Professional Documents
Culture Documents
Longo 2013
Longo 2013
review article
mechanisms of disease
Dan L. Longo, M.D., Editor
Uterine Fibroids
Serdar E. Bulun, M.D.
U
From the Department of Obstetrics and terine fibroids (leiomyomas) represent the most common tumor
Gynecology, Feinberg School of Medi- in women. These lesions disrupt the functions of the uterus and cause ex-
cine, Northwestern University, Chicago.
Address reprint requests to Dr. Bulun at cessive uterine bleeding, anemia, defective implantation of an embryo, recur-
Prentice Women’s Hospital, 250 E. Supe- rent pregnancy loss, preterm labor, obstruction of labor, pelvic discomfort, and uri-
rior St., Ste. 03-2306, Chicago, IL 60611, nary incontinence and may mimic or mask malignant tumors. By the time they reach
or at s-bulun@northwestern.edu.
50 years of age, nearly 70% of white women and more than 80% of black women
N Engl J Med 2013;369:1344-55. will have had at least one fibroid; severe symptoms develop in 15 to 30% of these
DOI: 10.1056/NEJMra1209993
Copyright © 2013 Massachusetts Medical Society women.1,2 Uterine fibroids in black women are significantly larger at diagnosis
than those in white women, are diagnosed at an earlier age, and are characterized
by more severe symptoms and a longer period of sustained growth.3-5 Approximately
200,000 hysterectomies, 30,000 myomectomies, and thousands of selective uterine-
artery embolizations and high-intensity focused ultrasound procedures are performed
annually in the United States to remove or destroy uterine fibroids. The annual eco-
nomic burden of these tumors is estimated to be between $5.9 billion and $34.4 billion.6
There may be one predominant uterine fibroid or a cluster of many fibroids
(Fig. 1). Very large fibroids can cause the uterus to expand to the size reached at
6 or 7 months of pregnancy. The location and size of the fibroid in the uterus are
critical determinants of its clinical manifestations. As compared with other fi-
broids, submucous fibroids that extend into the uterine cavity are the most disrup-
tive to endometrial integrity, implantation, and the capacity of the myometrium to
contract and stop menstrual bleeding from the endometrial blood vessels; thus,
even small submucous fibroids are associated with excessive or irregular bleeding,
infertility, and recurrent pregnancy loss. In contrast, subserous fibroids that grow
out into the peritoneal cavity can exert pressure that is sensed by the patient as
pelvic discomfort only if they reach a certain size. Intramural fibroids that reside
in the myometrial wall represent an intermediary group. Regardless of their size
or location, fibroids may have paracrine molecular effects on the adjacent endo-
metrium that are extensive enough to cause excessive uterine bleeding or defective
implantation.7
Uterine fibroids are monoclonal tumors that arise from the uterine smooth-
muscle tissue (i.e., the myometrium).8 Histologically, fibroids are benign neo-
plasms composed of disordered smooth-muscle cells buried in abundant quanti-
ties of extracellular matrix (Fig. 1). The cells proliferate in vivo at a modest rate.
Formation of the extracellular matrix also accounts for a substantial portion of
tumor expansion. Uterine fibroids are almost always benign.9
A striking feature of uterine fibroids is their dependency on the ovarian steroids
estrogen and progesterone.10 Ovarian activity is essential for fibroid growth, and
most fibroids shrink after menopause. The sharp elevations and declines in the
production of estrogen and progesterone that are associated with very early preg-
nancy and the postpartum period have a dramatic effect on fibroid growth.11-13
Gonadotropin-releasing-hormone (GnRH) analogues, which suppress ovarian ac-
tivity and reduce circulating levels of estrogen and progesterone, shrink fibroids
and reduce associated uterine bleeding.14
Fibroid
Obliterated
endometrial
cavity
Cervical
canal
ECM
50.0 µm 50.0 µm
A limited number of genetic defects transmit- other molecular defects involving transcription-
ted by germ cells have been associated with fa- al and posttranscriptional events, microRNAs
milial uterine fibroid syndromes.15 Most notable (miRNAs), and signaling pathways have also
are germline mutations causing fumarate hydra- been described.21-28 Although some of the ef-
tase deficiency, which predisposes women to the fects of uterine fibroids on cell proliferation,
development of multiple uterine fibroids.16 In apoptosis, and extracellular matrix formation
addition, a variety of somatic chromosomal rear- have been identified, little is known about their
rangements have been described in up to 40% of effects on other cellular processes in fibroid
uterine fibroids.17 Recently, whole-genome se- growth, such as autophagy and senescence. This
quencing showed that chromosomal rearrange- review focuses on some recent developments in
ments are often complex, best described as sin- fibroid research, including the role of stem cells,
gle events consisting of multiple chromosomal somatic genetic and epigenetic defects, and the
breaks and random reassembly.18 In an earlier action of estrogen and progesterone and their
study, a somatic single-gene defect was found in cross-talk with various signaling pathways.
a majority of uterine fibroid tumors; this group
of mutations affects the gene encoding mediator CEL LUL A R OR IGINS
complex subunit 12 (MED12).19
There are also genomewide differences in DNA The cellular origin of uterine fibroids remains
methylation between fibroid tissue and the adja- unknown. Several observations support the no-
cent normal myometrium.20 A large number of tion that each fibroid originates from the trans-
formation of a single somatic stem cell of the of β-catenin in uterine mesenchyme during em-
myometrium under the influence of ovarian hor- bryonic development substantially reduces uter-
mones. Early genetic studies suggest that fibroids ine size and replaces the uterus with adipocytes,
are monoclonal tumors.8 Human and mouse disrupting entirely the normal myometrial dif-
myometrial tissues contain multipotent somatic ferentiation or regeneration of smooth muscle.
stem cells. By means of asymmetric division, this This observation suggests that β-catenin plays a
subset of tissue cells undergoes self-renewal and key role in stem-cell renewal and in the differen-
produces daughter cells under the influence of tiation of stem cells into the smooth-muscle
reproductive hormones (possibly ovarian hor- phenotype observed in myometrial and fibroid
mones); this process is responsible for regenera- tissues.29 Conversely, selective overexpression of
tion.29-31 Human uterine fibroid tissue contains constitutively activated β-catenin in uterine mes-
fewer stem cells than normal myometrium.32,33 enchyme during embryonic development and in
However, stem cells derived from fibroid tissue adult mice gives rise to fibroidlike tumors in the
— not the myometrium — carry MED12 muta- uterus.36
tions, which suggests that at least one genetic hit Complex mechanisms regulate the biologic
initially transforms a myometrial stem cell, which functions of β-catenin. Secreted WNT proteins
subsequently interacts with the surrounding bind to cell-surface receptors of the Frizzled
myometrial tissue to give rise to a fibroid tumor family, causing the activation of a cascade of
(Fig. 2).33 proteins that leads to decreased β-catenin deg-
In vivo experimental models reveal that the radation in the cytosol and ultimately changes
growth of human fibroid tumors that are depen- the amount of β-catenin that reaches the nucle-
dent on estrogen and progesterone requires the us.37 Having escaped degradation, cytoplasmic
presence of multipotent somatic stem cells.33,34 β-catenin is able to enter the nucleus and inter-
As compared with the main fibroid-cell popula- act with chromatin and the family of T-cell
tion or with normal myometrial cells, fibroid transcription factor (TCF) proteins to regulate
stem cells express remarkably low levels of re- the expression of a large number of genes and
ceptors for estrogen and progesterone. The alter key cellular functions, such as cell fate,
growth of fibroid stem cells requires the pres- tumorigenesis, and differentiation.37 The size
ence of myometrial cells with higher levels of and number of fibroidlike tumors driven by
the estrogen and progesterone receptors and β-catenin increase with parity in mice, suggest-
their ligands, suggesting that the action of ste- ing that ovarian hormones may interact with the
roid hormones on fibroid stem cells is mediated WNT–β-catenin pathway to accelerate tumori-
by myometrial cells in a paracrine fashion.33,34 It genesis.36 The activated WNT–β-catenin pathway
is likely that this paracrine interaction with the has also been shown to stimulate the expression
surrounding cells supports the self-renewal of fi- of transforming growth factor β3 (TGF-β3),
broid stem cells (Fig. 2). Both myometrial and which induces cell proliferation and the forma-
fibroid multipotent somatic stem cells lack mark- tion of extracellular matrix in human fibroid
ers for smooth-muscle cells, and in addition to tissue.36,38 Fibroid-tissue–derived TGF-β3 may
their differentiation into smooth-muscle cells in also suppress the expression of local anticoagu-
vivo, they can be induced to differentiate into lant factors in adjacent endometrial cells, which
cells with adipogenic and osteogenic lineages.31,34 results in the prolonged menstrual bleeding as-
Signaling by the wingless-type MMTV inte- sociated with fibroids.7 These observations indi-
gration site family (WNT)–β-catenin pathway cate that there are critical interactions among
seems to play a role in somatic stem-cell func- activated WNT–β-catenin and TGF-β pathways,
tion in the myometrium and in uterine fibroid estrogen and progesterone, and stem-cell re-
tissue. Overall, total β-catenin levels in the myo- newal and that these interactions ultimately give
metrium and fibroid tissue are similar.35 But rise to the clonal formation of uterine fibroid
because the key effects of β-catenin are probably tumors (Fig. 3).
manifested at the level of stem cells, which make
up a very small fraction of fibroid or myometrial GENE T IC FE AT UR E S
tissue, differences in β-catenin levels would not
be detected when whole fibroid and myometrial Hereditary syndromes and somatic chromosom-
tissues are compared. In mice, selective deletion al aberrations associated with uterine fibroids
A Normal myometrium
Paracrine interactions induced
CH3 by estrogen or progesterone
Progesterone C O
CH3
CH3
Actively
O dividing myometrial
Myometrial cells Mature
stem cell myometrial
OH
Estrogen CH3 smooth-muscle
ERα and PR cells
HO
B Fibroid-tumor initiation
Paracrine interactions
induced by estrogen or
Genetic hit progesterone
C Fibroid-tumor growth
Paracrine interactions induced
by estrogen or progesterone
ERα and PR
Actively Mature
Progesterone dividing fibroid myometrial
cells smooth-muscle
cells
Mutated myometrial
Estrogen or fibroid stem
cell
Mature fibroid
smooth-muscle
Fibroid-tissue
cells
extracellular matrix
epigenetic and phenotypic abnormalities. ERαs and PRs are concentrated primarily in mature fibroid cells and pass
on estrogenic or progestogenic signals to stem cells through paracrine mechanisms. The single, transformed fibroid
stem cell eventually gives rise to a benign fibroid tumor with well-demarcated margins, which expands within the
myometrial tissue (Panel C). Extracellular-matrix formation contributes substantially to tumor expansion.
↑ TGF-β receptor
expression
WNT SMAD
Mature
ligands
myometrial or fibroid Mutant MAPK
smooth-muscle cell MED12
Frizzled
receptors
β-catenin–TCF ↑ Proliferation
Mutant
ERα–PR β-catenin–TCF MED12 ↑ Self-
renewal
Estrogen or ↑ Proliferation
progesterone
Fibroid stem
cell
↑ TGF-β ↑ Self-
renewal
↑ Extracellular matrix
formation
Figure 3. Interactions among Ovarian Hormones, the β-Catenin and TGF-β Pathways, and MED12 in Fibroid Cells.
Since ERα and PR levels are remarkably high in mature myometrial cells and fibroid cells as compared with stem
cells, estrogen and progesterone probably send signalsCOLOR to fibroid
FIGURE stem cells through hormone receptors in mature
cells in a paracrine fashion. Estrogen and progesterone Draft 3 may increase 9/20/2013 secretion of WNT ligands from mature smooth-
Author
muscle cells surrounding the stem cells. In both cell Bulun_ra1209993
types, WNT, acting through the Frizzled family of receptors,
Fig # 3
activates the β-catenin–T-cell transcription factor
Title (TCF) Uterinepathway,
Fibroids which induces the production of transforming
growth factor β (TGF-β) in mature cells and leads
DE toLongo
excessive formation of extracellular matrix. In stem cells, non-
ME
mutant MED12 may act as a physiologic modifier of Name
β-catenin action, whereas mutant MED12 (or the absence of
Artist Williams
MED12) may lead to the failure to accomplishPubthis 10/3/2013 The absence of MED12 or the presence of the mutant
Date function.
form in stem cells has also been linked to increased AUTHOR expression
PLEASE NOTE: of the TGF-β receptor, which leads to the activation
Figure has been redrawn and type has been reset
of its downstream signaling. This in turn activates thePlease mothers
check carefullyagainst decapentaplegic homologue (SMAD) and
mitogen-activated protein kinase (MAPK) family proteins, mediating stem-cell self-renewal and proliferation.
have been reviewed previously.15,39 Analysis of presses HMGA2.43 Thus, alterations in the Let7–
single-nucleotide polymorphisms in peripheral- HMGA2–p14ARF pathway in fibroid stem cells
blood DNA has revealed three chromosomal loci may favor self-renewal and offset senescence.
— 10q24.33, 22q13.1, and 11p15.5 — associated In their study of 225 fibroid tumors from 80
with uterine fibroids.40 Somatic mutations in- patients, Mäkinen et al. found that approximate-
volving high-mobility group AT-hook 2 (HMGA2) ly 70% contained heterozygous somatic muta-
and MED12 are discussed here. Rearrangements tions that affect MED12 on the X chromosome.19
involving chromosome 12q14-15 are observed in The mutated allele was either predominantly or
7.5% of fibroids. Most of the 12q15 breakpoints exclusively expressed in affected tumors.44 Other
are located upstream of the HMGA2 gene pro- studies confirmed these findings and estab-
moter, giving rise to full-length HMGA2 overex- lished that mutations in MED12 are also present
pression, and are strongly associated with large in small subsets of other mesenchymal tumors
fibroids.17 Hmga2 expression in murine neural of the uterus or in other tissues, although the
stem cells suppresses cyclin-dependent kinase uterine fibroid remains the most frequently af-
inhibitor 2a (Cdkn2a), which encodes the pro- fected tumor.44-47
teins p16Ink4a and p14Arf, negative regulators of MED12 encodes a subunit of the mediator
their self-renewal.41 In fibroid cells, HMGA2 ap- complex, which consists of at least 26 subunits
pears to inhibit senescence by down-regulating and regulates transcription initiation and elon-
p14ARF.42 Intriguingly, uterine fibroids are defi- gation by bridging regulatory elements in gene
cient in the Let-7 miRNA that targets and sup- promoters to the RNA polymerase II initiation
complex.19 The mediator complex is highly con- pression in uterine fibroid tissue and matched
served in all eukaryotes and is required for the normal myometrial tissue from 18 black women
transcription of almost all genes in yeast.48 revealed 55 genes in the two tissue types in
MED12, together with MED13, cyclin-dependent which there were differences affecting promoter
kinase 8 (CDK8), and cyclin C, also forms a me- methylation and mRNA transcription.20 The ma-
diator subcomplex (the CDK8 module) that regu- jority of these genes (62%) displayed hypermeth-
lates transcription.48 MED12 binds directly to ylation at promoter sites that were associated
β-catenin and regulates canonical WNT signal- with their silencing in the fibroid tissues.20 A
ing.49 Because MED12 limits β-catenin–depen- large number of tumor suppressors, including
dent tissue growth during embryonic develop- the gene encoding the transcription factor Krüp-
ment, a critical question is whether the absence pel-like factor 11 (KLF11), were among these
of MED12 or the presence of a defective version hypermethylated and repressed genes.20 KLF11,
in uterine fibroid stem cells or the main fibroid- also a target of progesterone or antiprogestins in
cell population causes β-catenin pathway–depen- uterine fibroid tissue, probably plays a distinct
dent tumor growth.50,51 The expression of WNT4, role in the fibroid development.20,59 These obser-
an activator of β-catenin, is markedly elevated in vations point to the important contribution of
fibroids with MED12 mutations as compared promoter methylation-mediated gene silencing
with those without these mutations (Fig. 3).47 in the pathogenesis of uterine fibroids.
In a further twist, MED12 deficiency activates
the TGF-β pathway, leading to drug resistance E S T RO GEN
and fibroid-cell proliferation mediated by mem-
bers of two signaling protein families in cancer A large body of experimental data and circum-
cells: the mothers against decapentaplegic homo- stantial evidence suggests that estrogen stimu-
logue (SMAD) and mitogen-activated protein lates the growth of uterine fibroids through es-
kinase (MAPK) (Fig. 3).52 It is postulated that trogen receptor α.60 The primary roles of estrogen
MED12 deficiency in somatic stem cells may and estrogen receptor α in fibroid growth are
trigger these events.48 These observations point permissive in that they enable tissue to respond
to a mechanism involving MED12 mutations, to progesterone by inducing the expression of
WNT–β-catenin activation, and hyperactive TGF-β progesterone receptor (Fig. 4).10 Fibroid tissue is
signaling that supports stem-cell renewal, cell exposed to ovarian estrogen and to estrogen pro-
proliferation, and fibrosis in uterine fibroid tis- duced locally through the aromatase activity in
sue (Fig. 3).48,53,54 fibroid cells.61
In fibroid tissue, multiple promoters con-
EPIGENE T IC FE AT UR E S trolled by a diverse set of transcription factors
contribute to the expression of a single aroma-
Epigenetic mechanisms such as DNA methyla- tase protein that converts circulating precursors
tion and histone modification may be inherited into estrogens.62 The mechanism underlying
and may regulate gene expression independently gonadotropin-independent expression of aroma-
of the primary DNA sequence. DNA methyltrans- tase in fibroid tissue is not completely under-
ferases catalyze the covalent addition of a methyl stood.63 It is likely that local aromatase activity
group to a cytosine in a cytosine–guanine se- in fibroids is clinically relevant because fibroid tis-
quence. As the degree of methylation of cyto- sue from black women — who have an increased
sine–guanine sequences in a gene promoter in- prevalence of uterine fibroids and an earlier age
creases, its expression decreases. This mechanism at diagnosis, as compared with white women
is particularly important for differential gene ex- — contain high levels of aromatase, which result
pression in stem cells.55-57 in elevated levels of estrogen in tissue.64,65 Most
The aberrant expression of specific DNA important, aromatase inhibitors are as effective
methyltransferases in uterine fibroid tissue as as GnRH analogues in shrinking fibroid volume,
compared with normal myometrial tissue prompt- despite stable levels of circulating estrogen.
ed further research into DNA methylation in These observations suggest that the inhibition of
these tumors.58 Genomewide profiling of DNA aromatase in fibroid tissue is a key mechanism
methylation and messenger RNA (mRNA) ex- in hormone-dependent fibroid growth (Fig. 4).66
Adrenal gland
Skin and Ovary
adipose tissue
Fibroid tissue
CH3
CH3
N
C O
CH3 H3C OH CH3
CH3
CH3
H
Aromatase O
H
inhibitor O
O OH
Progesterone Antiprogestin
↑ Proliferation
O
HO ↓ Apoptosis
Androstenedione Aromatase Estradiol ERα ↑ PR ↑ Extracellular matrix
formation
↑Tumor growth
pausal women in a dose-dependent manner, and immediately upstream of the BCL2 transcription
the addition of progestins to GnRH agonists di- start site, thereby inhibiting cell death in fibroid
minishes the inhibitory effects of these agonists tissue (Fig. 5).75
on leiomyoma size.67,68 The strongest evidence In addition to the direct transcriptional ef-
supporting the in vivo growth-stimulating ef- fects mediated by nuclear progesterone receptor,
fects of progesterone on fibroids comes from the binding of progesterone to cytoplasmic pro-
clinical trials of three different antiprogestins, gesterone receptors can rapidly activate the ex-
each of which showed that treatment consistent- tranuclear phosphatidylinositol 3-kinase–AKT
ly reduced tumor size (Fig. 4).69-72 signaling pathway in uterine fibroid cells.76 Con-
Progesterone receptor, a ligand-activated sequently, treatment of leiomyoma cells with an
transcription factor, mediates the actions of pro- AKT inhibitor reduces progesterone-induced
gesterone and antiprogestins and exerts broad proliferation and survival of fibroid cells, under-
biologic effects as a master regulator of hun- scoring the capacity of the progesterone receptor
dreds of genes at any given time (Fig. 5).73 to interact with cytoplasmic signaling pathways.76
Across the genome of fibroid smooth-muscle During pregnancy, progesterone and its re-
cells, the antiprogestin RU486–bound progester- ceptor are instrumental in the physiologic
one receptor interacts with more than 7000 DNA growth of myometrial tissue, which after deliv-
sites, most of which lie very far from transcrip- ery regresses almost to its original volume. This
tion start sites.74 More than 75% of RU486-reg- fact argues against the view that progesterone
ulated genes contain a progesterone-receptor– receptor exerts a primary tumor-initiating ac-
binding site that is more than 50,000 bp from tion. However, by signaling through its receptor,
their transcription start sites; these genes con- progesterone may play a central role in the
trol cell growth, focal adhesion, and the func- clonal expansion of genetically or epigenetically
tioning of the extracellular matrix.74 This mech- altered fibroid stem cells into clinically detect-
anism, in which genes are regulated by the able fibroids, and it may further the growth of
progesterone receptor, contrasts with that seen these tumors by affecting both stem cells and
in breast-cancer cells, in which the majority of differentiated fibroid cells.31 Since the stem-cell
genomic targets of the RU486-bound progester- population expresses much lower levels of pro-
one receptor reside within 5000 bp of a regu- gesterone receptor than the population of ma-
lated gene.74 These observations underscore the ture cells but serves as the key source of tissue
complexity of progesterone and antiprogestin growth, a paracrine signal originating from
action and account for the difficulties in identi- progesterone-receptor–rich differentiated cells
fying a single progesterone-receptor target gene may mediate the proliferative effects of proges-
for use as an effective therapeutic strategy. terone on fibroid stem cells (Fig. 2).33,34
In fibroid cells, the antiprogestin RU486-
bound progesterone receptor assembles a tran- SUM M A R Y
scriptional complex that forms a bridge between
a 20,500-bp distal DNA sequence and the tran- During a woman’s reproductive years, myome-
scription start site of the tumor-suppressor gene trial smooth-muscle cells undergo multiple cy-
KLF11, leading to an increase in gene expression cles of growth followed by involution under the
and protein levels (Fig. 5).59 Once encoded, influence of ovarian hormones or the hormones
KLF11 effectively inhibits the proliferation of of pregnancy. These cycles make stem cells vul-
fibroid cells.59 In contrast, progesterone-bound nerable to the development of mutations. A point
progesterone receptor maintains transcriptional mutation affecting the function of MED12, a
repression of KLF11 through the same regulatory chromosomal rearrangement increasing the ex-
DNA sequence; this transcriptional control oc- pression of HMGA2, or some other gene defect in
curs in addition to the epigenetic mechanism a somatic stem cell in the myometrium may be
discussed above (i.e., hypermethylation of the the initiating event of tumorigenesis. This origi-
KLF11 transcription start site).20,59 Progesterone, nal, single genetic hit may alter key signaling
on the other hand, increases the level of the pathways such as those involving β-catenin and
antiapoptotic protein BCL2 through the binding TGF-β, which regulate cell proliferation, surviv-
of progesterone receptor to a classical sequence al, and senescence and the formation of extracel-
CH3 PR H
H
O O
1 2 3 4 5 6 7 8 9 10 11 12
13 14 15
16 17 18 19 20 21 22
B PR interaction sites
PR Antiprogestin
Progesterone RU486
PR PR
bp
50 0
−2 0 ,
References
1. Baird DD, Dunson DB, Hill MC, Cous- 5. Peddada SD, Laughlin SK, Miner K, sumed leiomyoma and rapidly growing
ins D, Schectman JM. High cumulative et al. Growth of uterine leiomyomata leiomyoma. Obstet Gynecol 1994;83:414-8.
incidence of uterine leiomyoma in black among premenopausal black and white 10. Ishikawa H, Ishi K, Serna VA, Kakazu
and white women: ultrasound evidence. women. Proc Natl Acad Sci U S A 2008; R, Bulun SE, Kurita T. Progesterone is es-
Am J Obstet Gynecol 2003;188:100-7. 105:19887-92. sential for maintenance and growth of
2. Catherino WH, Parrott E, Segars J. 6. Cardozo ER, Clark AD, Banks NK, uterine leiomyoma. Endocrinology 2010;
Proceedings from the National Institute Henne MB, Stegmann BJ, Segars JH. The 151:2433-42.
of Child Health and Human Development estimated annual cost of uterine leiomyo- 11. De Vivo A, Mancuso A, Giacobbe A,
conference on the Uterine Fibroid Research mata in the United States. Am J Obstet et al. Uterine myomas during pregnancy:
Update Workshop. Fertil Steril 2011;95: Gynecol 2012;206(3):211.e1-211.e9. a longitudinal sonographic study. Ultra-
9-12. 7. Sinclair DC, Mastroyannis A, Taylor sound Obstet Gynecol 2011;37:361-5.
3. Marshall LM, Spiegelman D, Barbieri HS. Leiomyoma simultaneously impair 12. Rosati P, Exacoustòs C, Mancuso S.
RL, et al. Variation in the incidence of endometrial BMP-2-mediated decidualiza- Longitudinal evaluation of uterine myoma
uterine leiomyoma among premenopausal tion and anticoagulant expression through growth during pregnancy: a sonographic
women by age and race. Obstet Gynecol secretion of TGF-beta3. J Clin Endocrinol study. J Ultrasound Med 1992;11:511-5.
1997;90:967-73. Metab 2011;96:412-21. 13. Laughlin SK, Herring AH, Savitz DA,
4. Faerstein E, Szklo M, Rosenshein N. 8. Linder D, Gartler SM. Glucose-6-phos- et al. Pregnancy-related fibroid reduction.
Risk factors for uterine leiomyoma: a prac- phate dehydrogenase mosaicism: utiliza- Fertil Steril 2010;94:2421-3.
tice-based case-control study. I. African- tion as a cell marker in the study of leio- 14. Filicori M, Hall DA, Loughlin JS, Rivier
American heritage, reproductive history, myomas. Science 1965;150:67-9. J, Vale W, Crowley WF Jr. A conservative
body size, and smoking. Am J Epidemiol 9. Parker WH, Fu YS, Berek JS. Uterine approach to the management of uterine
2001;153:1-10. sarcoma in patients operated on for pre- leiomyoma: pituitary desensitization by a
luteinizing hormone-releasing hormone in uterine leiomyomas using the 100K microRNAs, and stem cell aging. Cell
analogue. Am J Obstet Gynecol 1983;147: single nucleotide polymorphism array. 2008;135:1013-6.
726-7. Exp Mol Pathol 2011;91:434-9. 42. Markowski DN, Helmke BM, Belge G,
15. Hodge JC, Morton CC. Genetic hetero- 28. Varghese BV, Koohestani F, McWil- et al. HMGA2 and p14Arf: major roles in
geneity among uterine leiomyomata: in- liams M, et al. Loss of the repressor cellular senescence of fibroids and thera-
sights into malignant progression. Hum REST in uterine fibroids promotes aber- peutic implications. Anticancer Res 2011;
Mol Genet 2007;16:Special Number 1: rant G protein-coupled receptor 10 ex- 31:753-61.
R7-R13. pression and activates mammalian target 43. Peng Y, Laser J, Shi G, et al. Antipro-
16. Tomlinson IP, Alam NA, Rowan AJ, et of rapamycin pathway. Proc Natl Acad Sci liferative effects by Let-7 repression of
al. Germline mutations in FH predispose U S A 2013;110:2187-92. high-mobility group A2 in uterine leio-
to dominantly inherited uterine fibroids, 29. Arango NA, Szotek PP, Manganaro myoma. Mol Cancer Res 2008;6:663-73.
skin leiomyomata and papillary renal cell TF, Oliva E, Donahoe PK, Teixeira J. Con- 44. Pérot G, Croce S, Ribeiro A, et al.
cancer. Nat Genet 2002;30:406-10. ditional deletion of beta-catenin in the MED12 alterations in both human benign
17. Hodge JC, Kim TM, Dreyfuss JM, et al. mesenchyme of the developing mouse and malignant uterine soft tissue tumors.
Expression profiling of uterine leiomyo- uterus results in a switch to adipogenesis PLoS One 2012;7(6):e40015.
mata cytogenetic subgroups reveals dis- in the myometrium. Dev Biol 2005;288: 45. McGuire MM, Yatsenko A, Hoffner L,
tinct signatures in matched myometrium: 276-83. Jones M, Surti U, Rajkovic A. Whole
transcriptional profiling of the t(12;14) 30. Szotek PP, Chang HL, Zhang L, et al. exome sequencing in a random sample of
and evidence in support of predisposing Adult mouse myometrial label-retaining North American women with leiomyomas
genetic heterogeneity. Hum Mol Genet cells divide in response to gonadotropin identifies MED12 mutations in majority
2012;21:2312-29. stimulation. Stem Cells 2007;25:1317-25. of uterine leiomyomas. PLoS One 2012;
18. Mehine M, Kaasinen E, Mäkinen N, 31. Ono M, Maruyama T, Masuda H, et al. 7(3):e33251.
et al. Characterization of uterine leiomyo- Side population in human uterine myo- 46. Ravegnini G, Mariño-Enriquez A,
mas by whole-genome sequencing. N Engl metrium displays phenotypic and func- Slater J, et al. MED12 mutations in leio-
J Med 2013;369:43-53. tional characteristics of myometrial stem myosarcoma and extrauterine leiomyoma.
19. Mäkinen N, Mehine M, Tolvanen J, et cells. Proc Natl Acad Sci U S A 2007;104: Mod Pathol 2013;26:743-9.
al. MED12, the mediator complex subunit 18700-5. 47. Markowski DN, Bartnitzke S, Löning
12 gene, is mutated at high frequency in 32. Chang HL, Senaratne TN, Zhang L, T, Drieschner N, Helmke BM, Bullerdiek
uterine leiomyomas. Science 2011;334: et al. Uterine leiomyomas exhibit fewer J. MED12 mutations in uterine fibroids —
252-5. stem/progenitor cell characteristics when their relationship to cytogenetic sub-
20. Navarro A, Yin P, Monsivais D, et al. compared with corresponding normal groups. Int J Cancer 2012;131:1528-36.
Genome-wide DNA methylation indicates myometrium. Reproductive Sci 2010;17: 48. Guo X, Wang XF. A mediator lost in
silencing of tumor suppressor genes in 158-67. the war on cancer. Cell 2012;151:927-9.
uterine leiomyoma. PLoS One 2012;7(3): 33. Ono M, Qiang W, Serna VA, et al. Role 49. Kim S, Xu X, Hecht A, Boyer TG. Me-
e33284. of stem cells in human uterine leiomyoma diator is a transducer of Wnt/beta-catenin
21. Mesquita FS, Dyer SN, Heinrich DA, growth. PLoS One 2012;7(5):e36935. signaling. J Biol Chem 2006;281:14066-
Bulun SE, Marsh EE, Nowak RA. Reactive 34. Mas A, Cervelló I, Gil-Sanchis C, et al. 75.
oxygen species mediate mitogenic growth Identification and characterization of the 50. Rocha PP, Scholze M, Bleiss W,
factor signaling pathways in human leio- human leiomyoma side population as pu- Schrewe H. Med12 is essential for early
myoma smooth muscle cells. Biol Reprod tative tumor-initiating cells. Fertil Steril mouse development and for canonical
2010;82:341-51. 2012;98(3):741.e6-751.e6. Wnt and Wnt/PCP signaling. Development
22. Mason HR, Lake AC, Wubben JE, 35. Tai CT, Lin WC, Chang WC, Chiu TH, 2010;137:2723-31.
Nowak RA, Castellot JJ Jr. The growth ar- Chen GT. Classical cadherin and catenin 51. Lin X, Rinaldo L, Fazly AF, Xu X. De-
rest-specific gene CCN5 is deficient in hu- expression in normal myometrial tissues pletion of Med10 enhances Wnt and sup-
man leiomyomas and inhibits the prolif- and uterine leiomyomas. Mol Reprod Dev presses Nodal signaling during zebrafish
eration and motility of cultured human 2003;64:172-8. embryogenesis. Dev Biol 2007;303:536-48.
uterine smooth muscle cells. Mol Hum 36. Tanwar PS, Lee HJ, Zhang L, et al. 52. Huang S, Hölzel M, Knijnenburg T,
Reprod 2004;10:181-7. Constitutive activation of Beta-catenin in et al. MED12 controls the response to
23. Laping NJ, Everitt JI, Frazier KS, et al. uterine stroma and smooth muscle leads multiple cancer drugs through regulation
Tumor-specific efficacy of transforming to the development of mesenchymal tu- of TGF-beta receptor signaling. Cell 2012;
growth factor-beta RI inhibition in Eker mors in mice. Biol Reprod 2009;81:545- 151:937-50.
rats. Clin Cancer Res 2007;13:3087-99. 52. 53. Lee BS, Nowak RA. Human leiomyo-
24. Gilden M, Malik M, Britten J, Delgado 37. Mosimann C, Hausmann G, Basler K. ma smooth muscle cells show increased
T, Levy G, Catherino WH. Leiomyoma fi- Beta-catenin hits chromatin: regulation expression of transforming growth fac-
brosis inhibited by liarozole, a retinoic of Wnt target gene activation. Nat Rev Mol tor-beta 3 (TGF beta 3) and altered re-
acid metabolic blocking agent. Fertil Steril Cell Biol 2009;10:276-86. sponses to the antiproliferative effects of
2012;98:1557-62. 38. Arici A, Sozen I. Transforming growth TGF beta. J Clin Endocrinol Metab 2001;
25. Norian JM, Owen CM, Taboas J, et al. factor-beta3 is expressed at high levels in 86:913-20.
Characterization of tissue biomechanics leiomyoma where it stimulates fibronec- 54. Catherino WH, Leppert PC, Stenmark
and mechanical signaling in uterine leio- tin expression and cell proliferation. Fer- MH, et al. Reduced dermatopontin ex-
myoma. Matrix Biol 2012;31:57-65. til Steril 2000;73:1006-11. pression is a molecular link between uter-
26. Halder SK, Goodwin JS, Al-Hendy A. 39. Walker CL, Stewart EA. Uterine fi- ine leiomyomas and keloids. Genes Chro-
1,25-Dihydroxyvitamin D3 reduces TGF- broids: the elephant in the room. Science mosomes Cancer 2004;40:204-17.
beta3-induced fibrosis-related gene ex- 2005;308:1589-92. 55. Dodge JE, Ramsahoye BH, Wo ZG,
pression in human uterine leiomyoma 40. Cha PC, Takahashi A, Hosono N, et al. Okano M, Li E. De novo methylation of
cells. J Clin Endocrinol Metab 2011; A genome-wide association study identi- MMLV provirus in embryonic stem cells:
96(4):E754-E762. fies three loci associated with susceptibil- CpG versus non-CpG methylation. Gene
27. Meadows KL, Andrews DM, Xu Z, et al. ity to uterine fibroids. Nat Genet 2011; 2002;289:41-8.
Genome-wide analysis of loss of hetero- 43:447-50. 56. Jaenisch R, Bird A. Epigenetic regula-
zygosity and copy number amplification 41. Hammond SM, Sharpless NE. HMGA2, tion of gene expression: how the genome
integrates intrinsic and environmental et al. High aromatase expression in uter- 70. Williams AR, Critchley HO, Osei J,
signals. Nat Genet 2003;33:Suppl:245-54. ine leiomyoma tissues of African-Ameri- et al. The effects of the selective proges-
57. Lister R, Pelizzola M, Dowen RH, et al. can women. J Clin Endocrinol Metab terone receptor modulator asoprisnil on
Human DNA methylomes at base resolu- 2009;94:1752-6. the morphology of uterine tissues after
tion show widespread epigenomic differ- 65. Sumitani H, Shozu M, Segawa T, et al. 3 months treatment in patients with
ences. Nature 2009;462:315-22. In situ estrogen synthesized by aromatase symptomatic uterine leiomyomata. Hum
58. Li S, Chiang TC, Richard-Davis G, P450 in uterine leiomyoma cells promotes Reprod 2007;22:1696-704.
Barrett JC, McLachlan JA. DNA hypo- cell growth probably via an autocrine/ 71. Donnez J, Tomaszewski J, Vázquez F,
methylation and imbalanced expression intracrine mechanism. Endocrinology et al. Ulipristal acetate versus leuprolide
of DNA methyltransferases (DNMT1, 3A, 2000;141:3852-61. acetate for uterine fibroids. N Engl J Med
and 3B) in human uterine leiomyoma. Gy- 66. Parsanezhad ME, Azmoon M, Alborzi 2012;366:421-32.
necol Oncol 2003;90:123-30. S, et al. A randomized, controlled clinical 72. Donnez J, Tatarchuk TF, Bouchard P,
59. Yin P, Lin Z, Reierstad S, et al. Tran- trial comparing the effects of aromatase et al. Ulipristal acetate versus placebo for
scription factor KLF11 integrates proges- inhibitor (letrozole) and gonadotropin- fibroid treatment before surgery. N Engl J
terone receptor signaling and prolifera- releasing hormone agonist (triptorelin) Med 2012;366:409-20.
tion in uterine leiomyoma cells. Cancer on uterine leiomyoma volume and hor- 73. Kim JJ, Sefton EC. The role of proges-
Res 2010;70:1722-30. monal status. Fertil Steril 2010;93:192-8. terone signaling in the pathogenesis of
60. Marsh EE, Bulun SE. Steroid hor- 67. Carr BR, Marshburn PB, Weatherall uterine leiomyoma. Mol Cell Endocrinol
mones and leiomyomas. Obstet Gynecol PT, et al. An evaluation of the effect of 2012;358:223-31.
Clin North Am 2006;33:59-67. gonadotropin-releasing hormone analogs 74. Yin P, Roqueiro D, Huang L, et al.
61. Bulun SE, Simpson ER, Word RA. Ex- and medroxyprogesterone acetate on Genome-wide progesterone receptor bind-
pression of the CYP19 gene and its prod- uterine leiomyomata volume by magnetic ing: cell type-specific and shared mecha-
uct aromatase cytochrome P450 in human resonance imaging: a prospective, ran- nisms in T47D breast cancer cells and
uterine leiomyoma tissues and cells in domized, double blind, placebo-con- primary leiomyoma cells. PLoS One 2012;
culture. J Clin Endocrinol Metab 1994;78: trolled, crossover trial. J Clin Endocrinol 7(1):e29021.
736-43. Metab 1993;76:1217-23. 75. Yin P, Lin Z, Cheng YH, et al. Proges-
62. Imir AG, Lin Z, Yin P, et al. Aromatase 68. Friedman AJ, Daly M, Juneau-Nor- terone receptor regulates Bcl-2 gene ex-
expression in uterine leiomyomata is reg- cross M, et al. A prospective, randomized pression through direct binding to its
ulated primarily by proximal promoters trial of gonadotropin-releasing hormone promoter region in uterine leiomyoma
I.3/II. J Clin Endocrinol Metab 2007;92: agonist plus estrogen-progestin or pro- cells. J Clin Endocrinol Metab 2007;92:
1979-82. gestin “add-back” regimens for women 4459-66.
63. Ishikawa H, Fenkci V, Marsh EE, et al. with leiomyomata uteri. J Clin Endocrinol 76. Hoekstra AV, Sefton EC, Berry E, et al.
CCAAT/enhancer binding protein beta Metab 1993;76:1439-45. Progestins activate the AKT pathway in
regulates aromatase expression via mul- 69. Murphy AA, Kettel LM, Morales AJ, leiomyoma cells and promote survival.
tiple and novel cis-regulatory sequences Roberts VJ, Yen SS. Regression of uterine J Clin Endocrinol Metab 2009;94:1768-
in uterine leiomyoma. J Clin Endocrinol leiomyomata in response to the antipro- 74.
Metab 2008;93:981-91. gesterone RU 486. J Clin Endocrinol Metab Copyright © 2013 Massachusetts Medical Society.
64. Ishikawa H, Reierstad S, Demura M, 1993;76:513-7.