Synthesis and evaluation of synthetic retinoid derivatives as inducers of stem cell differentiation

Article abstract of DOI:10.1039/b808574a

All-trans-retinoic acid (ATRA) and its associated analogues are important mediators of cell differentiation and function during the development of the nervous system. It is well known that ATRA can induce the differentiation of neural tissues from human pluripotent stem cells. However, it is not always appreciated that ATRA is highly susceptible to isomerisation when in solution, which can influence the effective concentration of ATRA and subsequently its biological activity. To address this source of variability, synthetic retinoid analogues have been designed and synthesised that retain stability during use and maintain biological function in comparison to ATRA. It is also shown that subtle modifications to the structure of the synthetic retinoid compound impacts significantly on biological activity, as when exposed to cultured human pluripotent stem cells, synthetic retinoid 4-(5,5,8,8-tetramethyl-5,6,7,8- tetrahydronaphthalen-2-ylethynyl)benzoic acid, 4a (para-isomer), induces neural differentiation similarly to ATRA. In contrast, stem cells exposed to synthetic retinoid 3-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)benzoic acid, 4b (meta-isomer), produce very few neurons and large numbers of epithelial-like cells. This type of structure-activity-relationship information for such synthetic retinoid compounds will further the ability to design more targeted systems capable of mediating robust and reproducible tissue differentiation.

Full text of DOI:10.1039/b808574a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis and evaluation of synthetic retinoid derivatives as inducers of stem
cell differentiation†
Victoria B. Christie,a,b Jonathan H. Barnard,c Andrei S. Batsanov,c Caroline E. Bridgens,a,c Emily B. Cartmell,c
Jonathan C. Collings,c Daniel J. Maltman,a,b Christopher P. F. Redfern,d Todd B. Marder,*‡c
a,b
Stefan Przyborski§ and Andrew Whiting*c
Received 21st May 2008, Accepted 7th July 2008
First published as an Advance Article on the web 8th August 2008
DOI: 10.1039/b808574a
All-trans-retinoic acid (ATRA) and its associated analogues are important mediators of cell
differentiation and function during the development of the nervous system. It is well known that ATRA
can induce the differentiation of neural tissues from human pluripotent stem cells. However, it is not
always appreciated that ATRA is highly susceptible to isomerisation when in solution, which can
influence the effective concentration of ATRA and subsequently its biological activity. To address this
source of variability, synthetic retinoid analogues have been designed and synthesised that retain
stability during use and maintain biological function in comparison to ATRA. It is also shown that
subtle modifications to the structure of the synthetic retinoid compound impacts significantly on
biological activity, as when exposed to cultured human pluripotent stem cells, synthetic retinoid
4-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)benzoic acid, 4a (para-isomer), induces
neural differentiation similarly to ATRA. In contrast, stem cells exposed to synthetic retinoid
3-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)benzoic acid, 4b (meta-isomer),
produce very few neurons and large numbers of epithelial-like cells. This type of
structure–activity-relationship information for such synthetic retinoid compounds will further the
ability to design more targeted systems capable of mediating robust and reproducible tissue
differentiation.
Introduction
Retinoids are a class of natural and synthetic analogues of all-
trans-retinoic acid (ATRA), which is a metabolite of vitamin A
(retinol), a fat-soluble vitamin required by all mammals. ATRA,
and its two naturally occurring isomers, 9-cis-retinoic acid (9CRA)
and 13-cis-retinoic acid (13CRA) (Fig. 1), are capable of mediating
many biological processes both during embryonic development
and in adult life, particularly in the nervous system.1 They are vital
in the mediation of cell proliferation, differentiation and apoptosis,
and they maintain these processes in both normal and tumour
cells, in vivo and in vitro. Due to the ability of different retinoids to
control differentiation and apoptosis of tumour cells, they have
potential therapeutic uses in the treatment and prevention of
Fig. 1 Structures for ATRA, 9CRA and 13CRA.
cancer.2 Indeed, ATRA and other commercially available retinoids
are currently used in a variety of cell differentiation therapies,3
rather than the naturally occurring retinoids, which contain a
polyene chain comprising five conjugated double bonds, and
are thus excellent chromophores which efficiently absorb light
in the region of 300–400 nm (depending on the solvent). This
makes these molecules particularly susceptible to photoisomeri-
sation, leading to degradation into a mixture of retinoic acid
isomers.4
Compound stability is an important feature when reagents are
routinely used for cell culture applications. During the preparation
and maintenance of cell cultures, the concentration levels of
retinoids have been shown to decrease markedly over time, and
aDepartment of Biological Sciences, Durham University, Science Laborato-
ries, South Road, Durham, DH1 3LE, UK
bReInnervate Ltd., Department of Biological Sciences, Durham University,
Science Laboratories, South Road, Durham, DH1 3LE, UK
cDepartment of Chemistry, Durham University, Science Laboratories, South
Road, Durham, DH1 3LE, UK
dNorthern Institute for Cancer Research, University of Newcastle upon Tyne,
Newcastle upon Tyne, NE1 7RU, UK
† Electronic supplementary information (ESI) available: X-Ray crystallo-
graphic information. CCDC reference numbers 688676–688681. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/b808574a
‡ Corresponding authors concerning Chemistry.
§ Corresponding author concerning Biology.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3497–3507 | 3497
©

View Article Online
this could be a direct consequence of both their degradation
and metabolism.5 Isomerisation of ATRA is understood to be
an important part of its metabolic pathways within cells, since the
resulting isomers have different mechanisms of action.5,6 Indeed,
this is an important point that is all too often over-looked by
users of natural retinoids in the cell culture laboratory. A warning
to this effect has been issued to cell biologists using retinoic
acids as inducers of cell differentiation by Murayama et al.,4
who reported that the isomers of ATRA differentially affect the
ability of mammalian stem cells to differentiate along alternative
lineages, and they stated that care should be taken to protect
retinoic acids from isomerisation. This is particularly relevant
when the cellular response is determined by the concentration(s)
of the isomer(s) present in solution. For example, the induction
of the differentiation of pluripotent stem cells using retinoids is
variable, resulting in the differential activation of key molecular
pathways involved in tissue development in a concentration-
dependent manner.7 In turn, this variation has the potential to
result in mixed proportions of alternative differentiating cell types
which leads to increased culture heterogeneity. To reduce such
variability in differentiation response and improve reproducibility,
it is essential that whatever is used to induce cell differentiation is
in the same form and concentration every time it is used. Currently,
this cannot be guaranteed, especially when using reagents such as
ATRA and its stereoisomers because they are highly susceptible
to isomerisation under sample preparation conditions, storage of
stock solution conditions and in culture.
It is because the different isomers have diverse effects on cells
that some attempts have been made to control ATRA’s sensitivity
and tendency to isomerise. For example, a number of additives
which inhibit either cis–trans interconversion or oxidation have
been evaluated, including bovine serum albumin (BSA), fibrogen,
lysozyme, phosphatidylcholine N-ethylmaleimide and vitamin C.8
However, the addition of such molecules to cell culture media is
often not viable because such additives may affect cell behaviour in
their own way. In addition, none of these additives can completely
prevent isomerisation and, for example, the use of BSA is not
possible in serum-free culture media.
to modulate tissue development; notably, the induction of cell
differentiation in a recognised pluripotent stem cell model of
human embryogenesis,9 is demonstrated.
Results and discussion
Synthetic retinoid synthesis
Retinoids have three structurally important regions: a hydropho-
bic section; a variable linker; and an acidic function, and extensive
chemical modifications to these motifs are possible. For example,
incorporating an aromatic ring to replace part of the polyene
chain forms a new class of retinoids, sometimes called arotinoids.
The major consequence of this replacement is an increase in
chemical and physical stability of the resulting retinoids compared
to natural retinoic acids, which are unstable under a wide variety
of conditions, including exposure to heat and light.4 A common
replacement for the trimethylcyclohexenyl ring and part of the
polyene chain in the retinoic acids is the 1,1,4,4-tetramethyl-
1,2,3,4-tetrahydronaphthalene unit,10 and hence, we decided to
incorporate such a function. In addition, it was expected that
an acetylene moiety would not only provide a suitable linear
structure, but it would also act as a non-isomerisable linker unit.
Hence, it was decided to access isomeric compounds of type
4, in which the isomeric placement of the carboxylate function
at either the para- or meta-positions might mimic some of the
different natural retinoic acid isomers, as has been reinforced by
competitive binding studies with nuclear receptor proteins.11 It was
envisaged that this series would be readily accessible, as outlined
in Scheme 1, using serial cross-coupling reactions starting from
bromide 1.
Although bromide 1 is known,12 an improved preparation
was developed (Scheme 2) involving boron trifluoride-catalysed
bromination. Subsequent Sonagashira cross-coupling of bromide
1 with trimethylsilylacetylene (TMSA) or 2-methylbut-3-yn-2-
ol, according to literature methods13 (with careful degassing to
remove oxygen, and thus avoid diyne formation) initially gave
no conversion to either the TMSA derivative 2a or dimethyl
carbinol product 2b. However, a combination of PdCl2 and PPh3
in conjunction with a Cu(II) co-catalyst gave improved results,
with the reaction of bromide 1 with TMSA yielding 2a in
82% yield. Larger scale reactions generally had small amounts
In this paper, we describe our recent work on the design, prepa-
ration and preliminary biological evaluation of synthetic retinoids
which remain stable to degradation for extended periods under
normal laboratory light. The ability of these types of compounds
Scheme 1 Proposed synthetic sequence for the synthesis of isomeric retinoid analogues 4.
Scheme 2 Synthesis of acetylene derivative 3 via bromide 5.
3498 | Org. Biomol. Chem., 2008, 6, 3497–3507 This journal is
The Royal Society of Chemistry 2008
©

View Article Online
of 1,4-bis(trimethylsilyl)butadiyne present as by-product, which
were readily removed later in the synthesis (Scheme 1). The
corresponding dimethyl carbinol 2b could also be accessed using
the same protocol, though in lower conversion (45%), and hence
the TMSA derivative 2a, which could be readily de-silylated
using KOH to give acetylene 3 (Scheme 2), was carried forward
for the synthesis of synthetic retinoids 4. It is clear that 1 has
unusually low reactivity for an aryl bromide, suggesting a strong
deactivation by the inductively donating alkyl groups of the
tetrahydrotetramethylnaphthalene unit.
As the coupling of bromide 1 with either TMSA or 2-methylbut-
3-yn-2-ol required relatively high palladium-catalyst loadings, a
more reactive coupling partner was considered, and iodide 6
appeared to be an ideal alternative reagent to bromide 1. However,
a suitable, efficient method for the synthesis of iodide 6 was
required.
Fig. 2 1H NMR at 400 MHz of ATRA in DMSO-d6 (5.0–8.0 ppm region)
in the absence of light.
Iodide 6 was prepared efficiently by an interesting, adapted
literature method14 in 71% yield (Scheme 1). Cross-coupling of
iodide 6 with TMSA under Sonogashira conditions occurred
readily at room temperature with only 1% Pd catalyst resulting
in an 88% yield of acetylene 2a. Subsequent deprotection of 2a
with hydroxide gave the alkyne 3, which was subjected to a second
Sonogashira coupling with both para- and meta-iodobenzoate
methyl esters to provide each of the isomeric esters 7a and 7b,
in 87% and 59% yields respectively (Scheme 3). Saponification of
the methyl esters 7 produced the corresponding carboxylic acids 4
in 72 and 78% yields, for the para- and meta-isomers respectively.
Stability of synthetic retinoids versus ATRA
Fig. 3 1H NMR at 400 MHz of ATRA in DMSO-d6 (5.0–8.0 ppm region)
The photostability of the synthetic retinoids 4 was tested and com-
pared with that of ATRA. The susceptibility of ATRA to undergo
photoisomerisation and degradation when exposed to ordinary
laboratory (fumehood) fluorescent light (see Experimental sec-
tion) in the visible to near-UV range was demonstrated first using
1H NMR spectroscopy. This showed that prolonged exposure to
fluorescent light led to progressive degradation of ATRA, and
after 3 days there was substantial isomerisation/degradation of
the ATRA (ca. 37% remaining), as shown in Fig. 3 (light-exposed)
compared with Fig. 2 (no light exposure). In contrast, ATRA was
found to be stable for at least 3 days when kept in the dark. When
after 3 weeks exposure to white fluorescent light at a distance of 40 cm.
the synthetic retinoids 4 were exposed to the same wavelength of
1
fluorescent light for 3 days, and the H NMR spectra compared
with control samples which had been left in the dark for the same
length of time, it was found that both 4a and 4b were completely
unaltered (Figs. 4 and 5). These results clearly demonstrate that
the replacement of the polyene chain in ATRA by a diphenylene
acetylene motif does result in a dramatic increase in photostability
compared with ATRA.
Scheme 3 Synthesis of synthetic retinoids 4 starting with iodide 6.
The Royal Society of Chemistry 2008 Org. Biomol. Chem., 2008, 6, 3497–3507 | 3499
This journal is
©

View Article Online
and 10 lM solutions of ATRA and the para-isomer synthetic
retinoid 4a slowed significantly, indicative of cells exiting the cell
cycle and committing to differentiation. However, the number
of viable cells in cultures treated with a 1 lM solution of the
meta-isomer synthetic retinoid 4b were not different to those of
the control and such cultures became over-confluent and were
terminated, indicating the inability of this compound to arrest
cell proliferation at this concentration. Cultures exposed to the
higher concentration of 4b (10 lM) did remain viable though
with reduced proliferation. Furthermore, the rate at which cells
committed to differentiate in response to 4b (10 lM) was slower
compared to that induced by either ATRA or 4a.
TERA2.cl.SP12 cells treated with natural or synthetic retinoids
responded positively and showed marked differentiation (Fig. 6).
The morphology of cultures exposed to 4a were similar to those
treated with ATRA, with both containing heterogeneous cell
populations and, in particular, containing structures resembling
neural rosettes typically seen in neural cell cultures (Fig. 6B and
C).15 Alternatively, 21 days of treatment with 10 lM 4b resulted
in highly confluent cultures containing distinct regions of large
flat epithelial-like cells (referred to as ‘plaques’) (see Fig. 6D).
Plaques formed by the large flat cells were distributed throughout
the culture and were large enough to be seen with the naked eye
against the bottom of the culture flask. Furthermore, the large
flat cells within the plaques showed strong staining for epithelial
cytokeratin-8 (Fig. 7). Such structures were observed to a lesser
extent in cultures treated with ATRA and 4a and occupied a
significantly smaller area compared to cultures treated with 4b
(Fig. 7D, E). Indeed, the proportion of the cell population that
Fig. 4 1H NMR at 400 MHz of 4a in DMSO-d6 (5.0–8.0 ppm region)
after 3 weeks exposure to white fluorescent light at a distance of 40 cm.
Fig. 5 1H NMR at 400 MHz of 4b in DMSO-d6 (5.0–8.0 ppm region)
after 3 weeks exposure to white fluorescent light at a distance of 40 cm.
Effect of synthetic retinoids on the differentiation of human
pluripotent stem cells
Having prepared synthetic retinoids 4, they were assessed for
their ability to induce the differentiation of the human embryonal
carcinoma stem cell line, TERA2.cl.SP12. Cell lines of the TERA2
lineage are proven models of human embryonic development9
and have frequently been used to study neural differentiation
in particular.15 To compare the effect of natural versus synthetic
retinoids on stem cell differentiation, cultures of TERA2.cl.SP12
cells were incubated with 1 lM and 10 lM concentrations of each
type of compound for up to 21 days. Control cultures consisted
of cells treated with the organic solvent (DMSO) alone (1 : 1000
dilution in tissue culture medium), which was used as the vehicle to
dissolve each synthetic retinoids tested. The number of viable cells
was determined during the first 7 days of treatment with each test
compound by using a standard MTS assay to monitor the rate of
cell proliferation (data not shown). Cell numbers were compared
to those of the control. Note that in all of our experiments using
ATRA, careful precautions were taken to ensure that only pure
ATRA was used, and that no degradation could occur during
the period of the experiments, by exclusion of light. On this basis,
accurate comparisons could be made on the effects all compounds
studied.
Fig. 6 Morphological appearance of TERA2.cl.SP12 human pluripotent
stem cells in the presence and absence of retinoid. (A) TERA2.cl.SP12
stem cells (control). (B) TERA2.cl.SP12 cells exposed to 10 lM ATRA
for 21 days. (C) TERA2.cl.SP12 cells exposed to 10 lM 4a (EC23)
for 21 days. (D) TERA2.cl.SP12 cells exposed to 10 lM 4b (EC19) for
21 days. TERA2.cl.SP12 stem cells retained a homogeneous ‘cobblestone’
appearance throughout the culture (A). Cultures exposed to ATRA
differentiate and become heterogeneous in appearance consisting of neural
rosettes [bordered area, (B)] and regions of flat epithelial-like cells (referred
to as ‘plaques’). Similarly, cultures exposed to 4a formed neural rosettes
[bordered area, (C)], but fewer plaques compared to ATRA. In contrast,
4b induced the formation of multiple, large plaques of flat epithelial-like
cells [bordered area, (D)] throughout the culture with no evidence of neural
rosettes. Scale bars: 100 lm.
Cultures treated with DMSO alone (control) continued to
proliferate, became over-confluent and were terminated after
7 days. In contrast, cell proliferation in cultures exposed to 1 lM
3500 | Org. Biomol. Chem., 2008, 6, 3497–3507
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 7 Induction of cytokeratin-8-positive epithelial plaques in response to retinoids. Immunofluorescence micrographs show cytokeratin-8 expression
in TERA2.cl.SP12 cultures exposed to 10 lM of either ATRA, 4a (EC23) or 4b (EC19) (A–C respectively). Plaque areas are outlined in each image. For
each condition, 12 randomly selected fields of view were collected. The number of plaques (D) and the area which they occupied (E) was determined per
field of view in cultures exposed to ATRA, 4a and 4b at a concentration of either 1 lM or 10 lM for 21 days. The number and average area covered by
epithelial plaques increased significantly in 4b cultures. As the concentration of ATRA or 4a increased, the number of plaques per field and their average
area occupancy per field of view decreased, being almost totally absent in cultures exposed to 10 lM 4a. Data represent mean SD, n = 12.
formed plaques was particularly low in cultures treated with a
10 lM solution of 4a. Cultures treated with solutions of either
10 lM ATRA or 4a resulted in slightly fewer cytokeratin-8-positive
plaques compared to those treated with a 1 lM dose of either
ATRA or 4a, but this was not significant.
However, in cultures treated with 10 lM 4b, SSEA-3 and TRA-
1-60 continued to decline to levels comparable to those cultures
treated with either ATRA or 4a for 14 days. These data indicate
that 4b can induce the differentiation of TERA2.cl. SP12 cells,
albeit at a slower rate than either ATRA or 4a. Commitment of
cells to differentiate in response to each of the retinoids tested
herein correlated with the regulated levels of cell proliferation as
determined by the MTS assay (as described above).
TERA2.cl.SP12 cells are well known for their ability to form
neurons in response to ATRA.15 Accordingly, flow cytometry
was used to evaluate the expression of antigens associated with
differentiating neural cell types, VINIS-53 (neuronal cell adhesion
molecule (NCAM)) and A2B5 (ganglioseries antigen marking
early-stage neural cells) (Fig. 8), markers which are expressed
during the early stages of neuronal development. Each retinoid
tested induced the expression of VINIS-53 and A2B5, although
there were some differences between the different compounds in
the rate of expression and the maximum levels attained. Notably,
cells exposed to 10 lM 4a showed the highest levels of VINIS-
53 and A2B5 during the period tested, whilst those treated with
4b were considerably slower at inducing the expression of neural
antigens, which correlated with the delayed decrease in stem cell
markers.
Induction of cellular differentiation in response to retinoids was
further evaluated by monitoring the expression profile of known
markers for stem cell and differentiated cell phenotypes (Fig. 8).
Flow cytometry was performed on samples of cells tested for the
expression of the stem cell antigens, SSEA-3 (globoseries stage
specific embryonic antigen-3) and TRA-1-60 (keratin-sulfate-
associated glycoprotein stem surface marker). The reduction in
the expression level of SSEA-3 and TRA-1-60 indicated cells
committing to differentiate in response to both ATRA and 4a. The
activity of ATRA and 4a was similar at both the 1 lM and 10 lM
concentrations. After 7 days exposure to either ATRA or 4a, the
stem cell antigens were heavily down-regulated; however, SSEA-3
and TRA-1-60 expression levels remained comparatively high in
cultures treated with 4b up to 7 days after addition. Nonetheless,
the decrease in stem cell antigen expression relative to control
cultures was marked, indicating that cellular differentiation had
been triggered in a proportion of cells. After 7 days, cultures treated
with 1 lM 4b were discarded due to overgrowth, as recorded above.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3497–3507 | 3501
©

View Article Online
Fig. 8 Flow cytometric analysis of the induction of TERA2.cl.SP12 cell differentiation in response to retinoids. Markers of stem cells (SSEA-3 and
TRA-1-60, A, B respectively) and neural differentiation (VINIS-53 and A2B5, C, D respectively) were differentially regulated during the 21 day culture
period in response to 1 lM or 10 lM ATRA, 4a or 4b. Both concentrations of ATRA and 4a resulted in a rapid decrease in expression of the stem cell
markers and reciprocal increase in neural marker expression. Similarly, 10 lM 4b induced differentiation as evidenced by the absence of the stem cell
markers after 14 days, although this down-regulation occurred at a slower rate than that induced by 4a and 4b. Treatment with 1 lM 4b did not induce
differentiation; cells continued to proliferate and the experiment was terminated after 7 days as a consequence of becoming over-confluent. Data represent
mean SEM, n = 3.
Confirmation of neural development of TERA2.cl.SP12 cells
was demonstrated by immunocytochemical staining of cultures
treated with each of the retinoid compounds over 21 days (see
Figs. 9 and 10). Nestin, a protein marker expressed by neuropro-
genitor cells, showed low levels of expression in TERA2.cl.SP12
stem cells which was significantly up-regulated throughout cul-
tures treated with either ATRA or 4a at either 1 lM or 10 lM
concentrations (Fig. 9). Treatment with both ATRA and 4a
induced large numbers of nestin-positive structures resembling
‘neural-rosettes’ which are characteristic of aggregations of cul-
tured neuroprogenitor cells. In contrast, cells exposed to solutions
of 4b resulted in lower levels of nestin expression which was
localised to discrete areas, and nestin expression was absent
in the large plaques consisting of flat epithelial-like cells. The
expression of neurofilament 200 (NF200) was also examined as
a cytoskeletal marker of mature neurons (Fig. 10). As expected,
TERA2.cl.SP12 stem cells did not express NF200. Large numbers
of NF200-positive neurons were identified in cultures treated with
either ATRA or 4a for both concentrations tested. In contrast,
cultures exposed to 4b produced few mature neurons and the large
epithelial plaques stained negative for NF200. Collectively, these
data indicate that the retinoids ATRA and 4a behave in a similar
manner and induce the differentiation of human pluripotent stem
cells to form mature neural derivatives. Quantification of the
number of NF200-positive perykarya in each culture treatment
was calculated as a mean from six randomly selected cultures
(Fig. 10). In rank order, the highest number of NF200-positive
perykarya (and by inference number of mature neurons), was
seen in cultures treated with 10 lM 4a > 10 lM ATRA >
1 lM 4a > 1 lM ATRA >> 10 lM 4b. Indeed exposure to
10 lM 4a resulted in approximately 38% more NF200-positive
cells compared to 10 lM ATRA. According to these data, the
stable synthetic retinoid 4a appears to be a more potent inducer
of neuronal differentiation compared to naturally occurring
ATRA. Although the results also indicate a degree of neuronal
differentiation in cultures exposed to 10 lM 4b, the levels were
very much lower than those observed following either ATRA or 4a
treatment.
3502 | Org. Biomol. Chem., 2008, 6, 3497–3507
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 9 Differential regulation of nestin, a protein marker of neuropro-
genitor cells, in response to retinoids. Immunofluorescence micrographs
show nestin expression in TERA2.cl.SP12 cultures exposed to (A) 1 lM
or (B) 10 lM ATRA, (C) 1 lM or (D) 10 lM 4a and (E) 10 lM
4b for 21 days. Undifferentiated cells displayed relatively low levels of
nestin expression (F). Both concentrations of ATRA and 4a elicited a
strong nestin expression response (A–D), although unlike 1 lM ATRA
(A), 1 lM 4a (C) repeatedly showed consistently high levels homogenously
distributed throughout the culture. In cultures exposed to 4b (E), nestin
expression was restricted to discrete areas and the large flat cells within the
plaques which dominated the 4b cultures stained negative for nestin. Panel
(E) includes an inset to show Hoescht 33342 counterstaining indicating
the location of all cell nuclei in the culture. Scale bars: 50 lm.
Fig. 10 Induction of neural differentiation in cultures of TERA2.cl.SP12
cells exposed to retinoids. Immunofluorescence micrographs show ex-
pression of neurofilament-200 (NF200), a marker of mature neurons, in
TERA2.cl.SP12 cultures exposed to (A) 10 lM ATRA, (B) 10 lM 4a
or (C) 10 lM 4b for 21 days. TERA2.cl.SP12 stem cells did not express
NF200 (D). Cultures treated with ATRA and 4a formed large numbers
of NF200-positive neurons (A, B respectively) whereas those exposed to
4b developed very few neurons and such cells were located between the
epithelial plaques (C). Inset images to show Hoescht 33342 counterstaining
indicating the location of all cell nuclei in the culture (C, D). Scale bars:
50 lm. Quantification of the numbers of NF-200 positive cell perikarya
induced in response to 10 lM ATRA, 4a, or 4b, was determined from 6
randomly selected fields of view per condition (E). Data represent mean
SEM, n = 6.
Conclusions
In this study, we have designed, synthesised and tested the stability
of two synthetic retinoids, 4a and 4b, and directly compared their
stability and biological activity to that of naturally occurring
ATRA. We have shown that both 4a and 4b remain stable and
are resistant to light under conditions which cause isomerisation
and degradation of ATRA. Our studies suggest that there are
some significant advantages to using synthetic retinoids such as
4a and 4b over using ATRA in the cell culture laboratory since
this avoids potential problems associated with solutions of ATRA,
which are very susceptible to photoisomerisation and which may
subsequently influence their function to modulate cell activity.4 It
is important to note that as well as observing normal handling
procedures, light needs to be excluded particularly when carrying
out in vitro studies with RAs, whereas 4a and b can be handled
under normal ambient light conditions. At present, we can not be
certain that all of the biological effects of ATRA are replaced by
4; however, ongoing studies may reveal situations in which there
are discernable differences. In the meantime, cell biologists can be
confident of retinoid stability when using the synthetic compounds
tested herein, and that many of the observable effects of ATRA
examined in these studies are reproduced. Testing the biological
effects of the synthetic retinoids showed that they possess the
ability to induce the differentiation of human pluripotent stem
cells. However, 4a in particular displayed the ability to induce
neural tissues to a degree that was at least as effective, or better
than that observed for ATRA. In contrast, 4b induced cell
differentiation, resulting in distinctly different cultures which were
composed primarily of epithelial cells with few neural phenotypes.
These data indicate that subtle differences in the molecular
structure can have a dramatic effect upon the biological function,
exemplified in this study by the isomeric synthetic retinoids 4b and
4a. Such information regarding structure–activity-relationships
will be important in the future design and application of new
synthetic compounds aimed at controlling cellular development
processes. The observed similarity in the function of 4a to
ATRA highlights that the para-isomer synthetic retinoid 4a is
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3497–3507 | 3503
©

View Article Online
an exciting compound for use in stem cell science since it has the
added benefit of stability. It is noteworthy that 4a and ATRA
both show competitive binding towards RAR receptors, and
not to RXR receptors,11 but it is premature to be certain that
the observed biological effects reported here are directly linked
to this. Further studies in this area will be reported in due
course.
syringe. After 18 h at 70 ◦C, triethylamine was evaporated and the
residue was passed through a short silica gel column (hexane as
eluent) to give the crude product as a viscous, pale yellow oil after
evaporation which slowly solidified to give an off-white solid which
was recrystallised from ethanol to give the trimethylsilylacetylene
adduct 2a (1.0 g, 81%); mp 51–52 ◦C. dH (400 MHz; CDCl3) 7.40
(1H, m, Ar), 7.21 (2H, m, Ar), 1.66 (4H, s, CH2), 1.26 (6H, s, CH3),
1.25 (6H, s, CH3), 0.24 (9H, s, CH3); dC (126 MHz, CDCl3) 145.7,
144.8, 130.2, 129.0, 126.4, 120.0, 105.8, 92.6, 34.9, 34.8, 34.2, 34.1,
31.7, 31.6, 0.0; m/z (EI) 284 (M+); Anal. Calcd. for C19H28Si C,
80.21; H, 9.92; found C, 80.04: H, 9.90.
Experimental
General synthetic chemistry
All cross-couplings were carried out under an N2 atmosphere
using standard Schenk techniques. Triethylamine was distilled
over CaH2 under N2. Other reactions were performed in the
air in reagent-grade solvents. The compound 1,1,4,4-tetramethyl-
1,2,3,4-tetrahydronaphthalene was obtained from Maybridge and
used without further purification. Trimethylsilylacetylene was
obtained from Fluorochem, 3-iodobenzoic acid and methyl 4-
iodobenzoate from Aldrich, and 2-methylbut-3-yn-2-ol from Alfa
Aesar.
Melting points were obtained using a Gallenkamp melting point
apparatus. NMR spectra were performed on Varian Mercury
(1H) and Bruker Avance (13C) spectrometers with chemical shifts
referenced to solvent resonances and reported relative to tetram-
ethylsilane. EI-MS analyses were performed on an Agilent 6890
GC with a 5973 Inert MSD, using UHP helium as the carrier gas.
ES-MS analyses were performed on a Thermo-Electron Corp.
Finnagan LTQ-FT mass spectrometer. Elemental analyses were
obtained using an Exeter Analytical CE-440 analyzer. UV-vis
spectra were recorded in CHCl3 on a Hewlett-Packard 8453
diode array spectrophotometer using standard 1 cm quartz
cells.
(5,5,8,8-Tetramethyl-5,6,7,8-tetrahydronaphthalen-2-
ylethynyl)dimethylcarbinol 2b
PdCl2 (0.331 g, 1.87 mmol), Cu(OAc)2 (0.274 g, 1.87 mmol), 1
(5.0 g, 18.71 mmol) and PPh3 (2.45 g, 9.35 mmol) were placed in a
500 ml Schlenk flask, and the flask was evacuated and back-filled
with N2 gas (3 ×). Triethylamine (150 ml) was added via cannula,
followed by 2-methylbut-3-yn-2-ol (4.72 g, 56.13 mmol). The
solution was stirred under N2 at 70 ◦C for 3 days. Triethylamine was
evaporated and the residue was passed through a short silica gel
column (hexane, then 10% EtOAc–hexane as eluent). The EtOAc–
hexane solution was washed with 1 M HCl solution (100 ml), dried
(MgSO4) and evaporated to give 2b as an off-white solid (2.25 g,
45%); mp 107–109 ◦C; dH (500 MHz, CDCl3) 7.36 (1H, s, Ar),
7.24 (1H, d, J 7.8 Hz, Ar ), 7.21 (1H, d, J 7.8 Hz, Ar), 2.05 (1H, s,
OH), 1.67 (4H, s, CH2), 1.62 (6H, s, CH3), 1,27 (6H, s, CH3), 1.26
(6H, s, CH3); dC (126 MHz, CDCl3) 145.6, 145.1, 130.1, 128.9,
126.8, 119.8, 92.8, 82.8, 65.8, 35.2, 35.1, 34.5, 34.4, 31.9, 31.9,
31.8; m/z (EI) 270 (M+); HRMS calcd. for C19H26ONa ([MNa]+)
293.18759, found 293.18776, and HRMS calcd. for C19H25 ([M −
OH]+) 253.19508, found 253.19522.
The crystal and molecular structures of compounds 1, 2a, 3,
6, 7a, and 7b were determined by single-crystal X-ray diffraction,
details of which are provided in the ESI†.
6-Iodo-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene 6
To a mixture of 1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene
(3.76 g, 20 mmol), iodine (2.04 g, 8 mmol) and periodic acid (0.92 g,
4 mmol) was added glacial acetic acid (20 ml), water (4 ml) and
concentrated H2SO4 (98%, 1 ml). The reaction mixture was heated
6-Bromo-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene 112
To a solution of 1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphtha-
lene 5 (10.0 g, 53.0 mmol) in DCM (60 ml) at 0 ◦C under N2 was
added Br2 (15.58 g, 97.5 mmol). BF3·Et2O (8.27 g, 58.3 mmol) in
DCM (10 ml) was added dropwise over 2 h. The reaction mixture
was diluted with 40 : 60 EtOAc–hexane (150 ml) and washed
with saturated Na2SO3 solution (100 ml), saturated NaHCO3
solution (100 ml), and H2O (100 ml). The organic layer was
dried (MgSO4), filtered and evaporated to give a dark brown oil.
Kugelrohr distillation (120 ◦C, 8 × 10−3 mbar) gave 1 as pale yellow
◦
to 70 C for 4 h. A precipitate formed upon cooling, which was
filtered off, dissolved in hexane and passed through a short silica
gel column (hexane as eluent). The hexane was evaporated and
the residue was re-crystallised from ethanol to give iodide 6 as a
white crystalline solid (4.45 g, 71%); mp 69–70 ◦C; dH (200 MHz,
CDCl3) d 7.59–7.62 (1H, d, J 2 Hz, Ar), 7.40–7.47 (1H, dd, J 8 and
2 Hz, Ar), 7.01–7.07 (1H, d, J = 8 Hz, Ar), 1.66 (4H, s, 2 × CH2),
1.28 (6H, s, 2 × Me), 1.26 (6H, s, 2 × Me); dC (100 MHz, CDCl3)
147.7, 144.6, 135.6, 134.6, 128.7, 91.1, 34.9, 34.8, 34.3, 34.1, 31.8,
31.7; MS (EI): 314 (M+); Anal. Calcd. for C14H19I C, 53.52; H,
6.10; found C, 53.66; H, 6.13.
◦
crystals (11.02 g, 78%); mp 43–45 C; m/z (EI) 267 (M+); Anal.
Calcd. for C14H19Br: C, 62.93; H, 7.17; Found: C, 62.81; H, 7.16.
All 13C and 1H NMR data were identical to those reported in the
literature.12
6-Trimethylsilylethynyl-1,1,4,4-tetramethyl-1,2,3,4-
tetrahydronaphthalene 2a
Trimethyl(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-
ylethynyl)silane 2a
6-Iodo-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene (3.14 g,
10 mmol), Pd(PPh3)2Cl2 (0.07 g, 0.1 mmol) and CuI (0.02 g,
0.1 mmol) were added to a 250 ml Schlenk flask which was
evacuated and purged with N2 (3 ×). Triethylamine (150 ml) was
added via cannula under N2, followed by trimethylsilylacetylene
PdCl2 (75 mg, 0.43 mmol), Cu(OAc)2 (77 mg, 0.43 mmol), 1 (1.14 g,
4.30 mmol) and PPh3 (0.56 g, 2.14 mmol) were placed in a 250 ml
Schlenk flask under N2. Dry, degassed triethylamine (100 ml) was
added via cannula and TMSA (0.7 mL 5.14 mmol) added via
3504 | Org. Biomol. Chem., 2008, 6, 3497–3507
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
(1.18 g, 12 mmol). After 12 h at RT, triethylamine was evaporated
and the residue was passed through a short silica gel column
(hexane as eluent) to give the crude product as a viscous, pale
yellow oil after evaporation which slowly solidified to give an
off-white solid, which was recrystallised from ethanol to give
the trimethylsilylacetylene adduct 2a (2.50 g, 88%). All spectro-
scopic and analytical properties were identical to those reported
above.
3-(5,5,8,8-Tetramethyl-5,6,7,8-tetrahydronaphthalen-2-
ylethynyl)benzoic acid methyl ester 7b
Methyl 3-iodobenzoate (0.52 g, 2 mmol), Pd(PPh3)2Cl2 (0.014 g,
0.02 mmol) and CuI (0.004 g, 0.02 mmol) were added to a 250 ml
Schlenk flask which was evacuated and purged with N2 (3 ×).
Triethylamine (50 ml) was added via cannula under N2, followed
by 6-ethynyl-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene 3
(0.44 g, 2.1 mmol). The reaction mixture was stirred overnight at
RT, then heated at 70 ◦C for 1 h. After cooling, the triethylamine
was evaporated and the residue was passed through a short silica
gel column (hexane, followed by hexane–DCM, 4 : 1 as eluent) to
give the crude product, which was re-crystallised from ethanol, to
6-Ethynyl-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene 3
To
a solution of 6-trimethylsilylethynyl-1,1,4,4-tetramethyl-
1,2,3,4-tetrahydronaphthalene (1.42 g, 5 mmol) in methanol
(50 ml) and diethyl ether (50 ml), was added NaOH (0.14 g,
3.5 mmol) in water (2 ml). After 4 h, the mixture was ex-
tracted with diethyl ether, washed with water (3 ×), dried
(MgSO4) and evaporated to give 6-ethynyl-1,1,4,4-tetramethyl-
1,2,3,4-tetrahydronaphthalene 3 as a pale yellow oil which slowly
give 7b as a white crystalline solid (0.41 g, 59%); mp 100–101 ◦C;
−1
≡
=
tmax (KBr disc, cm ) inter alia (C C), 1717 (C O); dH (200 MHz,
CDCl3) 8.23 (1H, s, Ar), 7.97–8.01 (1H, d, J 5 Hz, Ar), 7.72–7.68
(1H, d, J 5 Hz, Ar), 7.50 (1H, s, Ar), 7.38–7.45 (1H, t, J 5 Hz, Ar),
7.28 (2H, s, Ar), 3.95 (3H, s, MeO), 1.70 (4H, s, 2 × CH2), 1.30
(6H, s, 2 × Me), 1.28 (6H, s, 2 × Me); dC (100 MHz, CDCl3) 166.5,
145.9, 145.2, 135.6, 132.7, 130.4, 130.0, 128.9, 128.7, 128.4, 126.7,
124.1, 119.8, 90.9, 87.2, 52.0, 35.0, 34.9, 34.4, 34.2, 31.8, 31.7; MS
(EI) 346 (M+); kmax (CHCl3) 289 nm (e 23000 M−1 cm−1); Anal.
calcd. for C24H26O2: C, 83.20; H, 7.56; found C, 82.99; H, 7.56.
solidified to give a white solid (0.78 g, 74%); mp 48–49 ◦C; tmax
−1
≡
(KBr disc, cm ) inter alia 2105 (C C); dH (200 MHz, CDCl3) 7.47
(1H, s, Ar), 7.27 (2H, s, Ar), 3.03 (1H, s, CH), 1.70 (4H, s, CH2),
1.29 (6H, s, 2 × Me), 1.26 (6H, s, 2 × Me); dC (100 MHz, CDCl3)
146.1, 145.1, 130.5, 129.2, 126.6, 119.1, 84.3, 75.9, 34.9, 34.8, 34.3,
34.2, 31.8, 31.7; m/z (EI): 212 (M+); Anal. calcd. for C16H20 C,
90.51; H, 9.49; found C, 90.27; H, 9.57.
4-(5,5,8,8-Tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)-
benzoic acid 4a
A solution of 4-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphtha-
len-2-ylethynyl)benzoic acid methyl ester 7a (0.35 g, 1 mmol)
in THF (20 ml) was treated with aqueous 20% NaOH (20 ml).
After heating at 70 ◦C for 20 h, the reaction mixture was diluted
with diethyl ether (150 ml) and water (150 ml), then 1 M HCl
solution was added until mixture reached pH 1. The ether layer
was separated, dried (MgSO4) and evaporated to give an off-white
powder, which was re-crystallised from acetonitrile to give the
Methyl 3-iodobenzoate
A solution of 3-iodobenzoic acid (24.8 g, 100 mmol) in methanol
(300 ml) was treated with sulfuric acid (98%, 5.0 g). After heating
at 70 ◦C for 5 h, the mixture was extracted with diethyl ether,
washed with water (3 ×), dried (MgSO4) and evaporated. The
residue was passed through a short silica gel column (hexane
as eluent), to give the product as a white crystalline solid
(21.1 g, 81%); mp 53–54 ◦C (lit. 54–55 C1). All spectroscopic
and analytical properties were identical to those reported in the
literature.16
◦
product 4a as a white crystalline solid (0.24 g, 72%); mp 254–
256 ◦C; tmax (KBr disc, cm ) inter alia 2205 (C C), 1681 (C O);
dH (400 MHz, CDCl3) 8.09 (2H, d, J 8.5 Hz, Ar), 7.62 (2H, d, J
8.5 Hz, Ar), 7.51 (1H, s, Ar), 7.31 (2H, s, Ar), 1.68 (4H, s, CH2),
1.31 (12H, s, Me); dC (125 MHz, CDCl3) 171.5, 146.5, 145.4, 131.7,
130.4, 130.3, 129.6, 129.1, 128.4, 127.0, 119.7, 93.9, 87.8, 35.1, 35.0,
34.6, 34.5, 32.0, 31.9; m/z (ESI) 377 (M + 2Na), 331 (M − H); kmax
(CHCl3) 310 nm (e 26900 M−1 cm−1); HRMS calcd. for C23H23O2
331.16926 (M − H), found 331.16949.
−1
≡
=
4-(5,5,8,8-Tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)-
benzoic acid methyl ester 7a
Methyl 4-iodobenzoate (3.59 g, 13.7 mmol), Pd(PPh3)2Cl2 (0.09 g,
0.14 mmol) and CuI (0.03 g, 0.14 mmol) were added with stirring to
a 500 ml Schlenk flask. After evacuation and purging with N2 (3 ×),
triethylamine (200 ml) was added via cannula under N2, followed
by 6-ethynyl-1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphthalene 3
(3.50 g, 16.5 mmol). After 18 h at RT, the triethylamine was
evaporated and the residue was passed through a short silica gel
column (hexane, followed by hexane–DCM, 4 : 1 as eluent) to
give the crude product, which was re-crystallised from ethanol to
give the product 7a as a white crystalline solid (4.13 g, 87%); mp
3-(5,5,8,8-Tetramethyl-5,6,7,8-tetrahydronaphthalen-2-ylethynyl)-
benzoic acid 4b
A solution of 3-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphtha-
len-2-ylethynyl)benzoic acid methyl ester 7b (0.35 g, 1 mmol)
in THF (20 ml) was treated with aqueous 20% NaOH (20 ml).
After heating at 70 ◦C for 20 h, the reaction mixture was diluted
with diethyl ether (150 ml) and water (150 ml), then 1M HCl
solution was added until mixture reached pH 1. The ether layer
was separated, dried (MgSO4) and evaporated to give an off-white
powder. The product was re-crystallised from acetonitrile to give
128–129 ◦C; tmax (KBr disc, cm ) inter alia 2207 (C C), 1712
−1
≡
=
(C O); dH (400 MHz, CDCl3) 8.03 (2H, d, J 9 Hz, Ar), 7.59 (2H,
d, J 9 Hz, Ar), 7.51 (1H, s, Ar), 7.31 (2H, s, Ar), 3.94 (3H, s,
MeO), 1.71 (4H, s, 2 × CH2), 1.28 (12H, s, Me); dC (100 MHz,
CDCl3) 166.8, 146.3, 145.4, 131.7, 130.3, 129.7, 129.4, 129.0, 128.6,
127.0, 119.8, 93.3, 87.2, 52.4, 35.2, 35.1, 34.6, 34.5, 32.0, 31.9; kmax
(CHCl3) 310 nm (e 26400 M−1 cm−1); Anal. calcd. for C24H26O2 C,
83.20; H, 7.56; found C, 83.03; H, 7.59.
4b as a white crystalline solid (0.26 g, 78%); mp 223–224 ◦C; tmax
−1
≡
=
(KBr disc, cm ) inter alia 2210 (C C), 1690 (C O); dH (200 MHz,
CDCl3) 8.25 (1H, s, Ar), 8.00–8.04 1H, (d, J 5 Hz, Ar), 7.76–7.80
(1H, d, J 5 Hz, Ar), 7.50 (1H, s, Ar), 7.42–7.50 (1H, t, J 5 Hz,
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3497–3507 | 3505
©

View Article Online
Ar), 7.28 (2H, s, Ar), 1.70 (4H, s, CH2), 1.32 (6H, s, 2 × Me), 1.30
(6H, s, 2 × Me); dC (100 MHz, d6-DMSO) 167.0, 146.3, 145.5,
135.7, 132.4, 131.9, 130.1, 129.7, 129.0, 127.4, 123.4, 119.5, 91.1,
87.8, 34.9, 34.8, 34.5, 34.4, 31.9, 31.8; kmax (CHCl3) 289 nm (e 29
000 M−1cm−1); Anal. calcd. for C23H24O2 C, 83.10; H, 7.28; found:
C, 82.38; H, 7.24; HRMS calcd. for C23H23O2 331.16926 (M − H);
found 331.17050.
determined to indicate the status of cellular differentiation by
TERA2.SP12 cells.9 Suspensions of single EC cells or their
differentiated derivatives were formed by the addition of 1 ml
0.25% trypsin/EDTA solution and cell numbers determined using
a haemocytometer. Cells were added to a 96-well plate (0.2 × 106
cells per well) for incubation with primary (1 : 20) and FITC-
conjugated secondary antibody (Cappell, 1 : 100) as previously
described.9,15 Labelled cells were analysed in a BD Biosystems
FACSFlow cytometer. Thresholds determining the numbers of
positively expressing cells were set against the negative control
antibody P3X.
Exposure to light
ATRA and the synthetic retinoids NMR samples were prepared
and left to stand in ordinary laboratory light, with additional
irradiation from a standard fluorescent light ca. 30 cm above the
sample. The spectral characteristics of the light were measured
using a spectrophotometer and are shown in Fig. 11.
Immunocytochemistry
Cells were grown in 12-well culture plates for immunocytochemical
analysis. Confluent cultures of stem cells and their differentiated
derivatives were fixed in ice-cold 4% paraformaldehyde (Sigma) in
phosphate-buffered saline (PBS) for 30 min at room temperature
(RT), followed by 3 washes in PBS. Cell membranes were perme-
ablised by treatment with 1% Triton-X-100 (Sigma) in PBS for
10 min at room temperature. Non-specific binding of antibodies
was blocked using a solution of 1% goat serum (Sigma) in PBS
containing 0.2% Tween-20 (Sigma). Fixed cells were incubated
with the blocking solution on a bench-top shaker for 30 min
at room temperature. Primary antibodies: nestin (Chemicon,
1 : 100); NF-200 (Sigma 1 : 100); cytokeratin-8 (Sigma, 1 : 100) were
diluted in blocking solution and incubated with the cells for 1 h
at RT. After washing, cells were incubated in FITC-conjugated
secondary antibody (Alexafluor 488 1 : 600) for 60 min. After
washing in PBS, samples were mounted under coverslips using
Vectashield, (Vector Labs) containing 1 lg ml−1 Hoechst 33342
nuclear staining dye (Molecular Probes). Imaging was performed
on a fluorescent microscope under restricted light conditions
(Nikon Diaphot 300). Fluorescence photomicrographs, including
Hoecsht 33342, were acquired using the appropriate filter sets and
an adapted digital camera (Nikon).
Fig. 11 Red line = room lights (note shortest wavelength feature 366 nm).
Light blue = lamps used for irradiation of the samples, with a short
integration period. Dark blue = lamps used for the sample irradiation,
with a longer integration period (note shortest wavelength feature 366 nm,
similar to room light).
Tissue culture
Human pluripotent TERA2.cl.SP12 embryonal carcinoma stem
cells were maintained under standard laboratory conditions as
described by Przyborski.9 In brief, cells were cultured in DMEM
(Sigma) supplemented with 10% FCS (Gibco), 2mM L-glutamine
and 100 active units each of penicillin and streptomycin (Gibco).
Cultures were passaged using acid-washed glass beads (VWR)
unless a single-cell suspension was required for counting, in which
case a 0.25% trypsin EDTA (Cambrex) solution was used. Cultures
intended for flow cytometric analysis were set up in T25 flasks
(Nunc) while 12-well plates (Nunc) were used for MTS assays and
immunocytochemical studies. In preparation for use in cell culture
experiments, stock solutions of synthetic retinoids and ATRA
(Sigma) were prepared in DMSO (Sigma) to concentr◦ations of
10mM. Aliquotted stock solutions were stored at −80 C in the
dark.
Acknowledgements
We thank the MRC for a studentship to C. E. B., the EPSRC for a
DTA studentship to J. H. B., Durham County Council for a stu-
dentship to V. B. C., ReInnervate Ltd, North East England Stem
Cell Institute (NESCI), One NorthEast and Durham University
for support, J. Dostal for carrying out elemental analyses, and
the EPSRC Mass Spectrometry Service at the University of Wales,
Swansea. We also thank Dr A. Beeby (DU) for spectrophotometric
measurements.
References
1 (a) V. B. Christie, T. B. Marder, A. Whiting and S. A. Przyborski, Mini-
Rev. Med. Chem., 2008, 8, 601–608; (b) M. Maden, Nat. Rev. Neurosci.,
2007, 8, 755–765.
Flow cytometry
2 L. Altucci and H. Gronemyer, Nat. Rev. Cancer, 2001, 1, 181–193.
3 N. P. Mongan and L. J. Gudas, Differentiation, 2007, 75, 853–870.
4 (a) A. Murayama, T. Suzuki and M. Matsui, J. Nutr. Sci. Vitaminol.,
1997, 43, 167–176; (b) D. K. Bempong, I. L. Honigberg and M. N.
Meltzer, J. Pharm. Biomed. Anal., 1995, 13, 285–291; (c) T. Suzuki, S.
Rao Kunchala, M. Matsui and A. Murayama, J. Nutr. Sci. Vitaminol.,
1998, 43, 729–736; (d) S. Rao Kunchala, T. Suzuki and A. Murayama,
Ind. J. Biochem. Biophys., 2000, 37, 71–76.
Flow cytometric analysis was carried out on live cells using
antibodies recognising cell surface markers. The expression of
markers indicative of the stem cell (SSEA-3 and TRA-1-60
(generous gift from Prof. P. Andrews, University of Sheffield))
or neural cell (A2B5 (Abcam) and VINIS-53 (generous gift
from Prof. P. Andrews, University of Sheffield)) phenotype was
3506 | Org. Biomol. Chem., 2008, 6, 3497–3507
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
5 (a) G. Y. Han, B. S. Chang, M. J. Connor and N. Sidell, Differentiation,
1995, 59, 61–69; (b) M. Lansink, A. M. van Bennekum, W. S. Blaner
and T. Kooistra, Eur. J. Biochem., 1997, 247, 596–604; (c) H. Han, J.
Kwon, M. Park, S. Park, S. K. Cho, Y. Rho, J. Kim, H. Sin and S. Um,
Bioorg. Med. Chem., 2003, 11, 3839–3845.
6 R. Alvarez, M. J. Vega, S. Kammerer, A. Rossin, P. Germain, H.
Gronemeyerb and A. R. Lera, Bioorg. Med. Chem. Lett., 2004, 14,
6117–6122.
7 (a) F. Mavilio, A. Simeone, E. Boncinelli and P. W. Andrews, Differenti-
ation, 1988, 37, 73–79; (b) A. Simeone, D. Acampora, L. Arcioni, P. W.
Andrews, E. Boncinelli and F. Mavilio, Nature, 1990, 346, 763–766.
8 (a) P. Venepally, L. G. Reddy and B. P. Sani, Arch. Biochem. Biophys.,
1997, 343, 234–248; (b) R. W. Curley and J. W. Fowble, Photochem.
Photobiol., 1988, 47, 831–835.
11 C. J. Gambone, J. M. Hutcheson, J. L. Gabriel, R. L. Beard, R. A. S.
Chandraratna, K. J. Soprano and D. R. Soprano, Mol. Pharmacol.,
2002, 61, 334–342.
12 (a) G. Garipova, A. Gautier and S. R. Piettre, Tetrahedron Lett., 2005,
61, 4755–4759; (b) H. Kagechika, E. Kawachi, Y. Hashimoto, T. Himi
and K. Shudo, J. Med. Chem., 1988, 31, 2182–2192.
13 (a) M. Ebisawa, K. Ohta, E. Kawachi, H. Fukasawa, Y. Hashimoto and
H. Kagechika, Chem. Pharm. Bull., 2001, 49, 501–503; (b) V. Vuligonda,
S. M. Thacher and R. A. S. Chandraratna, J. Med. Chem., 2001, 44,
2298–2303.
14 H. Suzuki, Org. Synth., 1971, 51, 94–96.
15 (a) S. A. Przyborski, I. E. Morton, A. Wood and P. W. Andrews,
Eur. J. Neurosci., 2000, 12, 3521–3528; (b) R. Stewart, V. Christie and
S. A. Przyborski, Stem Cells, 2003, 21, 248–256.
9 (a) S. A. Przyborski, Stem Cells, 2001, 19, 500–504; (b) S. A. Przyborski,
V. B. Christie, M. W. Hayman, R. Stewart and G. M. Horrocks, Stem
Cells Dev., 2004, 13, 400–408.
16 H. H. Wuest, F. F. Frickel and A. Nuerrenbach (BASF A.-G., Fed.
Rep. Ger.), Ger. Offen. CODEN: GWXXBX DE 3434946 A1 19860403,
1986, 23 pp.; Pat. Appl. DE 84-3434946 19840922; Chem. Abstr., 105,
172069.
10 H. A. Bruson and J. W. Kroeger, J. Am. Chem. Soc., 1940, 62, 36–44.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3497–3507 | 3507
©