With the interface between the body and the environment, the functions of the skin may arguably be regarded as adaptations to the different environments that our ancestors were exposed to, shaping the geographical variation in skin characteristics, including pigmentation in modern humans. The apparent geographical variation in skin pigmentation has attracted both public and academic attention, and has been studied in many research areas such as evolutionary biology, clinical medicine, health science, physiology, and molecular genetics. Most studies on the genetic basis of human skin pigmentation, however, have focused on people of European ancestry, and only a few studies have focused on Asian populations. Therefore, this study investigates the association of skin reflectance and freckling with genetic polymorphisms in the Japanese population.

DNA samples were obtained from a total of 653 Japanese individuals (ages 19-40) residing in Okinawa. Skin reflectance was measured using a spectrophotometer and freckling status was determined for each individual. A lightness index (L*) of the inner upper arm and freckling status were not correlated with age, BMI, or ancestry (Ryukyuans or Main Islanders of Japan).

In Chapter 1, the association of skin reflectance and freckling with polymorphisms of the Melanocortin 1 Receptor (MC1R) is examined. The MC1R is located on human chromosome 16q24.3, and it encodes the protein MC1R, a G protein-coupled seven transmembrane receptor for a-inelanocyte simulation hormone (a-MSH). When a-MSH binds to MC1R, subsequent activation ofadenylate cyclase facilitates accumulation of cAMP inside the melanocyte. This accumulation ultimately promotes eumelanin (dark pigment) synthesis. Therefore, MC1R activation plays a key role in skin pigmentation because it results in a switch from pheomelanin (light and immature pigment) synthesis to eumelanin synthesis. Among the 10 nonsynonymous variants that were identified by direct sequencing of the coding region of MC1R, two variants were most common —R163Q and V92M—with the derived allele frequencies of 78.6% and 5.5%, respectively. The others were observed at low frequencies (<5%) and hence, are defined as "rare variants." As shown in the largest circle with many branches in the haplotype network (Figure 1), the majority of the estimated haplotypes had the 163Q allele. Tajima's D was calculated as -1.66 with P-value of 0.015, which suggested a deviation from neutrality and the possibility of purifying or positive selection on MC1R.

Stepwise multiple regression analysis showed that the 163Q allele and the presence of nonsynonymous rare variants were significantly associated with an increase in sex-standardized skin lightness (L* of CIELAB) of the inner upper arm, designated ZuaL* (Figure 2). Relative to the 92V allele, the 92M allele was significantly associated with increased odds of freckling. R163Q was reported as an example of East Asian selective sweep, although its functional significance was then unknown. This study thus shows the effect of the 163 Q allele on skin reflectance values. This association indicates that light-toned skin may have been subject to positive selection in East Asian people.

Each circle represents a haplotype with its amino acid change(s) annotated by or in the circle. The size of each circle reflects frequency of the haplotype. Base changes are annotated along the lines connecting circles, which are broad for nonsynonymous substitution. The three haplotypes estimated with the probability of 0.5 with R163Q are indicated by dotted lines.

"163R+rare" or "163Q+rare" means the rare variant is located on the same haplotype as the 163R or. 163Q allele, respectively. "0" means no rare variant is present. The bottom of the box is the 25th percentile, the center line is the median, and the top of the box is the 756 percentile. P-values were determined using the Mann-Whitney U test.

In Chapter 2, single nucleotide polymorphisms (SNPs) that had significant associations with skin pigmentation and freckling were identified through the analyses in the discovery study (86 females and 209 males), in the replication study (203 females and 155 males), and in the total sample.

In the discovery study, SNPs were selected if they satisfied either or both of the following criteria: 1) associations of the SNPs with skin pigmentation in people of European descent descent were reported, 2) positive selective sweep on the SNPs or its adjacent region in the sample of East Asians were suggested by the previous studies in population genetics. A total of 59 SNPs were genotyped using TaqMan assays, a DigiTag2 assay, or direct sequencing. Two SNPs in Oculocutaneous Albinism II (OCA2), rs12442916 (rs: reference SNP number from dbSNP, NCBI), and rs7465330 had P.-values smaller than 0.01for the correlation with ZuaL*, and hence were further examined in the replication study together with a,nonsynonymous polymorphism (rs1800414) located near these two SNPs. Only rs6058017 in Agouti Signaling Protein (ASIP) showed a significant (P<0.05) association with freckling, and the ancestral allele had significantly higher odds for freckling relative to the derived allele.

In the replication study, the correlations of the three SNPs in OCA2 with ZuaL* were confirmed. After merging the discovery and the replication studies, stepwise multiple regression analyses showed that rs7465330 and rs1800414 had significant associations with ZuaL* and suggested that the derived alleles would increase the lightness in an additive manner. A haplotype network of these three SNPs (Figure 3) showed that the derived alleles of rs74653330 (A481 T) and of rs1800414 (H615R) had occurred on the derived allele of rs12442916 (g.28233207G>A), and they do not share any haplotype except for a low-frequency haplotype observed in only one chromosome. This explained the apparent association of rs12442916 and skin reflectance, and confirmed the independent effects of A481T and H615R on the lightness of skin among Japanese people (Figure 4). OCA2 has been reported as one of the genes that had signatures of positive selection in Asia, and therefore, the results of this chapter also suggest that the light-toned skin was subject to selection in East Asia.

To summarize, the 163Q allele of MC1R and the 481T and 615R alleles of OCA2 were associated with the lightness of skin reflectance, and the 92M allele of MC1R and the ancestral allele of rs6058017 in ASIP showed higher odds for freckling. These results explain part of the difference in constitutive skin pigmentation between Asian and non-Asian populations as well as the north to south gradient in pigmentation across Asia. Association studies on admixed populations or comparative studies in allele frequencies among different populations are required to further elucidate the evolution of skin pigmentation among peoples of Asia.

Figure 1 MC1R haplotype network based on the estimated haplotypes of 1306 chromosomes evaluated in this study.

Figure 2 Boxplot of L* (lightness) categorized by R163Q genotype and clustered by rare haplotypes.

Figure 3 OCA2 haplotype network based on the estimated haplotypes of 1304 chromosomes evaluated in this study.

Figure 4 Boxplot of L* (lightness) categorized by R163Q genotype and clustered by rare haplotypes.

ヒトの皮膚色でみられる地理的変異は、様々な分野で研究の対象となってきた。特に進化人類学では、皮膚色の地理的変異の起源を、主として異なる環境への適応として説明してきた。すなわち、暗い皮膚色は、低緯度地域において過剰な紫外線に対する防御の役割を果たす一方で、高緯度地域においてはビタミンD合成を阻害するなどの有害な効果をもたらすため、紫外線照射量の違いに応じた皮膚色の地理的勾配が、自然淘汰によって生じたのだとされる。近年の分子遺伝学的研究は、ヒト皮膚色の多様性の遺伝学的基盤を徐々に明らかにしつつあり、この仮説の検証につながるものと期待される。しかしながら、これらの研究の多くはヨーロッパ由来の集団を対象としたものであり、アジア集団を対象としたものは依然として少数である。また、従来の研究には、被験者の皮膚色のデータを質問紙や色見本による分類に基づいて収集しているものが少なくなく、より客観的な皮膚色測定の方法が望まれる。本論文では、東アジア集団における皮膚形質と遺伝的多型との関連を調べることを目的として、沖縄在住の日本人653人からDNA標本を得るとともに、各被験者に対して、分光光度計による皮膚反射率の測定、および目視によるそばかすの有無の判定を行つた。

本論文は2章からなる。第1章では、日本人集団において、メラノコルチン1受容体(MC1R;melanocortin 1 receptor)遺伝子の多型と、皮膚形質との関連を検証した。MC1Rはメラノサイトにおけるメラニン合成に関与するタンパク質であり、それをコードする遺伝子、MC1Rにおける変異は、皮膚色に影響を与えることが知られている。また、MC1Rのコード領域はアプリカ集団では高度に保存されている一方で、ヨーロッパ起源の集団では多くの同義・非同義置換を含むことが報告されている。この知見は、アフリカの低緯度地域ではMC1Rに純化淘汰が作用しているのに対し、ヨーロッパではこの淘汰が緩和されているという仮説と合致する。本論文第1章では、日本人標本で、MC1Rのコード領域のDNA配列を決定し、10個の非同義変異の存在を確認した。これらのうち、R163QとV92Mの2つの1塩基多型(SNPs; single nucleotide polymorphisms)については、派生型アリルの頻度が、それぞれ78.6%、5.5%であった。また、他の8個のSNPsについては、派生型アリルの頻度がそれぞれ5%未満であったため、まれな変異として一括して扱った。先行研究によリヨーロッパ由来の集団において皮膚色との強い関連が報告されているアリルは、日本人標本では確認されなかった。重回帰分析により、R163Qの派生型アリルをもつこと、およびまれな変異の派生型アリルをもつことには、分光光度計で測定された上腕内側の皮膚色を、明るくする効果があることが示唆された。一方、V92Mの派生型アリルをもつことと、そばかすがあることとの間には有意な関連が見出された。MC1Rに関しては、東アジアにおいても最近の急速な自然淘汰の痕跡が報告されており、第1章の結果は、より明るい皮膚色が有利となるような自然淘決が東アジアで起こり、これによってR163Qの派生型アリルの頻度が増加したという可能性を提起する。

第2章では、上記の日本人標本をdiscovery sample(295人)とreplication sample(358人)に分け、皮膚形質と関連するSNPsの探索を行った。Discovery sampleでは、ヨーロッパ由来の集団において皮膚色との関連が示唆されているSNPs、および東アジア集団において最近の急速な自然淘汰の痕跡が報告されているか、そのような報告のある領域の近傍に位置するSNPs(合計57個)について、皮膚形質との関連を分析した。Replication sampleでは、discovery sampleでの分析の結果を踏まえ、皮膚形質と関連するSNPsの候補を絞り込んだ。これらの作業を通じて、OCA2(Oculocutaneous albinism II)遺伝子内の2つのSNPs(A418T、H615R)と上腕内側の皮膚色との間tおよびOCA2の10kb上流に位置するHERC2遺伝子内のSNP(rs916977)と額の皮膚色との間に、それぞれ関連が示唆された。OCA2内の2つのSNPsの派生型アリルは、相互に独立に皮膚色を明るくする効果をもつこと、また、HERC2内のSNPの派生型アリルは、皮膚色を暗くする効果をもつことが、それぞれ推定された。OCA2は、その産物がメラニン合成に関与し、眼皮膚白皮症II型の原因遺伝子として知られている。HERC2-OCA2領域について、アジアにおいて正の自然淘汰が働いてきた可能性が指摘されており、第2章の結果は、東アジアにおいて明るい皮膚色の進化が自然淘汰によって促進されたという仮説を支持する。

なお、本論文第1章は、渡邊千晶、川口亮、佐藤丈寛、中伊津美、進藤美咲、諸見里恵一、青木健一、石田肇、本村亮介との共同研究であり、また第2章は、凍邊千晶、川口亮、佐藤丈寛、進藤美咲、諸見里恵一、提嶋恵美、西田奈央、石田肇、本村亮介との共同研究であるが、いずれも論文提出者が主体となって分析及び検証を行つたもので、論文提出者の寄与が十分であると判断する。

以上より、博士(理学)の学位を授与できると認める。