oz, g, kg, pint, litre, ml, inches and cm). • We ONLY need to know when you ate food containing vitamin D, or vitamin D fortified (added) and also portions of fruit and vegetables NOT everything you ate during the day. • If you did not eat that category of food then just leave it blank. Explanation of example over leaf, please use a page per day. During the day our volunteer ate: • Breakfast; A bowl of cereal with milk and took a multivitamin at breakfast. • Lunch; A tuna sandwich with thick filling and butter on brown bread Milk was used in tea and coffee throughout the day. • Dinner; Steak with new potatoes, butter and peas. We have given the above as an example, we do not expect you to fill in every box every day. We want you to eat your normal diet, without changing anything to “look good” in the diary. The results will be anonymous. 266 Day 1 Food Product make Quantity date: Fortified foods - Vitamin D 30g (1 or omega 3 added. Tesco own brand medium (e.g. some cereal, yoghurt, fruity flakes cereal sized juice) bowl.) Red Meat 2” x 4” x Steak (e.g. Beef, lamb, pork, liver) 1” Eggs (e.g. whole boiled, fried, omelette or quiche) Total milk used in the day Semi skimmed milk ¾ pint (inc drinks, sauces) Total fat spread used for the Lurgold Butter 2 tbsp day Supplements containing Multivitamin from vitamin D (e.g. multivitamin, my health 400iu fish oil/ omega 3) multivitamin Fatty fish (e.g. Sardines, mackerel, Tinned Tuna in Half a tuna, herring, salmon – Brine Tesco tin or specify if tinned, farmed or own brand 92.5g wild) 267 Day 1 Food Product make Quantity date: Fortified foods - Vitamin D or omega 3 added. (e.g. cereal, yoghurt, juice) Red Meat- (e.g. Beef, lamb, pork, liver) Cheese (e.g. Swiss cheese) Eggs (e.g. whole boiled, fried, omelette or quiche) Total milk used in the day (inc drinks, sauces) Total fat spread used for the day Supplements containing vitamin D (e.g. multivitamin, fish oil/ omega 3) Fatty fish (e.g. Sardines, mackerel, tuna, herring, salmon – specify if tinned, farmed or wild.) 268 Appendix 3: Simplified Psoriasis Index Taken from Chularojanamontri et al., 2013. 269 Appendix 4: Research letter publication 270 271 272