Subcellular Localization-function Relationship
Study in Human Antiquitin
CHAN, Chi Lung
A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Philosophy in Biochemistry
The Chinese University of Hong Kong September 2011 ITi f ？ I 4/J - %、、-r. I J . ；(-^T、^ ”T; ^ J T x I 、 2 I . ！ \ s i J ： : / i , 一 •(..，，) ， 、； -、.... 少 ！ 一，力〕
li
. r .
. : .〈v Thesis/Assessment Committee
Professor Christopher H.K. Cheng (Chair)
Professor Wing-Ping Fong (Thesis Supervisor)
Professor Kam-Bo Wong (Committee Member)
Professor Shih-Jiun Yin (External Examiner)
I Doc la ration
1 doclarc that the thesis here submitted is original except lor sourcc material explicitly acknowledged, and thai the same or related material has not been previously submitted tor the same or dilTerent courses. I also acknowledge thai 1 am aware of University policy and regulations on honesty in academic work, and of the disciplinary guidelines and procedures applicable to breaches of such policy and regulations, as contained in the website http://www.cuhk.edu.hk/pQlicy/academichonesty/.
16/9/2011 Signature Date
CHAN CHI LUNG 1009045100 Name Student ID
BCHE8012 Thesis Research Course code Course title
\ii Ackiiowlcdjiciiiciits
It IS my pleasure to lake this opportunity to express my most sinccrc graliludc
to my supervisor Prol： Wing-Ping Fong for his patient guidance throughout the study
and his invaluable advice in the preparation of this thesis.
Many grateful thanks are due to Prof. Christopher U.K. Cheng and Prof.
Kam-Bo Wong, especially to my external examiner Prof. Shih-Jiun Yin, National
Defense Medical Center, Taiwan, for their willingness to be members of my thesis
committee.
Special thanks go to my colleagues Ms. Judy Wong, Ms. Esther Yeung and
Ms. Pui-Wah Choi for their patience and technical support.
I am particularly grateful to my friends for their encouragement during the preparation of this thesis. Last but not least, I would like to make my particular and precious thankfulness and appreciation to my beloved family, especially during the sickness of me and my dearest mother, for their encouragement and their pray for my research life.
Financial supports from the Research Grant Council of Hong Kong Special
Administrative Region (Project No. 464407) and The Chinese University of Hong
Kong are gratefully acknowledged.
lii 摘处
Antiquitin(ALl)lI7)乃f彳？肌(丨iffe (ALl)丨丨）超級家族之成i丨’從它Y丨：人ft'丨
物的iiUiii牧排丨？；中rtfl丨似率A'l丨，它‘柯沾化.IJh彳沐:广〕:之制‘丨。1人1此
ALD117呢忍ra扮演iRi剧〈丄卞现作川。Gyi巡--步r解此iiin ’我們池tr广多力
ifif的研究。
在對人體ALDH7 A1亞細胞足位的研究1-11 ’經計党機模擬分析(iPSORT)及
驗研究均發現一段長28粒氨基酸的肽位於它的氨基端。此肽爲一腺粒體目標訊
號。利用ALDH7A1-增強綠色螢光蛋白偶聯物，結果發現ALDH7A1的腺粒體及胞
液亞細胞定位是憑有與沒有該氨基末端腺粒體目標訊號在該蛋白上達到。流式細
胞儀對從細胞中分離的腺粒體及亞細胞分離法分析顯示腺粒體中存在自然的
ALDH7A1。免疫螢光染色法配合ALDH7A1抗體於WRL68細胞染色展現腺粒體、
胞液及潛在的細胞核亞細胞定位。ALDH7A1的腺粒體亞細胞定位與其發生於腺
粒體基質的賴安酸(Lysine)分解代謝的生理作用一致。
在生理作用的層面上，除了已知ALDH7A1多樣的生理作用外，免疫螢光
染色展示ALDH7A1存在於細胞核，這發現建議它可能有扮演控制細胞生長及控
制細胞周期的角色。在這硏究中，ALDH7A1的表達水平於被同步的嬰兒肝細胞
株WRL68及嬰兒賢細胞株HEK293受到監測。實驗結果發現ALDH7A1表達水平於
&-S相變期間上調。於被同步的細胞中利用免疫螢光染色展示出ALDH7A1於Gi-S
相變期間的蛋白表達增加及於細胞核中積聚。利用小髮夾核醣核酸（shRNA)敲
除ALDH7A1的細胞轉染實驗發現個別調節細胞周期的蛋白表達水平有改變。然
而，於細胞生長率的分析中未能發現被敲除ALDH7A1的細胞有明顯的生長率變
化。
ALDH7A1於癌細胞的表達水平利用蛋白印跡法進行檢測，因爲癌細胞一
般丨M現出不正常細胞生長及細胞周期。結束顯示ALDH7A1在癌細胞並非必然地
h调。於超低附乾性培養碟中培養的細胞，與其二維細胞比較之後亦沒丨/必然的
。較前的硏究中報〈I?ALDH7A1Y丨:iW列腺癌幹細胞的過
In ti^A^速，於li!：次研究中利川iW列腺松細胞株1)⑴/I川1他fif ’扣M的趨
_剛浏彳似‘;細胞RMG1中IRJii。Ai丨胞液及腺粒體AL1)H7八1的佔丨人I及催化突變耶
C302S 的n粒败轉染节細胞逃行過敗农述。轉染的細胞過度招紅Hiif -
ALDH7A1都)/《小丨I丨較空俄體控制W驗iJI 1快的細胞生书。）J站见標小
ALDH7A1以非催化形式濟在參與控制細胞先長。S记轉染的細胞亦被川作測試
ALDH7A1對Aldefluoi•受質的活撥性，Aldefluor—般被視爲辨認幹細胞的「黄金標
準」。結果顯示ALDH7A1對Aldefluor有活撥性’這些結果亦帶出利用Aldefluor測
試作爲辨認及分離幹細胞的可靠性的重新評估。
\ Abstract
Ami quit in (ALl)117) belongs to the supcrfamily of aldehyde dehydrogenase
(AU)U). Il is an evolutionarily conserved protein as shown from the high amino acid sequence identity belwecn human and plant countcrparls. 1 hcrcforc, ALI)I17A1 is believed to have important physiological functions. To further investigate this protein, ditYerent attempts have been made in this study.
In subcellular localization study of ALDH7A1, it was found in both in silica iPSORT analysis and in ri/ro studies that a 28 amino acid peptide is potentially located at the N-terminal of ALDH7A1. This peptide is a mitochondrial-targeting signal. Studies using ALDH7A1 -enhanced green fluorescence protein conjugates showed that the mitochondrial and cytosolic localizations of ALDH7A1 could be achieved by the presence and absence of the mitochondrial-targeting sequence in the
N-terminal. Flow cytometric analysis of isolated mitochondria and subcellular fractionation indicated the presence of ALDH7A1 in the mitochondria. The mitochondrial localization of ALDH7A1 is consistent to its physiological function in lysine catabolism which takes place in the mitochondrial matrix. On the other hand, immuno-fluorescence staining of WRL68 cells using anti-ALDH7Al antibody revealed mitochondrial, cytosolic as well as potential nuclear localizations of
ALDH7A1.
The presence of ALDH7A1 in the nucleus suggests that it may also play a role in cell growth and cell cycle progression. In this investigation, the expression level of
ALDH7A1 was monitored in synchronized WRL68 and HEK293 cell lines. The protein was up-regulated during Gi-S phase transition, especially in the nucleus as shown by immuno-fluorescence staining. Knockdown of ALDH7A1 using shRNA transfcction resulted in changes in the levels of several key cell cycle-regulating proteins. However，it did not cause any significant changes in the rale ol'ccll growth.
\ i C^inccr cclls have abnormal ccll growth and ccll cyclc. Thcrdoru, llic expression level of AU)117八 1 was checked to see if the abnomalitics arc related to
ALD117A1 de-regulation. It was observed that ALDI17A1 was not obligatorily up-regulated in cancer cells as compared to normal cells such as I IEK293 or WR1.68.
Recent study reported the over-expression of ALDH7A】 in prostate cancer cells, especially in the cancer stem cells. Similar results were obtained in this study using
DU145 prostate cancer cells as a model and surprisingly the trend recurs in the ovarian cancer cells RMGl. Cells cultured in ultra-low attachment plate, a system used for culturing stem-like cells, showed no obligatory iip-regulation of ALDH7A1 as compared to their 2D cell counterparts. To see whether the potential effect on cell growth is due to ALDHTAl's catalytic activity, plasmids containing cytosolic and mitochondrial ALDH7A1 genes as well as a catalytic mutant C302S were transfected to cells to allow over-expression. Stable transfectants overexpressing either one of these constructs demonstrated enhanced growth rate compared to the empty vector control. The results indicated that ALDH7A1 might affect cell growth in a non-catalytic manner. The stable transfectants were also tested for the reactivity of
ALDH7A1 towards the Aldefluor substrate which is the ‘golden standard’ for identifying stem cells with ALDHlAl overexpression. The results pinpointed the reactivity of ALDH7A1 towards the substrate and these results call for a re-evaluation of the reliability of the Aldefluor assay in stem cell identification and isolation.
\ ii List of Ahhrcviatioiis a-AASA a -Ainiiioadipic scniialdchydc a-AAA a -Aminoadipic acid
ALDl I Aldehyde dehydrogenase
ALDH+ Aldefluor-active cell population
ALDH7 Antiquitin
ALDH7A1 Human antiquitin
ATCC American Type Culture Collection
BAAA-DA BODIPY-aminoacetaldehyde diethyl acetal
BAA BIDOPY-aminoacetate
BCA Bicinchoninic acid
3-FGF P-Fibroblast growth factor
BSA Bovine serum albumin
CDE Cell cycle dependent element cDNA Complementary deoxyribonucleic acid
C. elegems Caenorhabditis elegans
CHR Cell cycle genes homology region
Cox IV Cytochrome C oxidase subunit IV
CSC Cancer stem cell
DEAR Diethyl aminobenzaldehyde
DMEM Dulbecco's Modified Eagle's Medium dNTP Deoxynucleoside triphosphate dsRNA Double-stranded ribonucleic acid dsDNA Double-stranded deoxyribonucleic acid
E. coli Escherichia coli
WW l-nhaiiccd chciniluniincsccncc
I" I) I A Mthylcncdiaiiiinctclraacclic acid
1 Epidermal growth factor
[{nhanccd green niioresccncc protein
Fi rC Fluorescein isothiocyanale
FL Full length
FSC Forward side scatter
HEPES N-2-Hydroxyethylpiperazine-N'-2-ethanesulfonic acid
HRP Horseradish peroxidase
IPTG Isopropyl-(3-D-thiogalactopyranoside
K Kelvin degree
LB Luria-Berlani medium
MOPS 3-Morpholinopropanesulfonic acid
MTS Mitochondrial-targeting sequence
MTR Mito Tracker Red CMX-Ros
NAD+ Nicotinamide adenine dinucleotide
NAD(P)+ Nicotinamide adenine dinucleotide (phosphate)
NCBI National Center for Biotechnology Information
NES Nuclear export signal
NLS Nuclear localization signal
OH Hydroxyl radical
ORF Open reading frame
P5C Pyrroline-5-carboxylate
P6C Piperideine-6-carboxylate
PBS Phosphate bu lie red saline
i\ Polymerase chain read ion
Protein data bank
Pyridoxine-dependcnl seizure
PH Phycocrythrin
PI Propidiiim iodide
PIG-A Phosphatidylinositol glycan class A
PLP Pyridoxal-5'-phosphate
PMSF Phenylmethylsulfonylfluoride
PPAR Peroxisome proliferators-activated receptor
PVDF Polyvinylidene difluoride
ROS Reactive oxygen species
RPMI Roswell Park Memorial Institute-invented culture medium
RXR Retinoid X receptor
SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis shRNA Small hairpin ribonucleic acid
SSC Side scatter
SLS Sjogren-Larsson Syndrome
TBS-T Tris-buffered saline-Tween-20
Tris Tris(hydroxymethyl)aminomethane
\ List of res
I' iguic 1.1 The nineteen A 1.1)11 genes in human and their evolution relationship
Figure 1.2 Structure of human ALD1I7A1 monomer
Kigure 1.3 Proposed catalytic mechanism of non-CoA dependent AL13I I towards
aldehyde substrates
Figure 2.1 Location of the potential mitochondrial-targeting signal (MTS) at the
N-terminus of ALDH7A1
Figure 2.2 Western blot analysis of ALDH7A1 in different subcellular fractions
of WRL68 cells
Figure 2.3 Flow cytometric analysis of isolated mitochondria from WRL68 cells
Figure 2.4 Immunofluorescence analysis of ALDH7A1 subcellular localization
in WRL68 cells
Figure 2.5 Confocal microscopy of WRL68 cells transiently expressing different
ALDH7A1-EGFP conjugates
Figure 2.6 The role of ALDH7A1 as an a-aminoadipic semialdehyde (a-AASA)
dehydrogenase
Figure 3.1 Analysis of ALDH7A1 in cells synchronized at different phases of
the cell cycle
Figure 3.2 Immunofluorescence analyses of ALDH7A1 subcellular distribution
in (A) WRL68 and (B) HEK293 cells synchronized at different
phases of the cell cycle
Figure 3.3 Western blot analyses on cells transfected with ALDH7A1 -shRNA
Figure 3.4 Growth analysis of transfected WRL68 cells by counting the cell
number
Mgure 3.5 Expression of ALDH7A1 in BL21 (DE3)pLysS cells and isolation ot
\i the ovcr-cxpressed proteins
Mgurc 3.6 C)vcr-cx press ion of various ALDl I7A1 constructs in 111:K293 cells
I'igurc 3.7 Growth analysis of trans fee ted IIPX293 cells by counting the ccll
number
Figure 3.8 Expression of ALDH7A1 protein in normal and cancer cell lines in
2D and ultra-low attachment culture
Figure 3.9 ALDH7A1 expression level in 2D and stem-like ovarian cancer cells
Figure 3.10 Flow cytometric analysis on cells over-expressing ALDH7A1
variants stained by Aldefluor reagent
Figure 3.11 Nuclear localization signal (NLS) and nuclear export signal (NES) in
ALDH7A1
Figure 3.12 Morphology of cells cultured in ultra-low attachment plate
Figure 3.13 A schematic diagram for the Aldefluor assay
\ii List of Tables
Tahlc 2.1 Comparison of I he putative M TS of ALDI17A1 with those ol" known
mitochondrial proteins
Table 2.2 Nucleotide sequence ol、the forward and reverse primers used for
cloning the various constructs of ALDM7A1
Table 2.3 Prediction of subcellular localization of antiquitins from different
species
Table 3.1 Experimental conditions used for synchronizing cells to different
phases of the cell cycle
Table 3.2 Activity of recombinant ALDH7A1 and the catalytic mutant C302S
\iii l ahlc of Content rhcsis Assessment Coniniitlcc i
Declaration ii
Acknowledgements iii ii^r^ iv
Abstract vi
List of Abbreviations viii
List of Figures xi
List of Tables xiii
Table of Content xiv
Chapter 1 General Introduction
1.1 Classification of the aldehyde dehydrogenase siiperfamily 1
1.2 Structures and catalytic mechanism of ALDH 4
1.3 Multiple functions of ALDH 8
1.4 Antiquitin — background and recent discoveries 12
1.5 Aim of study 19
Chapter 2 Mitochondrial and Cytosolic Localizations of ALDH7A1
2.1 Introduction 21
2.2 Materials and Methods 26
2.2.1 Cell culture 26
2.2.2 Subcellular fractionation 26
2.2.3 Western blot analysis 27
2.2.4 Flow cytometric analysis of mitochondria in WRL68 cells 28
2.2.5 Transient transfection of various EGFP constructs 29
2.2.6 Immunofluorescence staining 31
\i\ ：..^ Rcsulls V、
2.3.1 Pfcscncc of ALDl 17A1 in cylosol and mitochondria in WRI/)8 cclls 33
2.3.2 Milochondrial-targeling N-terminal sequence in ALDl I7A1 34
2.4 Discussion 40
2.4.1 In si/ico and in vilro subcellular localization studies on ALDH7A1 40
2.4.2 Significance of mitochondrial and cytosolic localizations of ALDH7A1 ..45
2.4.3 Comparison of animal ALDH7A and plant ALDH7B enzymes 48
Chapter 3 ALDH7A1: A Potential Regulator for Cell Growth, Cell Cycle and a
Potential Biomarker for Cancer (Stem) Cells
3.1 Introduction 51
3.2 Materials and Methods 55
3.2.1 Cell synchronization 55
3.2.2 Semi-quantitative determination of DNA amount in synchronized cells...55
3.2.3 Total protein extraction 55
3.2.4 Western blot analysis 57
3.2.5 Immunofluorescence staining 57
3.2.6 Expression and purification of ALDH7A1 and its mutant 57
3.2.7 Kinetic analysis of ALDH7A1 and its mutant 58
3.2.8 Generation of native ALDH7A1 and mutant for transfection 58
3.2.9 Generation of stable cell line transfectants 59
3.2.10 2D cell culture and ultra-low attachment cell culture 59
3.2.11 Collection of total cell lysates 60
3.2.12 Western blot analysis 60
3.2.13 Growth analysis 61
3.2.14 AldcHuor assay 61
w Results 62
3.3.1 I'xprcssioii level ofALDl 17A1 at diffcrcnl phases of the ccll cyclc 62
3.3.2 Subcellular distribution ofAUJI I7A1 in synchroni/cd cclls 64
3.3.3 Changes in the expression level of key cell cycle regulators and ihc growth
rate alter ALDH7A1 knockdown 68
3.3.4 Absence of catalytic activity in the purified ALDI I7A1 mutant C302S....68
3.3.5 Over-expression of ALDH7A1 variants in HEK293 cells 73
3.3.6 Growth rates of cells overexpressing different ALDH7A1 variants 73
3.3.7 Expression level of ALDH7A1 in various 2D cell types and stem-like
cells 76
3.3.8 Aldefluor assay on cells over-expressing different ALDH7A1 variants...79
3.4 Discussion 82
3.4.1 Nuclear localization of ALDH7A1 82
3.4.2 Potential role of ALDH7A1 in cell cycle 86
3.4.3 Non-catalytic role of ALDH in cell growth and development 86
3.4.4 Relationship between ultra-low attachment culture and stem-like cells....89
3.4.5 Up-regulation of ALDHs in cancer and CSCs and the evaluation of
applicability of Aldefluor assay in CSC isolation 93
3.4.6 Comparison on ALDH7A1 expression level in primary and stem-like
cells 98
Chapter 4 Future Prospects
References 103
\\ I (liiipttr 1 (aiuial Introduction
1.1 ( lassificatioii of the aldehyde deliydrojijenase superfamily
ogcnasc (ALDl I; 1.2.1.3) refers to a category ol'cn/.ymcs
catalyzing the conversion of aldehydes to carboxylic acids with the help of、the
cofoctor nicotinamide adenine dinucleotide (NAD') or nicotinamide adenine
dinucleotide phosphate (NADP+). Sequence comparison between bacterial and human
ALDHs reveals extensive similarities, indicating that the ALDH genes are originated
from a common ancestral gene (Vasilioii el a I., 1999). The high similarities of the
ALDH genes in different organisms also pinpoint the cellular physiological
importance of this superfamily of proteins. The ancestral ALDH proteins are believed
to be essential for organisms to counteract the toxic effects of aldehydes generated in their living habitats with extraordinary high temperature and high oxygen level.
Endogenous aldehydes can be generated, among others, through peroxidation of lipids in plasma membrane and amino acid degradation during protein catabolism. These aldehydes are so reactive that they damage many biomolecules such as proteins and
DNA which are rich in chemical groups like thiol and amino groups. The strong electrophilic nature of aldehydes will therefore lead to undesirable reaction and inactivation of these biomolecules, causing disturbance in normal cellular functions and more seriously cell death. The presence of extensive members of ALDH and the evolutionarily conserved nature of ALDHs pinpoint the needs for clearance of various forms of endogenous as well as exogenous reactive aldehydes for maintaining the normal physiology of organisms.
Members of ALDH are classified using a standardized gene nomenclature system (Vasiliou e/ aL, 1999). They are categorized into families and subfamilies based on the percentage of amino acid identity. ALDH proteins with amino acid scqucncc identity equal to or greater than 40% arc assigned to a particular faiiiih
1 (designated h\ an .Arabic luinibcr, such as the、T m 八whereas those shariiiu amino acid soqucncc identity greater than 60% arc grouped in the same sLiMamily
(designated by a Idler, such as the ‘八’ in AU)I17八).These cul-olT values Ibllovv the original rccomnicndalion by Margaret DayholT who llrst applied to cn/ymcs of ihc
CN lochromc P-450 supciiamily (Negcrl and Wain, 2003).
In the ALDH siipeiiamily, there arc 19 functional genes and 3 pseudogencs found in the human genome (Vasiliou and Nebert, 2005). The functional ALDH genes, with distinct chromosomal location, are classified based on the nomenclature mentioned above as shown in Figure 1.1. Gene Chromosorno
「-ALDH4A1 1p36.13
LALDH7A1 5q31
J-ALDH1A1 9q21.13
R—\^ALDH1A2 15q22,1
_ALDH1A3 15q26,3
r- 「ALDH1B1 9p11.1 LalDH2 12q24.2
~R 「ALDH1L1 3q21.2 |— \-ALDH1L2 12q23.3 Common ancestral gene _ ALDH9A1 1q23 1
(〜3 billion years ago)「 ALDH5A1 6p22.2
^ALDH8A1 6q23.2
ALDH6A1 14q24.3
「ALDH3A1 17p11.2 r^ALDH3A2 17p11.2 ALI)H7A1 「ALDH3B1 11q13.2 Root s>iiibol」LaLDH3B2 11q13.2
sutennh- ^ALDH18A1 10q24.3 IiKln-idual gene LALDH16A1 19q13.33
Figure 1.1 The nineteen ALDH genes in human and their evolution relationship.
The clustering dendogram illustrates the origin of modern ALDH gene as a common ancestral gene existed approximately 3 billion years ago. The construction of this tree diagram also shows the evolutionary relationship between the 19 human ALDH genes.
This figure was adapted from Marchitti el uL, 2008. 1.2 St met 11 res ami catalytic iiuchaiiism of AIJ)II
l、he catalytic activity of Al.DIl is conserved by ihc prcscncc of invariant residues. For example, two glycinc (G246 and (]298, numbering based on the human
ALDH：! sequence), one glulamale (H399) and one phenylalanine (1:401) residues arc important for the binding of the colaclor. G246 and F401 are responsible for interacting with the nicotinamide ring of NAD+. H399 serves as a general base for the catalytic reaction and coordinates the nicotinamide ribose in the enzyme-NAD+ binary complex. G298 is important for the proper orientation of the catalytic niicleophile, cysteine (C302). This catalytic thiol is present in most of the ALDHs.
One exception is the cephalopod n-crystallin which has an arginine residue in this position and therefore it lacks catalytic activity (Zinovieva ct a/.’ 1993).
Due to the advancement in protein purification and X-ray crystallography, many ALDH proteins have been solved in atomic and three-dimensional structural details. For example, bovine ALDH2 (Steinmetz ef ciL 1997), sheep ALDHl (Moore et ciL 1998), Escherichia coli YdcW (Gruez et a I., 2004), rat 10-FTHFDH
(Tsybovsky et a I., 2007), human ALDH7A1 (Brocker et cd., 2010), etc. Most of the
ALDH enzymes exist as homotetramers and the overall folding of the monomer is conserved in the whole superfamily. In the ALDH monomer, three domains can be found, viz. NAD+-binding domain, catalytic domain and oligomerization domain
(Figure 1.2). Both the NAD+-binding and catalytic domains adopt the (3-a-P secondary structure while the oligomerization domain consists of (3-sheets. The interface of the catalytic and NAD+-binding domains is a funnel passage of catalytic pocket. The upper part of the pocket is believed to govern the substrate specificity whereas the lower part of the pocket is the catalytic site where hydride transfer from substrate to
CO factor lakes place (Liu el cd.’ 1997).
•) Figure 1.2 Structure of human ALDH7A1 monomer.
The diagram shows the structure of a human ALDH7A1 monomer with
NAD^-binding domain (in blue), catalytic domain (in green) and oligomerization domain (in red). C and N stand for carboxyl- and amino- terminals of the protein. The figure was adapted from Brocker et cd.’ 2010. \'Ill' cat~dytic tnechanisnl or ALI)II has been proposed (IS an irreversihle (1 1](.1 cutnpulsl)ry order rcaction (Figure 1.3). Th~ lirst st~p invol v ~ s th~ re v ~r s ibl ~ bi ndin g or N.-\o t to /-\LOll. Thc binding of NAO t will activat~ th~ catalytic thi ol for t h ~ attack or substratc's carbonyl carbon and thus f~lcilitat e the Connation of t h~ thiohclniacetal intennediate. Hydride is then transferred to the cofactor, leading to th e fonnation of an acyl intern1ediate. The carboxylic acid product will then be released by the hydrolysis of the acyl intennediate through a water molecule which is activated by the conserved glutalnate residue. Finally, the reduced cofactor NAOH will be released and the enzylne is now available for another round of reaction.
The binary con1plex (ALDH: NAO+) of n1any ALDH delnonstrated flexibility of nicotinan1ide on the cofactor as shown by a weak electron density of the nicotinalnide ring and ribose. The cofactor NAD+ can exist in two conformations during catalysis. One conforn1ation has the nicotinan1ide ring in an extended position
\vhich favors hydride transfer. Another conforn1ation has the reduced forn1 of the nicotinalnide ring pointing out of the catalytic pocket and thus favors the positioning of water n10lecule for hydrolysis of the acyl intern1ediate.
() C)、、 , . OH
li f , . • * 八山H ) NA1.),IT y:' I
NAL)(P)I I 、• A11)H Cvs-S ^ ALDH NAUP)' < ! \ „ /./ \\ ^^ / \ y ；i \ \ I \ I NA[.)(P 丨.
丨 z" • H ’ z.O ALDH ——Cys-S. NAD(P)|-i \ 力 z 〇 、、.——R ALDH Cys-S ！ I 、、\ 〜Gki / Thioacvlenzvme ,::’ 、 / R 丨O、 ，, / 3 intermediate \ H广 、H // ‘ \ NAD(Pr H / \\ z''. 5 \ ALDH:— Cys-S .~o N \ \ \ NAD(P)- H P zZ • 广. H N、ALDH 八LDII — Cys-S、：::iU~〇、、_[；] zZ ^ ^ 4 \ /} Oxyanion thiohemiacetal H \、、ALDH , intermediate H
Figure 1.3 Proposed catalytic mechanism of non-CoA dependent ALDH towards aldehyde substrates.
Enzyme is symbolized as ALDH-Cys-SH with SH representing the catalytic thiol
C302. The enzyme first binds to NAD(P)+ and subsequently nucleophilically attack the substrate aldehyde. After hydride transfer to NAD(P)+, deacylation occurs with the help of the nucleophilic water molecule with sequential release of carboxylate and
NAD(P)H. The figure was adapted from Marchitti et cd, 2008.
/ 1.3 Multiple timet ions ot ALDH
One of the physiological riindions ol、AlJ)ll is lo cataly/.c ihc conversion of endogenous and exogenous aldehydes lo rcspcctivc carboxylic acids with the help of the CO factor NAl)+ or NAI)P'. For example, the mitochondrial ALI3112 is involved in the second step of clhanol metabolism by catalyzing the conversion of cthanal to ethanoic acid (Klyosov et a/., 1996). ALDI I2 is also reported to be essential for the bioaclivation of nitroglycerin used lo treat angina and heart failure (Chen and Slamler,
2006). ALDHlAl (commonly known as retinal dehydrogenase) is involved in the catalytic oxidation of all-/厂肌s- and 9-c7"v-relinal to retinoic acid (Dockham e( ul.’
1992). Retinal is important for light vision by cooperating with its binding protein rhodopsin for the generation of nerve impulse upon light stimulation. On the other hand, the carboxylic acids all-//Y//7.v-retinoic acid and/or Q-c/.v-retinoic acid are important for growth, development and differentiation by serving as ligands for the retinoic and retinoic X receptor (RXR) (Duester, 2000). ALDH3A1 is involved in the oxidation of 4-hydroxynonenal derived from lipid peroxidation during oxidative stress
(Pappa et ciL, 2003). Chronic exposure to products from lipid peroxidation has been reported to be involved in atherogenesis (Witztum, 1994). tumourigenesis as well as cataractogenesis (Awasthi et ciL 1996). ALDH4A1, a mitochondrial matrix homodimer, catalyzes the second step in the proline degradation pathway - the irreversible conversion of A‘-pyrroline-5-carboxylate (P5C), derived either from proline or ornithine, to glutamate. This reaction is a necessary step connecting the urea and tricarboxylic acid cycles. ALDH9A1 catalyzes the dehydrogenation of y-aminobulyraldehyde and aminoaldehydes derived from polyamines. It has also been characterized as betaine aldehyde dehydrogenase with its ability lo convert bclainc aldehyde lo glycine betaine which is an important osmolyle used by organisms lo ovcrcomc osmotic stress (Gruncwald and Fxkstein, 199S). Apart from catal> zing the
S oMdalioii ot、nh\smlogiL、:il aldehyde substrates, some ALDII has also been reported to
catalYA? xcnobiolic compounds.「or example, AU:MI1A1 is related to ihc increased
resistance of canccr cclls towards a category o�anti-cantx T drug callcd
oxa/.aphosphorincs (some common examples include cyclophosphamide and
itbsfamide) by detoxifying the major metabolites (Sladek, 1999).
ALDH also possess non-catalytic funclions. For example, ALDHlAl has been
reported as a cytosolic thyroid hormone-binding protein in Xenopus (Vasilious et a I.,
2004). In addition, ALDHl ATs abundant expression in the mammalian cornea and
lens categorizes it as a corneal crystalline based on the similarity in the level of
expression to lens crystallins (Piatigorsky, 1998). Although the exact function of this enzyme in the cornea and lens remains a matter of controversy, there is evidence suggesting that it may play a role in the development and/or maintenance of tissue transparency through its effects on light-scattering and absorption (Jester ef "/., 1999).
It may also have a metabolic role in mitigating the effects of oxidative stress (Manzer
"1., 2003). ALDH3A1 is constitutively expressed at considerably high level
(10-40% of the water-soluble protein) in human cornea (King and Holmes, 1993). It may have been recruited by gene sharing as major corneal proteins, the crystallins which serve as structural elements in the eye (Piatigorsky, 1998). The specific roles of
ALDH3A1 are a matter of current investigations and may include maintenance of corneal transparency, protection of lens crystallins by hydroxyl radical (-OH) scavenging (Uma et al., 1996), direct absorption of UV-light (Abedinia et ciL 1990) and generation of NAD(P)H for glutathione reduction and UV-B absorption.
Further investigations have implicated ALDH3A1 as a negative cell cycle regulator as observed by elongated cell cycle, lengthened celTs doubling lime, clccrcascd colony formation efficiency and DNA synthesis as well as altering checkpoint protein expression levels in cclls with ovcr-cxprcsscd A1.I)113A1 (l)uppa
0 l '{ cd .. ~OO)). l:or e:\anlple. a high e:\pression le ve l or /\LI)II] /\ I in IICL ce ll s ( h Ul11 ill1 l"urlle~lI L'pithelial cell line) leads to decreased cyclin /\- and cyclin n-der end ent kina~c activity. decreased cyclins A. 1-3 and E as well as 1 ~2F l and cyclin-derendcnt kinasL' inhibitor p21. It is speculated that the reduction in DN/\ sy nthesi s and ce ll proliferation is ilnportant to protect corneal epithelial cells against ox idati ve stress . rvlore recent tinding delnonstrates that ALDH3A 1 expression is in ve rse ly correlated
\\'ith the activation of Peroxison1e Proliferators-Activated Receptors (PPARs). a category of orphan nuclear honnone receptors, in hUlllan cells (O raldi et a!. , 2011).
PPARy is invol ved in cell proliferation. Induction and over-expression of PPARy in
A549 (Hulllan lung adenocarcinollla epithelial cell line) and NCTC 2544 (hulllan skin keratinocyte cell line) cells by ligands and transfectioll respective ly caused a decrease in AL DH3 A 1 and inhibition of cell proliferation. On the contrary, reduction of PPARy expression using siRNA increased ALDH3A 1 expression and cell proliferation rate.
In addition, ALDH3A 1 expression increased during the period of proliferation ~ whereas PPARy expression decreased. Therefore, it is likely that through n10dulation of PPARy or ALDI-I3A 1 activity, reduced and enhanced cell proliferation can be achieved in tunlor cells and recovering tissues respectively.
Regarding the important physiological functions of various ALDH in organisms, it is anticipated that de-regulation and nlutation of ALDH genes \\f ill lead to metabolic errors and diseases. For exanlple, 11lutation of ALDH3A2 \vill cause
Sjogren-Larsson Syndrolne (SLS), an autosonlal recessive disorder characterized b) the ichthyosis, mental retardation and spasticity (Rizzo and Carnet, 2005). Mutations in ALDH4A 1 lead to type II hyperprolinenlia characterized by accunlldation of PSC and proline re sulting ill neurological nlanifestations including seizures and ln ental retardation (Geraghty et al., 1998). Mutations in both the coding and non-coding scglllents of /\LDHSA1 lead to a functional deficiency in the gene product. Thi~ r ~H(,
10 inclabolic disorder y-aniinobiilyric ((iAI认）acid clcgraclalion is characlcri/cd 1)\ the accunuilation of CiABA and y-hydroxybutyric acid leading to a variety ol neurological disorders such as mental retardation, ataxia and seizures ((iibson cf a/..
1()()7). A general increase in ALDH activities has been reported in tumor cclls, especially in lumor-inilialing cells (or canccr-slcm cclls (CSCs)) and the increased expression of ALDH is suspected lo be related to increasing invasiveness, drug-resistance and self-renewal of this category of cells (Marcarto e( a/., 201 la).
11 1.4 Aiitiquiti" - hack^roiiiul and recent discoveries
Anliquilin (AL1)I17) is classilicd as Family 7 in the AU)I1 superfamily
(N'asilioiis ct ‘//., 1()9()) with its scqucncc sharing approximately 30% identity to othcr
ALOll members. This family of proteins has dilTcrent sub-families as it can be round in different living organisms. ALDH7A refers to the proteins found in animals such as human (ALDH7A1) and C. dedans (ALDII7A2). ALDII7B refers to the proteins found in plants such as green garden pea (ALDH7B1) and apple (ALDH7B5).
ALDH7C refers to those found in insects such as Dn)s()philci melcmogaster
(ALDH7C1). The name ‘antiquitin，was coined lo this ALDH owing to its antique nature (Lee et aL, 1994). Comparison between amino acid sequences of ALDH7A1 and ALDH7B1 shows 〜60 % homology. Such a high percentage of homology between proteins in plants and animals can only be observed in actin, histone and heat-shock proteins (Doolittle, 1992) which have indispensable physiological functions. Therefore, the highly conserved nature of antiquitin in plants and animals indicates its essentiality in living organisms.
Antiquitin was first discovered as a 26g turgor protein in green garden pea
(Guerrero et aL, 1990). It is a 508 amino acid protein with 〜30 % amino acid identity to several ALDHs. The protein was found to be tiirgor-responsive as its RNA level increased in shoots by 4-6 fold during dehydration but not in heat shock and abscisic acid treatment. After rehydration, the RNA level was found to drop back to normal.
This 26g turgor protein was later renamed as ‘antiquitin，due to its highly conserved nature during evolution (Lee et aL, 1994).
Scientists later on were trying to elucidate the physiological functions of this protein. In an investigation using Brasssica napiis, antiquitin (ALDH7B3) was induccd by osmotic stress, cold shock, heat shock, abscisic acid treatment and high sail challenge (Slrochcr et aL, 1995). Similar inductions were observed in Arahidopsis
1： ilhiliiifhi (Kirch cf (//., 2005) and /fonlciini viil^circ (O/lur d "/.’ 2002). I he iikJucccI expression of anliquilin was believed to be important in couiilcracling osmotic stress.
During dclndralion or high sail challenge, Ihc intracellular osmotic potential will be higher than the surrounding and thus there will be a net movement of water out of ihc cell. The undesirable water loss will alTecl ccll turgor as well as the normal physiological tunclions. Relent ion of water under this situation is utmost important and this can be achieved by the generation of a category of small molecules named osmoproteclanls (Burg el c//., 1996). The synthesis and accumulation of osmoprotectants help to retain water without affecting normal functions of the plant cell (Chen a al., 2002). In plant, ALDH9 (betaine aldehyde dehydrogenase) is involved in generating betaine glycine from betaine aldehyde (Wyn Jones and Storey.
1981). Betaine glycine is a quaternary ammonium soluble salt well-known as an osmoprotectant. Therefore, it is speculated that the increase in ALDH7 level under stress might be important for generating aldehyde-derived osmoprotectants in a fashion similar to ALDH9.
The enzymatic activity of ALDH7 is also important for plant's detoxification by removing reactive aldehydes generated during oxidative and/or osmotic stresses.
Reactive oxygen species (ROS) are generated during these stresses and they react with plasma membrane, causing lipid peroxidation and the formation of reactive aldehydes such as nonenal and malondialdehyde. Many members of the ALDH superfamily such as ALDH3A1 and ALDH7A1 are capable of oxidizing these aldehydes (Brocker et al” 2010). Therefore, the induced expression of ALDH7 in plants may serve to quickly remove stress-induced aldehydes so as to minimize damages to cellular proteins, DNA and membrane lipids. Experimental c\'idcnces were obtained in a study using transgenic plants overcxprcssing soybean antiquilin-likc gene. The transgenic plants showed greater lolcrancc against oxidative stivss (Rodrigiics d ‘//., 2()()()). I'lirlhcr studies also suggested the parlicipalioii ol
plant aiitiquilin. together with other cn/yincs such as calalasc, in the I'jipliorhid
defense systems lor the prolcclion against biolic and abiotic stresses (Mura cl ci/..
2007).
Antiquilin in plants is also found to be imporlanl in the ripening of fruits. The
development of fruils involves a scries of processes including cell division, cell
enlargement and senescence. These processes are accompanied by a series of
physiological and biochemical changes which are closely related to gene expression
and manipulation. Some proteins have been found to be expressed specifically during
ripening of fruits (Lopez-Gomez and Gomez-Lim. 1992). In a study using apple, a 60
kDa protein which is highly expressed in fruit development was purified and
identified as antiquitin (ALDH7B5) (Yamada el ai. 1999). ALDH7B5 was believed
to be essential for fruit development as its induction happened in the enlargement
stage with sugar and water accumulation in fruit cells. It is also possible that the
induced expression is important for water retention by the increased synthesis of osmoprotectants.
Animal antiquitin was first discovered in a study analyzing a cDNA
subtraction library prepared from the rat intestinal mucosa (Lee et cd.’ 1994). A clone showing a remarkable degree of homology to the cDNA of green garden pea antiquitin was found. Later on, a complete cDNA of ALDH7A1 was also found using this original rat cDNA clone. Amino acid sequence analysis demonstrated 60 % identity and 74 % similarity between ALDH7A1 and ALDH7B1. With regard to the highly conserved nature of antiquitin in plants and animals, it is expected that the protein's function is also highly conserved in human. Scientists first find some hints by studying the causativc genes of hearing defects and deafness using a human fetal cochlca cDNA library. AU)n7Al cl)NA was isolated as one of the potential
1.1 c、uis、Ui\c genes in which mutations or dc-rcgulalioii result in hearing clclccls (Skvoiak
L'/ 1^)^)7). Subsequent Northern hlol analysis using Ictal iiiKNA samples
dcnionslralcd ihc expression of ALI)II7A1 in brain, thymus, cochlca, spleen, skeletal
musclc, lung, ovary, tongue, eye, heart, adrenal, liver and kidney. An exceptionally
high level of gene expression was found in kidney and cochlca where osmoregulation
is important. Rirther investigation of anliquilin expression in cochlea found thai the protein was only expressed in the ouler but not the inner hair cells or vestibular type 1 hair cells. The ouler hair cells are under positive hydrostatic pressure and would collapse when exposed to increased osmolarily (Chertoff and Brownell, 1994).
Therefore, it is possible that anliquilin may play a role in maintaining proper osmolarity in the outer hair cells. Analysis of ALDH7A1 amino acid sequence discovered a hydrophobic region (Lee et a/., 1994). It was first proposed to be a membrane-spanning region which helps antiquitin to regulate osmotic balance by controlling water movement in and out of the outer hair cells. However, later crystallographic study showed that this hydrophobic region is important for NAD+ cofactor binding (Tang ct cd.、2008a).
Referring to the participation of antiquitin in cochlea's osmoregulation, the protein was once suspected to be the candidate gene for Meniere's disease which is characterized by the triad of tinnitus, vertigo and sensorineural hearing loss (Andrews and Honrubia, 1996). This disease is expected to be initiated by excessive accumulation of endolymph in the inner ear. However, genetic analysis of a pedigree of nine patients from eight families showed no mutations of the antiquitin gene.
Therefore, ALDH7A1 gene may not be related with Meniere's disease (Lynch ct a/.,
2002).
Unlike plant antiquitin, animal antiquilin seems to lack inducibilil\ under various stresses, l-or example, ALDII7A1 mRNA level was nol altered by clcln dralioii. shock, chcniical hypoxia, hydrogen peroxide and forskoliii ircalincnls in lll:lv，)3 (llumaii cnibryoiiic kidney ccll line) and I lcp()2 (Human hepatoma ccll line)
(Wong, 2003). The results show that AU)117A1 is conslilulivcly expressed. Similar rcsiihs were obtained in a sliidy using black scabrcani as a model to study the control ot animal anliquilin expression under osmotic stress, oxidative stress as well as hormonal fluctuation such as thyroxine, growth hormones and Cortisol (Chan and
Fong, unpublished data).
The potential role of animal anliquilin in growth and development was exposed in a study monitoring the protein expression pattern in pig oocytes
(Ellederova et al., 2004). Proteomic analysis on pig oocytes showed a significant increase in antiquitin expression in the first meiosis and second meiosis stages compared to the initial germinal vesicle oocytes. Proteins synthesized within the period of oocyte growth allow the oocyte to have full meiotic and developmental competence (Motlik et cd.，2000). High expression of antiquitin is therefore important for the oocytes to undergo successful fertilization and zygotic development.
A direct linkage between mutations of antiquitin and human disease was first identified in the disease pyridoxine-dependent seizures (PDS) in 2006. PDS was first described in 1954 (Hunt et al., 1954). It is an autosomal recessive inherited disorder with an incidence of 1:400,000 to 1:750,000 (Been ct 2005). Symptoms include intractable seizures shortly after birth and even in the womb (Bok et al. 2007). Such symptoms are responsive only to intravenous administration of pyridoxine (Hunt et al..
1954) but conventional anticonvulsants give no relief. Systematic analysis found mutations in the ALDH7A1 gene in children with PDS (Mills et ciL 2006).
ALDH7A1 is an a-aminoadipic semialdehyde (a-AASA) dehydrogenase oxidizing a-A AS A into (x-aminoadipale (a-AAA) in Ihc presence ol NAD' in the pipccolic acid metabolism (Chang d a/., 1990). Any AIJ)117A1 gene mulalions in 1)I)S patients will
l(、 lead lo iiKictiN iilicMi ol (he cii/viiics. 1 herd ore, (x-A AS A will he accuiiuikitccl and lead
to an incivasc in its equilibrium product pipcridcinc-6-carb()xylalc (1)6(,) (Tsai and
1 Iciklcrson, 1974). 1)6〔，will complex with pyrid()xal-5-ph()sphalc (IM.Pj through
K.noc\cnagcl condensation. 1)U) is the active form of vitamin B6 and serves as a
prosthetic group in some enzymes. Thus, pathways involving degradation of
ncurotransmillors will be affected. The incomplete clcavagc of neurolransmiUors will
over-stimulate the patient's nervous system, leading to symptoms such as seizures or
other neurological defects. A similar disease hyperprolinemia type II with ALDH4A1
mutations also causes depletion of PLP through complex formation with proline
degradation intermediate (Farrant et uL 2001). Increasing number of mutation sites
on ALDH7A1 gene has been identified including splicing-mutates and translational
mutations (Plecko et cil.�2007; Salomons et uL 2007; Striano et a I., 2008; Mills et a I..
2010; Scharer et a I., 2010; Schmitt ef ciL. 2010; Millet ct ciL 2011). These mutations
are potentially valuable as biomarkers for prenatal, neonatal as well as postnatal
diagnosis.
At the protein level, the purification of black seabream {Accmthopagnis schlegeli) (Tang et cd., 2002) and grass carp {Cfenopharynogodon idella) (Chan et al,
2003) antiquitins characterized animal ALDH7 as a tetramer with subunit molecular
weight 〜58 kDa. Kinetic analysis using purified fish antiquitin also demonstrated
aldehyde-oxidizing activity towards aliphatic propionaldehyde， aromatic
benzaldehyde as well as a-AASA (Tang et cd.’ 2008a). More recent findings showed
that ALDH7A1 is catalytically active towards a wide variety of aldehydes including
stress-induced long chain aliphatic aldehydes such as nonenal, octanal and even the
osmoprotectant precursor betaine aldehyde (Brocker ct al., 2010).
Three-dimensional structures of seabream and human antiquitins ha\ c been
solved recently. They arc highly conserved with other members in ihc A1,011
17 supcrfainilN ( Tang ct ‘".，2()()(Sa). l-or the scahrcani aiUiciuilin, the native prolciii
consists of 4 inononicrs iiUcmcUng in a cliincr-of cliiiicr fashion which is common lo
other ALDI 1 members. l{ach monomer showed 3 domains: NAlV-hincling domain,
catalytic domain and oligomcrizalion domain. Two main slructural features ol" ihc
scabrcani antiquitin help explaining the catalytic properly of the enzyme, hirstly, a
large, conserved hydrophobic region is important for NAD+-binding in the binary complex. Based on sequence analysis, this region was oncc assumed to be a trans-membrane region (Lee ct ai. 1994). Secondly, the presence of two oppositely charged residues. El21 which is negatively charged and R301 which is positively charged under physiological conditions govern the specificity of antiquitin towards its substrate a-AASA. These two residues are located at the entrance of the crescent substrate-binding pocket with the catalytic thiol C302 located at the bottom of the pocket. These two residues are likely to interact with the substrate's a-ami no and a-carboxylate groups through charge-charge interactions.
Later on. the three-dimensional structure of the human ALDH7A1 was also solved (Brocker et ul.’ 2010) and the overall structure is similar to that of the seabream counterpart. This result is expected as their amino acid sequence is highly similar to each other. The two charged residues governing substrate specificity are conserved from seabream to human in terms of both amino acid sequences as well as three dimensional structures. This observation further supports that a-AASA is highly specific for antiquitins in animal.
IS 1.5 Aim of study
Our group slarlcd the study of、antiquitin by using black scabrcairi as a model and successfully purified and partially characlcri/cci the native cn/ymc (Tang cl
-002). Better understanding on both structural and llinclional characteristics ()厂 antiquitin was achieved by furlhcr studies on grass carp (Chan c( "/.， 2003), recombinant scabream as well as ALDH7A1 (Tang e( 2005; Chan e! a I., 2007). In terms ot、functions, the indueibilily of antiquitin was systematically assessed in
HEK293 and HepG2 cells. Unlike its plant counterpart, our results demonstrated the lack of inducibilily under various forms of treatments which mimic osmotic and oxidative stresses (Wong, 2005).
Recent findings on ALDH7A1 can be separated into two fronts, namely the role of o\'erexpression of ALDH7A1 in attenuation of stress-induced cytotoxicity
(Brocker ef a I., 2010: 2011) and the possible roles of the protein in disease other than
PDS, for example osteoporosis (Giio et ai. 2010). ovarian autoimmunity responses
(Edassery et aL, 2010). type 2 diabetes (Zhou ct ciL. 2010) and Huntington's disease
(Sorolla et aL, 2010). However, most of these investigations did not provide a clear picture on how the subcellular localization of ALDH7A1 correlates with the physiological significance and the diseases. The subcellular localization of a protein is important for its physiological function. In this regard, the previous demonstration of the ‘cytosolic’ localization of seabream antiquitin (Tang et aL. 2005) is not consistent with its presumed physiological function in lysine metabolism which occurs in the mitochondrial matrix (Blemings et aL, 1994; Mills et c//., 2006). Thus, further confirmation is needed. The difference in the subcellular localization of proteins can be explained by the use of different start codons (Claudiani et 2005; Rentero ct al.,
2006). In ALDl I7A1, an extra start codon exists upstream the presumed one. The region between the start codons is believed lo be important lor coding an N-tcrniinal
卜） targctm^」signal for the (mnsporl ()l、AU)I17A1 t(> niilochondria. Thcrdorc in the
present cxpcriinciual study, the subcellular localization oi'A\A)\\7A\ and the lunction
or the cxlra 28 amino acids codcd by the region between the two start coclons will bu
elucidated.
The discovery of correlation between antiquitin with fruit growth and maturation in apple (Yamada et aL 1999) and pig oocytes development (Ellederova et
‘//., 2004) provided clues for a possible involvement of the protein in growth and development. Our preliminary study in HEK293 and HepG2 cells over-expressing full length ALDH7A1 showed the involvement of the protein in cell growth (Wong. 2005).
Therefore, in functional analysis of ALDH7A1 in cell growth, the expression level of the protein during cell cycle progression as well as its changes in the subcellular localization or distribution in cell cycle will be monitored. As the protein is believed to play important roles in cell growth, its up- and down-regulation should affect the cell's doubling time. Over-expression of different ALDH7A1 variants as well as knockdown using shRNA transfection will be conducted to find the relationship between ALDH7A1 and the growth rate of transfected cells. As ALDH7A1 might play a role in cell growth, its expression level will also be studied in cancer cells which are characterized by their high growth rate and abnormal cell cycle. c liaplcr 2 iNlitoclioiuIrial and ( ytosolic i.ocali/ations ot AIJ)II7AI
2.1 Introduction
Pol) peptides synthesized on rough endoplasmic reticulum arc targeted lo different organelles inside a ccll, for example mitochondria, cytoplasm, lysosomc. nucleus, clc. Sonic proteins can be targeted lo more than one organelle and the amount of protein distributed lo the differenl organelles may vary. The phenomenon for a protein being distributed to more than one compartment in comparable amount is called "isoprotein distribution’ whereas ‘eclipsed distribution' refers to non-comparable amount of distribution (Regev-Rudzki and Pines, 2007).
Targeting of proteins to organelles is achieved by the presence of signal sequence either at the N-terminal, C-terminal or certain intermediate region of a polypeptide. These signal sequences can be recognized by specific receptors on the surface of organelles. It can also bind to specific receptors in the cytosol and then transported to the destined organelle. The reasons for proteins being targeted to more than one subcellular compartment include, first, introduce specific proteins into the organelles for specific functions or processing; and second, synergize with alternative splicing to reduce the number of genes required in a cell. The same protein may be involved in different pathways in different compartments. For example, aconitase in yeast cytosol is responsible for the formation of important metabolic intermediates from simple precursors in the shortage of main carbon sources. However, in the mitochondria, aconitase serves as an enzyme in the Kreb's cycle for the conversion of citrate to iso-citrate (Regev-Rudzki and Pines, 2005).
Multiple-targeting of a protein can be achieved by various mechanisms
(Danpure, 1995). For example, two mRNA molecules are transcribed from two genes in which only one carries a targeting sequence-coding region (e.g. alcohol dehydrogenase 2, van Loon and Young, 1986). It is also possible that two mRNA
21 inolcciilos arc transcribed IVoiii one gLMic, with only one encoding with llic lar^'cdii^'
s“iUL、ik、L、(e.g. h-tRNA synthetase. Small cl 丨Mallhcvvs ci a/.. 2005).八
non-spliccd mRNA and the spliccd mRNA can also lead lo ihc loss ol" target scqiicncc-coding region (e.g. dU'I Pasc, Ladner and Caradonna, 1997). When ihcrc is only a single mRNA transcribed IVom a gene, allcrnalivc initiation codons are utilized to translate two forms of protein with one of them containing the organcllc-targcting sequence (e.g. A'-isopenlenyl pyrophosphate: tRNA isopentenyl transferase, Gillman t" (//., 1991). Other mechanisms which allow a protein lo achieve multiple localization include incomplete targeting, reverse translation with the protein being exported from an organelle to the cytosol (e.g. fumarase. Sass ct 2003), ambiguous signal recognized by multiple organelles (Chew ct cd.’ 2003) and post-translational modifications such as phosphorylation on specific serine residues. A representative example for the control of protein targeting by post-translational modification is the increase in import of mouse cytochrome P4502E1 to the mitochondria by the phosphorylation of SI29 upon an increase in endogenous protein kinase A activity
(Robin e/a/., 2002).
ALDH7 is an evolutionarily conserved protein in both plants and animals.
Animal antiquitin is believed to participate in osmoregulation and detoxification. The physiological function of the human protein is in lysine catabolism which takes place in the mitochondria matrix (Mill et aL 2006). However, in a study using seabream antiquitin-green fluorescence protein conjugate, only cytosolic localization can be observed which contradicts the physiological function of animal antiquitin. In addition, experimental evidence pinpointing the mitochondrial localization of
ALDH7A1 in human cells as well as the mechanism leading to such localization is lacking.
The most possible mcchanism I or the mitochondrial targeting of AU)1I7A1 is till' prl':";l'lll'l' or ~l nlitocholldrial-targ~ting sequcnce (MTS). !\p'- lrt !"rolll th e preSLlI1H.: d
:..;t~lrt codt)1l of /\1 ,1)117.1\ I gene \vhich codes for the full len gth protein, th cre i\
~lllotlh..'r i ll- fralne I\TG potential start codon found in the !\ LD I 17!\ 1 gene in the Ne B I database updated on 2006 (Figure 2.1). Initiation of translation at that new upstreanl
.-\TG site \\"ill potentially lengthen the protein by 28 anlino acids in the N-tcrnlinus.
This short peptide is suspected to be a MTS as in silico analysis shows significant sequence silniliarity \vith other known MTS such as those in ALDI-12 and cytochrollle c (Table 2.1). Therefore, in the present study the subcellular localization of
ALDI-I7 A 1 and the function of this potential 28 anlino acid MTS will be clarified using WRL68 cells (Hlllnan fetal liver cell line). Four different 111ethods are employed, including subcellular fractionation, transient transfection of WRL68 cells with
ALDH7 A 1. with or \vithout the MTS conjugated with enhanced green fluorescence protein (EGFP), i111111unofluorescence staining of ALDI-I7 A 1 and finaily in1n1uno-staining followed by flo\v cyton1etric analysis of isolated 111itochondria fro111
WRL68 cells. Taking together, the relationship between the MTS and the nlitochondrial localization of ALDH7 A 1 can be evaluated.
1 .... . ' New upstream ATG V 181 cccqggctca gtAI^t^gcg- ccttcctCQC gcqctiqtqtq tqcjcactQc
• � 2 J . ^ : J一一
231 d ci cj Qcj cccig c ^^qctcLcLq gcj c c t L ggu g cd gq cc L q c c qccl (. cATG t
281 ccactctcct catcaatcaq ccccaqtatq cqtqgctgaa aqagctgqqq
Fiwe 2.1 Location of the potential mitochondrial-targeting signal (MTS) at the
N-terminus ol ALDH7A1.
The nucleotide sequence (gene accession number nin—OOl 182.4) showed the additional upstream start codon in ALDII7A1 gene. The underlined sequence codes for the putative MTS. This figure was adapted from Wong et 2010. ,i'a h h.' 2. I ( ' 0 III par is () n 0 f t h l' P11 tat iv e M r r S 0 f A L n 117 A I "V i t h t h 0 s C 0 f k no \ 't' n
III itol'hond rial proteins.
(;l'l1d~al1k Prot~il1 l1al1l ~ Positioll Amino acid s ~ qll c n cc
aL'L' l' SS lUll 110 .
NP 00 11 73.2 ALDII 7 A I 1-2g MWRI J I ) R A I J C VII AA K ~ I ' S KI J SG I) WS RPAA I :
NP OO() SI .2 ALDI1 2 1-17 M LRAAA RF G PRI ,G RRLLS
NP 00 185 2.1 Co.\ IV 1-30 MI ,ATRVFSIVG KRAIS TSVCVRAHESVV KS
NP 06 1820.1 Cyt oc hrome c 1-30 MGDVE KGKKIFI MKCSQCHTVE KGG KHKTG
------
The positivel y charged an1ino acid residues which n1ight lead to n1itochondrial targeting of the protein are sho'vvn in bold. The table is retrieved fron1 Wong et al.,
2010.
2 ~ 2.2 Materials and Methods
2.2.1 C cll cultIIre
\VRL()8 cclls were obtained froni American l ypc Culture C\)llccli()n (A'l (,(,).
cclls were cullurcd in l)ulbecco,s Modified I:aglcs\s medium (1)M1‘:M)
supplemented with 10 % fetal bovine serum (v/v) and 1 % penicillin-streptomycin
(v/\.). Cells were cultured at 37 ()C in a humidified incubator containing 5 % CXh. All
the cell culture materials used were purchased from Invilrogcn.
2.2.2 Subcellular fractionation
Approximately two million of WRL68 cells were collected at 600 x g for 3 min at 4''C and washed twice with phosphate-buffered saline (PBS) containing 100 jiM sodium orthovanaciate. Ice-cold cell lysis buffer containing 1 % sodium dodecyl sulfate, 1 % Triton X-100, 100 ^iM phenyimethylsulphonyl fluoride (PMSF), 100 jiM sodium orthovanadate (dissolved in isopropanol) in PBS with the same volume as the cell pellet and 1 cocktail protease inhibitor (Sigma) were added to resuspend the pellet. Cells were lysed on ice for 10 min with occasional vigorous mixing. Lysates were centrifuged at 20,000 x g for 7 min at 4 and supernatant was obtained as the
'total protein, of WRL68 cells.
Fractionation to separate particulate and cytosolic proteins was carried out by using the Biovision Cytosol / Particulate Rapid Separation Kit according to the manufacturer's protocol. Briefly, approximately two million of WRL68 cells were collected at 600 x g for 3 min at 4()C and washed twice with PBS containing 100 jiiM sodium orthovanadate. The cells were resuspended in an equal volume of cell suspension buffer and were kept on ice. One microlitre of cocktail protease inhibitor,
0.4 |il 100 mM PMSF, 0.4 jil 100 mM sodium orthovanadate were added to the 40 pi cytosol releasing buffer on ice. Forty microlitrcs of the particulate layer was added into a separated cppcndoii' and layered on lop, without mixing, with the 500 fil oil la\LT on icc. I hc cvtosol-rclcasiiig hullcr with protease inhibitors was added ；tiid
mi\“i with ihc suspended cclls. Wilhin 30 see, the lysatc was layered on lop of ihc oil
Li〉\、r and ccnlrifugcd in a bcnch lop ccntrifugc at 14,000 rpm for 1 min at 4 ()(,. 1 he
top supernatant was collcclcd as the ^cytosol fraclioiV and Ihc bottom layer was
collcctcd as the 'particulalc fraclionV Cell lysis bulTcr used in total protein extraction
was added to solubilise and further lyse the pellet in the ^particulate fraction, before being used for Western blot analysis.
Another separation of the mitochondria and cylosolic proteins was carried out.
Approximately ten million of WRL68 cells were collected at 600 x g for 3 min at 4 ''C and washed twice with PBS containing 100 ^iM sodium orthovanadate. Cells were resuspended to 3 X lO?/ ml in cytosolic extraction buffer (250 mM sucrose, 70 mM
KCl. 137 mM NaCL 4.3 mM Na2HP04, 1.4 mM KH2PO4. 200 i^g/ml digitonin, 100 uM PMSF, 1:100 (v/v) cocktail protease inhibitor, pH 7.2) for 3 min on ice. Lysed cells were centrifiiged at 1,000 x g for 5 min at 4 ''C. Supernatant ('cytosol fraction,) was completely removed and saved with the addition of 1 |al cocktail protease inhibitor. The pellet was solubilised in two volumes of mitochondrial lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 2 mM EGTA, 0.2 % (v/v) Triton X-100,
0.3 % (v/v) Tergitol-type NP-40 (NP-40), 100 ^M PMSF, 1 jil cocktail protease inhibitor, pH 7.4) for 5 min on ice. The suspension was centrifiiged at 10,000 x g for
10 min at 4 "C and the supernatant was saved as the ‘mitochondrial fraction,.
2.2.3 Western blot analysis
The protein concentration of the fractions was determined by the bicinchoninic acid (BCA) protein quantification method, using bovine serum albumin (BSA) as standard. Twenty of protein were resolved in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with 5 % stacking gel and 12 % running gel. After electrophoresis at 120 V’ the separated proteins were hi oiled onto
27 [H^lwiiiylidenc fluoride (PVl)l-) nicnibranc (Immohilon, Milliporc). Target proiciiis
\\ci\、dctoclcd by Ihc corresponding primary antibody al 4 T overnight (actin’ lubulin
& cytochrome C，oxidase subunit IV (CY)x IV) : 1:1 ()(){) with antibody in 5 % milk in
Tris-buffored saline (TBS, 50 mM Iris and 150 mM NaCl with pM adjusted lo 7.6 by
HCl): ALDII7A1: 1:3000 with antibody in 5 % milk in TBS-T; anli-aclin &
anti-ALDH7Al antibodies: raised from rabbit; anli-tubulin & anti-Cox IV antibodies:
raised from mouse), followed by secondary antibody conjugated with horseradish
peroxidase (HRP) in room condition for 2 hr (tubulin. Cox IV: anti-mouse-HRP
1:1000 with antibody in 5 % milk in TBS-T; actin, ALDH7A1: anti-rabbit-HRP
1:1000 with antibody in 5 % milk in TBS-T). The protein bands were visualized by
film development using chemiluminescence.
2.2.4 Flow cytometric analysis of mitochondria in WRL68 cells
Mitochondria were isolated from the WRL68 cells by the same method described in Cytosol / Mitochondria fractionation (Section 2.2.2). The mitochondria pellets were washed twice with PBS at 1,000 x g for 3 min. Washed pellets were resuspended in 100 |li1 PBS followed by a slow addition of 1 ml ice-cold absolute methanol with vortex and then kept at 4〔'C overnight. The fixed mitochondria were re-pelleted at 1,000 x g for 5 min at 4。(：. Pellets were washed with PBS and re-pelleted again at 600 x g for 3 min twice. To detect ALDH7A1 or Cox IV (positive control), mitochondrial fractions were incubated with the corresponding primary antibody (1:100, ALDH7A1 antibody: raised from rabbit; Cox IV antibodies: raised from mouse) in high potassium buffer (80 mM HEPES, 5 mM EDTA, 2 mM MgCk
0.1 % Triton X-100, pH adjusted to 7 by KOH) at 4 "C for 48 hr. Unbound primary antibody was removed by centrifugation at 600 x g for 3 min and the pellets were washed twice with the high potassium buffer. Secondary antibody (1:100, ALDI I7A1: anti-rabbit antibody—fluorescein isothiocyanate (FITC, Zymcd); Cox IV: aiili-iiioiisc
:s anlihoJ>-l'l \ C in high potassium biil'fcr) was added lo solubilisc the jicllcl. Allcr incubation a I 4 V overnight in dark, the unbound secondary antibody was removed -J h〉ccntrifugalion al 600 x g for 3 min and ihc pellets were washed and rc-pcllclcd with high potassium buffer al 600 x g lor 3 min Iwicc. The washed pellet was rcsuspendcd in 500 [il PBS be (ore How cytometric analysis in a I-ACSCanto flow cytomeler equipped with an argon laser and CellQuest software (Becton Dickinson,
San Jose, CA, USA). Thirty thousand events were acquired from each sample. The populations were identified by their lighl-scallering characteristics, enclosed in electronic gates, and analyzed for the intensity of the Ouorescenl signal.
2.2.5 Transient transfection of various EGFP constructs
The ALDH7A1 cDNA clone was purchased from Clontech. It was designed according to the information in the NCBI database in 2003. The signal peptide sequence. abbreviated as SP (5*-
ATgTggCgCCTTCCTCgCgCgCTgTgTgTgCACgCTgCAAAGACCAgCAAgCTCT
CTggACCTTggAgCAggCCTgCCgCCTTC - 3\ Accession no. NM—001182.3) and primers (Table 2.2) were purchased from Invitrogen. To generate appropriate constructs for cell transfection and protein overexpression, full length ALDH7A1 (FL) and the signal peptide variants (SP and SP-FL) were constructed by polymerase chain reaction (PGR) with restriction sites for XlwX and £CY;RI, which are used for insertion into the multiple cloning sites of the pEGFP.Nl or pEGFP.C3 plasmids (Clontech).
These result in fusion proteins with EGFP conjugated either to the carboxyl- or amino-terminal of ALDH7A1. All constructs were verified by sequencing. Tahh,' 2.2 NIH.'h'otitll\. sl'qul\nCe of the f()J4,,'ard and reverse p."irn ers used for l"on ing thl\ \' a .. ious cons t .. ucts of A L D 117 A I.
( 'oll structs Forward primt.... I{{'vt."rse pri III er "Ill plico l\
(S' -7 J') (S'-73' ) siz e
(hp)
FL TgAgCCC'I'Cg!\g;\TgTCC!\CTCTCCTC fU C ggCrC!\g!\!\TrCgCTg!\!\!\CTTg!\TT 1,533
SP TgAgCCCTCg!\gAI'g'rggCgCCTTCCTCgC ggCTC!\g!\AI'TCgg!\!\ggCggC!\ggCC'r 84
SP-FL Tg.'\ gCCC-I·Cgi\gATgTggCgCCTTCCTCgC ggCTC!\g!\!\TTCgCTg!\i\!\CTTg!\TT 1. 6 17
" () \\'l《l.(iS cclls (0.2 million) were seeded on a 22 X 22 mm glass covcrslip in a
2.5 cm Petri dish (Mal'lck). Cells were incubatccl in a humidified incubator overnight
(37、\、，5 % CX)：) lo allow ccll adhesion. Complete I)IVII:M was removed and cclls
were washed with plain DMliM (i.e. DM KM without supplement oflO % fetal bovine
serum and 1 % penicillin-streptomycin) twice and finally covered by 1 ml plain
DMEM.
Transfection of cells with plasmids was performed using Lipofectamine and
Plus reagent (Invitrogen) according to the manufocUirer，s specifications. Briefly, one
microlitre of Plus reagent and one microgram of DNA plasmids were mixed and
incubated at room condition for 15 min. Four microlitres of lipofectamine was then
added for a further incubation of 15 min. Plain DMEM (50 |il) was added to the
mixture for dilution and then introduced to the washed WRL68 cells. The cells were
incubated in a humidified incubator (37。C, 5 % CO2) for 6 hr for DNA transfection.
Cells were washed twice and covered with DMEM for overnight incubation
50/0 CO2). MitoTracker Red CMX-Ros (MTR, Molecular Probes) with a final
concentration of 500 nM in culture medium was added to cells for staining at 37 in
a humidified incubator containing 5 % CO2 for 45 min. Cells were washed with PBS
twice and 1 ml PBS was added after washing. Cells were visualized under a confocal
microscope (Leica SP5).
2.2.6 Immunofluorescence staining
WRL68 cells (0.35 million) were seeded on a 22 X 22 mm glass cover slip in
a 2.5 cm Petri dish. Three samples (labeled as 1, 2 and 3) were prepared and incubated
in a humidified incubator overnight (37 ''C, 5 % CO2). Except for Sample 3 with
double antibody staining, Samples 1 and 2 were stained with MTR before cell fixation.
One microlitre of 1 mM MTR stock was added to 50 PBS. The diluted MTR was added lo cclls in 1 ml medium lor furlher incubalion at 37 ‘'C in a humidillcd
、、1 uk、iiha“、i- containiiio S % C'O： lor 45 mm. Mccliuiii was icniovcti unci culls of all 1 samples were washed Iwicc with PBS. Two millililrcs of icc cold absolute niullianol
added to pcmicabili/c and fix the cclls for 6 hr al 4 X. Absolute methanol was rcmm cd and rcplaccd by 0.2 % bovine scrum albumin in high potassium buflcr lor 15 min at room tcnipcralurc lo block any non-specific binding. Cells were then incubalcd w ith 1:200 primary antibodies in the same bulTcr Ibr two overnight incubation al 4 ()C
(samples 2 and 3 ALDH7A1: antibody raised from rabbit; samples 1 and 3 Cox IV: antibody raised from mouse). Primary antibody was removed and cells were washed twice with high potassium buffer. Cells were stained with secondary antibody in high potassium buffer (1:200, ALDH7A1 in Samples 2 and 3: anti-rabbit-FITC; Cox IV: anti-mouse-FITC in Samples 1 and 2 or anti-mouse- phycoerythrin (PE, Invitrogen) in
Sample 3) and incubated in dark for 6 hr at 4。C. Secondary antibody was removed and cells were washed twice with PBS. Immuno-stained WRL68 cells were covered by 1 ml PBS and were observed under a confocal microscope (Leica SP5). 2J Results
2.3.1 Presence of ALDHTAl in cytosol and mitochondria in WRI.68 cells
nie Siihcclluar fractions obtained from WRL68 cclls were analysed using
Western blot analysis. Two fractionation methods were employed. Tubulin and Cox
IV were used as markers tbr the cytosol and mitochondria respectively. Preliminary
fractionation indicated the presence of ALDH7A1 in both the cytosol and particulate
fractions (Figure 2.2A). Tubulin was present predominantly in the cytosol fraction,
although a minute amount was also observed in the particulate fraction.
To confirm the presence of ALDH7A1 in mitochondria, fractionation was
subsequently carried out to obtain cytosol and mitochondrial fractions. The intense
ALDH7A1 signal in the mitochondrial fraction revealed the presence of abundant
ALDH7A1 (Figure 2.2B). Cox IV was found only in the mitochondrial fraction
u.hereas tubulin was found only in the cytosol fraction, demonstrating the successful
separation of the two fractions. To sum up, ALDH7A1 has two subcellular
localizations in WRL68 cells, cytosol and the mitochondria.
In another experiment, intact mitochondria were isolated from WRL68 cells
for flow cytometric analysis. Intact mitochondria without being immuno-stained were
used to measure the endogenous fluorescence (Figure 2.3A). Staining using anti-Cox
IV antibody gave a significant peak shift (Figure 2.3B), indicating the mitochondrial
nature of the sample preparation. Peak shift was also observed after staining of
ALDH7A1 (Figure 2.3C). Therefore, it could be concluded that ALDH7A1 is a
protein present in the mitochondria.
The subcellular localization of ALDH7A1 was also studied by immuno-fluorescence staining of WRL68 cells. In Cox IV plus MTR staining (Figure
2.4 row (A)), the overlapping of their signals (Figure 2.4A3)’ green from Cox IV
(Figure 2.4A1) and red from MTR (Figure 2.4A2), ensured proper immuno- and dye- slaiiiiiig of fixed inilochoiuli ia in WRI/)(S cclls.
I lk、staining of ALI)1I7A1 gave grccii cytosolic lluorcsccncc (I-'igiircs 2.4B1 and 2.4C1) as well as yellow mitochondrial lluorcsccncc (iMgures 2.41^3 and 2.4C3j in \\’RL68 cclls' milochondrial staining separately with M \ R (iMgurc 2.4 row (B)) and Cox IV (iMgure 2.4 row (C)). Apart from the milochondrial and cytoplasmic stainings, ALDH7A1 green fluorescence was also found in the nucleus aggregated in the nucleoplasmic region (Figures 2.4B3 and 2.4C3).
2.3.2 iMitochondrial-targeting N-terminal sequence in ALDH7A1
Different ALDH7A1-EGFP conjugates were employed to study the subcellular localization. Green fluorescence represents EGFP signal and red fluorescence represents the MTR signal. When only native EGFP was expressed, the protein was evenly distributed in the transfected cell (Figure 2.5A). Full length ALDH7A1-EGFP conjugate (FL-EGFP) was observed only in the cytosol (Figure 2.5B). However, in the presence of the N-terminal 28 amino acid putative signal peptide, both full length
ALDH7A1-EGFP conjugate (SP-FL-EGFP) (Figure 2.5C) and EGFP (SP-EGFP)
(Figure 2.5D) were targeted completely to the mitochondria, as shown by the yellow fluoresecence obtained from the merged images of EGFP and MTR. These results verified that the N-terminal signal peptide of ALDH7A1 is mitochondrial-targeting.
In a parallel set of experiments, FL (Figure 2.5F), SP-FL (Figure 2.5G), SP
(Figure 2.5H) were fused to the C-terminal of EGFP using the pEGFP.C3 vector.
From confocal analyses, the fusion proteins showed no specific subcellular localization as no yellow fluorescence signal was observed. Total |m)tcm ( ytosol Ihiclioii Parliculatc Iraction (A)
,.、、〖" I, 42 kl)a
Tubulin MiaHPMiA 55 kDa
^••MBflBMatfifr^ 57 kDa
,D、 Total protein Cytosol Fraction Mitochondrial Fraction ()
j-Actin •丨"I丨 42 kDa
Tubulin —^•^^ommm^g 55 RDa
17 kDa
ALDH7A1 -'ma
Figure 2.2 Western blot analysis of ALDH7A1 in different subcellular fractions
ofWRL68 cells.
(A) Total protein, cytosol and particulat e fractions (B) total protein, cytosol and mitochondrial fractions. Proteins in the fractions were resolved in SDS-PAGE before
Western blot analysis. Cross-contamination between fractions was checked using fraction-specific markers (Tubulin for cytosol and Cox IV for mitochondria). The figure shown is a set of representative data of three independent trials. A K\ cuts
(A) Without immuno-stciining
__‘ (B1) Cox IV immuno-stainin g (C)ALDH7A1 immuno-stainino; ‘ FITC-A
Figure 2.3 Flow cytometric analysis of isolated mitochondria from WRL68 cells.
Mitochondrial fraction was isolated from WRL68 cells to perform flow cytometric analysis (A) Without immuno-staining; (B) Cox IV immuno-staining; (C) ALDH7A1 immuno-staining. The figure shown is a set of representative data of three independent trials. IBr T Til 1 I •• I I _ •• I I: I
I ij^mc 2.5 C oil focal microscopy of WRI.68 cells transiently expressing; di He rent
A1.1)H7AI-FX;FP conjugates.
Different constructs were transiently Iransfecled into WRL68 cells using
Lipofcctaminc and Plus reagent for 48 hr. (A) pEGFP.Nl vector only, (B) VL-\'X}\'\\
(C) SP-FL-BGFP, (D) SP-EGFP, (E) pEGFP.C3 vector only, (F) EGFP-FL, (G)
EGFP-SP-FL, and (H) EGFP-SP. Mitochondria were stained with 500 nM of MTR for 45 min. The fluorescence signals were obtained by confocal microscopy. The figure shown is a set of representative data of three independent trials. 2.4 Discussion
2.4.1 In silico aiid in vitro siibcclliiiar localization studies on ALI)II7A1
Subcellular localization of a protein can be studied by both in silico and in
yiffo methods. Before the ALDII7A1 protein was purified, sequence analysis of the
ALDH7A1 gene found a large hydrophobic region between amino acid residues 155 and 208. This region was once suggested lo be a trans-membrane segment of the protein (Lee et al.. 1994). Subsequent analysis showed that this region can also be found in other ALDH members. Crystal structures of ALDH7A1 binary complex
(PDB: 2J6L), seabream ALDH7 (Tang et al., 2008a) and other ALDH members
(Steinmetz et al. 1997; Johansson et cd.’ 1998; Moore et al, 1998) demonstrated that this hydrophobic region is actually part of the NAD+-binding domain rather than serving as a trans-membrane domain. This suggestion is consistent with the aldehyde-oxidizing property of ALDH7A1 (Mills et cd., 2006) which requires NAD+ as the cofactor.
Based on the sequence alignment with other ALDHs, ALDH7A1 was once believed to be a 511 amino acid enzyme. In silico analysis of the entire amino acid sequence can be used to predict the subcellular localization of a protein. Analysis of
ALDH7A1 by the WoLF PSORT program (http://wolfpsort.org/) (Horton et al., 2007) gives no significant subcellular localization signal. In addition, previous seabream antiquitin purification experiment and N-terminal amino acid sequencing of the
'purified' protein suggested that animal antiquitin is likely to be a cytosolic protein
(Tang et al., 2002). However, the deduction of the ‘whole’ amino acid sequence from purified native protein may not be accurate. For example, the targeting signal might be cleaved by proteases after entering a specific organelle, leading to the absence of the targeting signal peptide in the purified protein and thus a false deduction on the protein's real subcellular localization. In ALDII7A1, a detail analysis of the
40 nucleotide scqucncc indicates iinolhcr upstream potential start coclon which mi|jht lead to I he addition of a 28 amino acid peptide fragment al the N-tcrminal of" ihu protein. /// si/ico analysis by WoLF 1)S0RT shows that this extra scqucncc is likely lo be a M I S. MVS is usually found al the N-terminal and is characterized by cnrichcd and alternate basic residues, lack of acidic residues, and ability to form amphiphilic a-helix.
Experimental proofs for the above prediction were obtained in our in vitro study. Direct evidence for the mitochondrial-targeting nature of the extra 28 amino acid N-terminal sequence was obtained by constructing and observing the subcellular localizations of different ALDH7A1-EGFP conjugates. The vector pEGFP.Nl was used for the fusion of protein to the N-terminal of EGFP and the cells transiently transfected with the constructed vectors were examined under a confocal microscope.
In the mock transfection control using only the pEGFP.Nl vector, green fluorescence was detected in nearly all compartments of the transfected cells except in mitochondria. It implied that the vector pEGFP.Nl could be successfully transfected into the WRL68 cells and the cells expressed functional EGFP. The red fluorescence signal represented the location of mitochondria in WRL68 cells as they were stained by MTR. The red and green fluorescence signals did not overlap impling that EGFP was not targeted to mitochondria.
To investigate whether the 28 amino acid long N-terminal sequence possesses the MTR, vector construct with the extra 28 amino acids directly conjugated to EGFP was used. From the confocal analysis, it could be observed that the green fluorescence signal and the signal from MTR superimposed completely, resulting in an intense yellow fluorescence signal. In addition, green fluorescence signal was absent in the cytosol and nucleus of transfected cells. Thus, the experimental result indicates that this 28-amino acid peptide does contain a MTS which might be crili cal for the
•n initochoiklrial locali/alion of ALDI 17A1. The milochondrial-targcling properly of this
N-lcrniinal scqucnce was rcconfirnicd when ihc upslrcam 2S amino acids was conjugated lo Al.Dl 17A1 followed by From con local analysis, ihc green tl 110rosecnce signal showed a complete overlapping with the signal from MTR. On the contrary, when the construct with ALDH7A1 conjugated lo EGFP was used, the green fluorescence signal could only be detected in the cytosol as il showed no overlapping with the fluorescence signal of MTR in transfected cells. The absence of yellow fluorescence signal indicated that this fusion protein was fully retained in the cytosol.
In the present study, the pEGFP.C3 plasmid was also included such that the presence of EGFP at the N-terminal of the protein will not mask any potential signal peptide at the C-terminal of the protein. The results failed to demonstrate any specific targeting for all constructs, including EGFP-FL, EGFP-SP-FL and EGFP-SP.
To conclude, it was observed that ALDH7A1 could be found in both the cytosol and mitochondria using EGFP conjugates. The mitochondrial localization is achieved by the presence of MTS. It remains to be elucidated which start codon is the one actually used in translation under physiological conditions.
Subcellular localization of a protein can be studied using the EGFP-conjugate approach. However, it is an artificial method and the results will be questionable when there is more than one potential start codon. Thus, the subcellular localization of
ALDH7A1 was also elucidated using Western blot analysis of the subcellular fractions of WRL68 cells. To determine whether ALDH7A1 is localized in more than one subcellular compartment, total protein, cytosol fraction and particulate fraction of
WRL68 cells were first obtained using BioVision Cytosol / Particulate Rapid
Separation Kit. Particulate fraction refers to those membrane-bound organelles such as lysosomes, peroxisomes and mitochondria. From the experiemental results, the prcscncc of a pale band representing tubulin in the particulate fraction was suspcclcd
1： lo he due lo the minute aniounl ol、tubulin allachcd to sonic of ihc organelles, loi example, mitochondria anchor lo ihc tubulin cytoskclcton for intracellular movcnicnl and thus a minute amount of tubulin unavoidably remained in the particulate fraction.
1 ho abscncc of Cox IV in the cylosol fraction indicated no cross contamination ol" c>tosol tVoni particulate fractions. Intense bands representing AI.DII7A1 were observed in total protein, cytosol fraction and particulate fraction. It can be concluded that ALDH7A1 is localized in more than one subcellular compartment, cylosol and particulates.
To verify the presence of ALDH7A1 in both mitochondria and cytosol, the mitochondrial fraction of WRL68 cells was isolated. From the Western blotting result, the absence of tubulin in the mitochondrial fraction and the absence of Cox IV in the cytosol fraction indicated that there was negligible cross-contamination between these two fractions. Similar to the previous Western blot results using cytosol/particulate fractions, ALDH7A1 was present in the cytosol of WRL68 cells. In addition, the presence of a band representing ALDH7A1 in the mitochondrial fraction implied that the enzyme is present in its anticipated subcellular compartment.
Apart from using Western blot analysis, the presence of ALDH7A1 in mitochondria was verified using flow cytometric analysis. From the experimental results, the peak in unstained mitochondria represented auto-fluorescence from mitochondria in the region of FITC-A. The sources of these fluorescence signals were probably from the mitochondrial proteins with exposed aromatic residues and heme-like groups. Enzymes involved in oxidative phosphorylation are examples of proteins using heme-like groups as the cofactors. The result served as a blank for the measurement of fluorescencefro m FITC in immuno-stained samples. Any increase in
I'l rC-A intensity (i.e. peak shifts to the right) indicates the presence of thai protein targeted by the corresponding Fn C-conjugalcd antibody. A signillcant peak shift was
•I > ohscrxcd in the sample with C\)X IV iniiiuiiio-staining. Cox IV is a niiloclioiulria
nuirkcr protein and there (ore ininnino-slaining of Cox IV served as a positive control
on the isolated mitochondria. The great peak shift indicated a large increase in
nuorosccncc intensity of 1"H C\ meaning that Cox IV is highly expressed in
mitochondria. The purity of the mitochondrial fraction was confirmed by staining the
fraction with other organelle-specific protein markers for example, PMP70 for peroxisomes, Grp94 for lysosomes and LAMP2B for endoplasmic reticulum. Our results demonstrated negligible contamination by other organelles in the mitochondrial Iraction (data not shown). Peak shift was also observed in isolated mitochondria immuno-stained for ALDH7A1. The peak shift was smaller than that in the Cox IV immuno-stained sample, indicating that the relative amount of ALDH7A1 in the mitochondria in WRL68 cells is lower than that of Cox IV.
To further explore the physiological control on the subcellular localization of
ALDH7A1， localization of the native protein was studied by the immunofluorescence method in which the cells were directly stained with ALDH7A1 antibody. Similar to the previous experimental results, ALDH7A1 was found to be localized in the cytosol and mitochondria. Surprisingly, positive staining was also observed in the nucleus as aggregates. This implied that ALDH7A1 might have some unknown functions in the nucleus. For example, nuclear ALDH7A1 may be important for regulating cell growth or serving as a reserve pool of the cytosol protein. However, the exact physiological functions of nuclear ALDH7A1 remain to be elucidated.
Using three different methods, ALDH7A1 was confirmed to be localized in both the cytosol and the mitochondria. Together with the results given by
EGFP-conjugate experiments, it is obvious that the dual localization of ALDH7AI is achieved by the absence and presence of a 28 amino acid long N-temiinal MTS. More rcccnt findings on ALDH7A1 mRNA showed the presence of two mature niRNA
44 \arKints，one willi (he scqiiciicc coding for Ihc largcliiig scqiiciicc while Ihc other docs
noi (Mrockcr d al., 2010).
Si^iiificaiKc of iiiieoclioiulrial and cytosolic 里ociili/ations of ALDMTAl
「he subcellular localization of an enzyme is lightly related to its physiological
mlc. Ihc prcscncc of ALDI17A1 in mitochondria is consistent to its proposed
physiological role in the oxidation of a-AASA, an intermediate in lysine calabolism
(Figure 2.6). The process takes place in mitochondrial matrix or peroxisomes, but not
in the cytosol (Rao ct a I., 1993). Proteomic studies have also demonstrated the
presence of ALDH7A1 in placenta (Lescuyer et al., 2003) and heart (Gaucher et al,
2004) mitochondria. Mutation of ALDH7A1 will lead to the accumulation of its
equilibrium product piperideine-6-carboxylate (P6C), resulting in the formation of
adduct with PLP (Figure 2.6). The depletion of PLP will inhibit the degradation of
neuro-transmittor, leading to pyridoxine-dependent seizure.
The cytosolic localization of ALDH7A1 has been proposed to be important in
detoxification by oxidizing reactive aldehydes generated during hyperosmotic and/or
oxidative stresses. Over-expression of ALDH7A1 has been found to increase the
resistance of cells against hyperosmotic stress and the protein has been shown to be
catalytically active towards a wide variety of stress-induced aldehydes such as nonanal, octanal and betaine aldehyde (Brocker et aL 2010). These aldehydes are usually generated from the peroxidation of membrane lipids in the presence of excessive endogenous ROS. The removal of these aldehydes is essential as they readily react with proteins and DNA, leading to inactivation as well as mutations. The cytosolic protein pool may also be important for the potential nuclear localization of
ALDH7A1. From our EGFP conjugate experiment, the SP-FL-EGFP was completely targeted lo the mitochondria. Therefore, it is not likely for ALDH7A1 to undergo retrograde translocation and release back to the cytosol and then transported into the
43 nu c k'll S. .'\ L l) Il7!\ 1 \vas prev iousl y proposed to be local ized in the: n lI C Ic:u s t() rL' glllalL' CL'II gro\vth or ccll cycle progression (Marchitti et aI. , 20(8). Pr e: li nl inary
L'~pcrin1L'ntal cvidence delll onstrating the presence of ALDH7 A 1 in th e nu cleus was obtained by our in1nluno-tluorescence staining experinlent on WRL68 cell s and also in a recent publication on the physiological functions of ALDI-I7 Al (Brocker et al. ,
2010).
Regarding the dual localization of a protein with site-specific functions, it is noted that ALDH3A2, a fatty aldehyde dehydrogenase (FALDH), can be found in both the endoplaslnic reticululn and peroxisomes (Ashibe et al. , 2007). FALDH-V is localized in the peroxisonles by the presence of an anlino-tenninal myc sequence
\vhile FALDH-N is found in the endoplasmic reticulum. These FALDH variants are produced by alternative splicing and share the ilnportant role in protecting against oxidative stress in an organelle-specific nlanner. FALDH-V, but not FALDH-N, is the key ALDH in the fatty acid degradation pathway and protects peroxisonles from oxidative stress. In contrast, both FALDHs have a protective effect against oxidative stress induced by lipid peroxidation.
Lt () 、’ o
�� \ • � ：\ H , 、） \丨 . \ A山H7A1 O , . • ,) • NH . O /.._ ‘.」H O / S -. A:p[ia-an),noad,p.c scn.aldonydo NADH A plia arnifioadipato •
^ OH .、 。
. OH
、\ ,Z , N ,
P:pendcino-6-carboxylato (P6C) Pyridoxal phosphate (PLP)
〇H I OH
, i i； ：i
o ' ' 、、.. o
！I M 卜们。，P、OH ^ HO 〇，，-OH I I Oil I OH 人N'夕丨 ，;、•:-.」 P6C/PLP adducts j
:PLP depletion
T
F'yndoxine-ciependent seizures
Figure 2.6 The role of ALDH7A1 as an a-aminoadipic semialdehyde (a-AASA) dehydrogenase.
This figure was adapted Irom M arc hi It i et "/., 2008.
•17 2.4.3 ( omparison ot animal ALDIITA and plant ALI川7B enzymes
I o have a bcUcr undcrslanding on the subcellular localization (“、antiquitin in both plants and animals, Ihc nucleotide sequences of antiquitin gene from dilTcrcnt species were compared. 1 he mRNA sequences of di fie rent antiquitins were translated by ORF finder (http:/八vwvv.ncbi.nlrn.-nih.gov/projects/gorf/)’ followed by subcellular localization analyses using the prediction program WoLF PSORT.
The results show that most animal antiquitins are predicted to have a dual localization in mitochondria and cytosol. However, their plant counterparts are mostly cytosolic or trans-membrane but not mitochondrial (Table 2.3). In general, animal antiquitins differ from plant antiquitins in having an additional potential start codon.
Translation started at the additional start codon will lead to synthesis of
MTS-ALDH7A and the protein will thus be transported into the mitochondria. For plant, a recent study using fluorescent microscopy provided direct experimental evidence that rice ALDH7 is a cytosolic protein (Shin et al., 2009). Interestingly, the prediction results show nuclear localization of antiquitin in some species {Danio rerio and Glycine max) but not in the case of ALDH7A1. Nevertheless, we and others
(Brocker et al., 2010) have shown the presence of ALDH7A1 in the nucleus.
Therefore, it is possible that the protein exists in the nucleus in both plants and animals. Regarding the difference in subcellular localization, it is anticipated that
ALDH7 from the two kingdoms might not have identical physiological function even though the protein is highly conserved through evolution.
•IS lahic 2.3 Prediction of subcellular localization of aiitiquitins from (lifTcrciit spccies.
Predicted subcellular Protein localization
Animal aiitiquitins
” . Mitochondria, Homo sapiens NP_001 173.2 — cytosol —mitochondria
,, , … Mitochondria, Macaca mulatto XP—001 1 1 1886.1 . — cytosol —mitochondria
Mitochondria, XP—001 111963.1 cytosol —mitochondria
Mitochondria, XP—001095681.1 . cytosol —mitochondria
Mitochondria, XP—001096111.1 cytosol —mitochondria
Mitochondria, XP—001096223.1 . cytosol —mitochondria
Mitochondria, Rattus norvegicus XP—214535.3 . — cytosol —mitochondria
Plasma membrane, Mus muscuius NP—613066.2 . — mitochondria
Plasma membrane, NPOO 1120810.1 peroxisome
Mitochondria, Acanthopagrus schlegelii AAX54912.1 cytosol —mitochondria
Danio rerio AAI65754.1 Cytosol, cytosol—nucleus
Mitochondria, AAH44367.1
cytosolmitochondria
Continued on next paj^e Plant aiUiquilins
Arcihiclopsis (hu/ia/ui NP_8498()7.1 Cytosol, plasma membrane
NP l 75812.1 Cytosol, plasma membrane
Orrza sativa (jcinonicci . 、 NP 001063281.1 Chloroplast, cytosol cultiyar-<^roup) —
Glycine max AAP02957.1 Nucleus, cytosol
Mollis X domestica BAA75633.1 Cytoskeleton, cytosol
Piswu satirum CAA38243.1 Cytosd，endoplasmic reticulum
The mRNA sequence of antiquitin from different species was translated by ORF finder (http:/Vww\v.ncbi.nlm.nih.gov/projects/gorf/). The subcellular localization was predicted by using the WoLF PSORT program (http://wolfpsort.org/). This table was adapted from Wong et a I., 2010.
>0 (liapkr 3 ALI)II7A1: A Potential Kej^ulator for C ell (;ro�vUi (, ell ( yclc and a
Potential Hiomarkcr for Cancer (Stem) C ells
3.1 Introduction
Apart from a role in lysine calabolism, another direction lo follow the llinclion of ALDH7A1 is its possible involvement in cell growth and cell cycle progression.
For example in apple, the expression of its antiquitin ALDH7B5 is up-regulated during fruit development and maturation (Yamada et a/., 1999). In addition, oxidative and osmotic stresses result in changes in ALDH7 expression as well as inhibition of growth and development simultaneously (Zhii, 2002). In animals, several pieces of experimental data have provided insights for the participation of ALDH7A in growth and cell cycle progression. Firstly, human cells with PIG-A mutation showed defects in ALDH7A1 expression but the cells demonstrated enhanced proliferation rate compared to normal cells (Kanai et cd.，1999). Secondly, the maturation of pig oocytes was accompanied by an elevation of ALDH7 expression (Ellederova et cd.,
2004). Thirdly, functional characterization of seabream ALDH7 gene discovered binding sites for c/、’-elements of cell cycle-dependent element (CDE) / cell cycle genes homology region (CHR) in the promoter region (Chan et cd.’ 2007). Such elements are important in regulating the expression of several cell cycle regulatory genes such as cyclin A, aurora B and cyclin B2. In ALDH7A1 gene, it has also been reported that its promoter region contains other c/^'-acting elements such as Spl, AP-2. and NF-IL6 which are believed to be important in cell growth and differentiation
(Wong, 2005).
Regarding the potential function of ALDH7A1 in cell cycle regulation, cells with abrupt cell cycle may have abnormal ALDH7A1 expression and therefore cancer cclls arc potential targets to be examined. In the last decade, the locus on ALDH studies shifts from basic kinetic charactcrization to physiological and clinical
SI sigiiilicanco of the cn/ynics. [{merging invcsligalions report the correlation bclvvccii the dc-rcgulalioii of ALDl 1 expression and canccr development. I-or example,
ALDl 11B1 was Ibimd to be highly expressed in human colon canccr cclls (Chen e! a I.,
2011). A general up-regulation of ALDH3B1 was found in many cancerous tissues such as lung，breast, ovarian and colon. Such an up-regulation was believed to be induced by the increased oxidative stress (Marchitti et ul., 2010).
In carcinogenesis, currently, there are two controversial models, the classic
‘stochastic modeF and the Miierarchical model，(Reya et cil., 2001). The classic
•stochastic model, proposes that the cell population within a tumor is heterogeneous, but that all cells are equal in acquiring mutations and initiating a tumor. The
•hierarchical model, (based on the CSC hypothesis) proposes that only a distinct and small subset of cancer cells is responsible for initiating tumors, while the remaining and majority of tumor cells are differentiated cells. The ‘hierarchical’ model is more widely accepted in cancer cell growth. A large number of、recent publications report the discovery and isolation of cancer stem-like cells in tumors from different origins.
The isolation of CSCs can be achieved by many methods and one of the methods is the Aldefluor assay which makes use of the reactivity of ALDHlAl towards the Aldefluor substrate. Stem cells and CSCs usually have ALDHlAl over-expression. Based on this characteristic, these stem cells would produce more
Aldefluor product which can be detected by flow cytometric analysis. At first, the assay was intended to isolate hematopoietic stem cells (Jones et al., 1995; Christ ct a I.,
2007). Later on, the assay was employed to isolate CSCs and the isolated cells demonstrate many stem cell characteristics. For example, isolated colorectal CSCs showed the expression of several ‘sternness’ markers such as CD133, CD44, CD24,
CD29, CD]66 and Lgr5 (Kemper ei al., 2010). In another study isolating myeloma stem cclls, the isolated cells showed much higher resistance and lower scnsili\ il> to “、mmon anti-canccr drugs. Tlicy were able to develop hind limp paralysis and can be
dctcctccl ill NOlVSni) micc even after 4 to 6 months ()r injection (Matsui ct "/.’
2008、. However, the specificity of Aldelluor substrate for various isozymes ol" human
AU)H remains lo be elucidated. More importantly, recent findings suggested that
cancer stem-like cells are associated with the abnormal expression of various ALDI I
isozymes but not solely ALDHlAl. For example, ALDH 1 A3 was reported to be
elevated in breast cancer cells and tissues, especially in breast CSCs (Marcato et al.,
2011b). Moreover, ALDH7A1 was also found to be highly expressed in
tumor-initiating and metastasis-initiating cells in human prostate cancer (van den
Hoogen et al.. 2010). Therefore, it is questionable whether those CSCs without
ALDHlAl over-expression could be successfully isolated by the Aldefluor assay.
In the present investigation, we will start to study the potential function of
ALDH7A1 in cell cycle by looking at the change in protein expression level during
cell cycle progression in HEK293 and WRL68 cells. The subcellular localization of
ALDH7A1 during cell cycle progression will also be determined to verify the relationships between subcellular localization, especially nuclear localization, and
function of ALDH7A1, especially the change in nuclear localization and cell proliferation rate. Besides, changes in several key cell cycle regulators will also be studied in cells transfected with ALDH7Al-shRNA for ALDH7A1 knockdown. On the other hand, over-expression of ALDH7A1 variants, including active and inactive cytosolic forms as well as mitochondrial form will be included to see if the protein confers growth advantage to cells in a catalytic or non-catalytic manner. The expression level of ALDH7A1 in some normal and cancer cell lines will be examined.
The expression level will also be examined in cells cultured in the ultra-low attachment system, a specially-designed system commonly employed to obtain stcm-likc cell spheroids (Miki el al., 2007; Dcy e( al.. 2009). The reactivity of
5、、 > ~ r 'Z
-.....j > >
:r e
er r. 3.2 Materials and IVltthods
3.2.1 C ell synch roil i/at ion
义’’ and WRL68 cclls, obtained from 八厂CC, were cultured in complclu
DMHM (with 10% hcal-inaclivaled horse serum, 1% non-essential amino acids and
lo/o penicillin-streptomycin, GIBCO) and complete RPMl (with 10% fetal bovine
serum and 1% penicillin-streptomycin, GIBCO) media respectively in 150 cm^
culture flask until 70-80% confluerice. For cell synchronization, different cell cycle
inhibitors were added to the medium (Table 3.1, method modified from Stephen et cd.
(1996)). Cells were kept in a humidified incubator at 37。C with 5 % CO2.
3.2.2 Semi-quantitative determination of DNA amount in synchronized cells
Synchronization efficiency was checked by flow cytometry with propidium
iodide (Sigma-Aldrich) staining, using a FACSCanto flow cytometer (Becton
Dickinson). Briefly, synchronizied cells were trypsinized to obtain 1 X 10^ cells. Cells were washed with 1 ml PBS once and resuspended in 0.1 ml PBS. One ml of 70 % ethanol was added drop wise with vortexing to completely resuspend the cells. Cells were kept at 4 ''C for 30 min. Ethanol was removed and cells were washed once with
1 ml PBS. After washing, cells were resuspended and incubated in dark at 37 ''C for
15 min in 1 ml ice-cold DNA staining solution containing 4 |_ig/ml RNase, 5 |ig/ml propidium iodide, 10 mM EDTA, 0.1 % Triton X-100 and 0.1 % sodium citrate.
Staining solution was removed and cells were resuspended in 1 ml PBS before being analysed in a flow cytometer.
3.2.3 Total protein extraction
The synchronized cells were trypsinized and lysed in PBS containing 1% SDS,
1% Triton X-100, 100 ^iM PMSF and 100 ^iM sodium orthovanadate for 10 min on ice. Lysatcs were ccntrifuged cil 20,000 x g for 7 min. The supernatant co Hoc led was taken as Ihc total protein.
s s r:山It�3. 1Kxpcriiiieiital conditions used tor synchroiii/in^ cclls to (lilTcrcnl phases of the cell cyclc.
Phase Treatment Incubation time (hr) Ci() Serum starvation 72 Cii 400 f-iM L-Mimosine 24
CJI-S 5 ^ig/ml Aphidicolin 24 S 2 |LIM Hydroxyurea 24 G.-M 5 |ag/ml Aphidicolin + 2.5 ^ig/ml II33342 24 + 48 M 6 laM Nocodazole 24
The synchronization conditions were modified from Stephen et al (1996). 3.2.4 Western blot analysis
Forty ^ig of protein samples were analyzed by SI)S-PA(JI'：. 'I he separated proteins were blotlcd onto PVDl- membrane. Target proteins were probed with the corresponding primary antibodies (IZpilomics, 1:1000) Followed by Iir^P-conjugalccl anti-rabbit or anti-mouse IgG secondary antibody (GE Healthcare), using the ECL
Western blotting reagents (GE Healthcare) for signal detection.
3.2.5 Immunofluorescence staining
Cells were seeded on 35 mm glass-bottomed confocal dishes (MatTek) until
60% confluence. Cells were synchronized as described (Section 3.2.1). Before fixation, cells were stained with 500 nM MTR for 1 hr. After fixing with paraformaldehyde (4%) for 15 min, the cells were permeabilized by incubating with
0.5% Triton X-100 for 10 min. Cells were then stained overnight with the target primary antibodies (Epitomics, 1:200) at 4 "C followed by secondary antibodies conjugated with Alexa Fluor 647 (Invitrogen, 1:200) at room temperature for 6 hr.
Cells were also stained with SYTOX Green (Invitrogen, 1:5000) for 10 min before being observed under a confocal microscope (Leica SP5).
3.2.6 Expression and purification of ALDH7A1 and its mutant
Wild type full-length ALDH7A1 cDNA was amplified by PGR performed in the GeneAmp PGR System 9700 (Applied Biosystem) using a pair of ALDH7A1 specific primer, viz. forward primer 5,-TGAGCCCATATGTCCACTCTCCTCAT-3’ and backward primer 5 ‘ -GGCTCAGAATTCTTACTGAAACTTGA-3 ^, designed according to the ALDH7A1 sequence (GenBank BT007823) using a commercially available pDNA-Dual/ALDH7Al plasmid (OpenBiosystem GH00232L1.0) as the template. The recombinant plasmid pRSETA/ALDH7A 1 was generated with the digested pRSETA (Invitrogen) and the ALDH7A1 insert containing compatible ends after enzyme digestion. For sitc-direeled mutagenesis, mutalion was introduced by
>7 using spccitic mutagenic primers designed according lo the QuikChanyc II
sitc-dircctcd mutagenesis kit (Qiagcn) with pRSl'lTA/ALDI I7A1 as the dsDNA
template. l’he VCR products were subcloncd into pRSI^TA vcctor (Invilrogcn). DNA
sequencing was performed lo confirm the sequence of the mutants. The antiquitins
were expressed in E. coli BL21(DE3)LysS and purified by affinity chromatography
on Affi-gel Blue Gel, using the procedure as described in Tang et al. (2002). The
purity of the preparations was checked by using SDS-PAGE with stacking and
running gels of 5 % and 12 % respectively.
3.2.7 Kinetic analysis of ALDH7A1 and its mutant
The enzymatic activity of ALDH7A1 was measured by following the initial
rate of NADH formation at 340 nm in an assay medium containing 2.5 mM NAD+
with acetaldehyde or a-AASA as the substrate. The buffers used for the assay of
and a-AASA were 0.1 M sodium pyrophosphate (pH 9.5) and 0.1 M
MOPS (pH 7.8) respectively. a-AASA was synthesized as described in Rumbero et al.
(1995). Briefly, 18.9 mg allysine ethylene acetal (Chiralix B.V.) was mixed with 40
mg Amberlyst-15 (dry) resin (Sigma-Aldrich) in 1 ml water for 10 min with constant
shaking. An equilibrium mixture of a-AASA with A'-piperideine-6-carboxylate (P6C)
was obtained from the filtrate. The amount of P6C formed was determined by reaction
with 2-aminobenzaldehyde (Sigma-Aldrich) (Soda et aL, 1968). The absorbance was
measured at 465 nm and an extinction coefficient of 2,800 was used for the
calculation.
3.2.8 Generation of native ALDH7A1 and mutant for transfection
The cytosol and mitochondrial forms of ALDH7A1 were generated as described (Section 2.2.5) with insertion of the constructs into the multiple cloning sites of the pBUDCE4.1 vector. The catalytic mutant ALDH7A1-C302S was generated by reaction using specific mutagenic primers designed according lo the
NS Ouikduingc 11 silc-dircclcd nuilagcncsis kil (Qiagcn) with 1/ALI)I I7A I
as the dsDNA template as dcscribcci (Section 3.2.6). AH constructs were vcrillcd by
sequencing.
3.2.9 Ceneration of stable cell line transfectants
rransfeclion of cells with plasmids was performed using Lipofectamine and
Plus reagent according to the manufacturer's specifications as described (Section
2.2.5). After transaction, cells were grown to near 100 % confluence. Medium was
removed and replaced by complete medium containing 250 jag/ml Zeocin antibiotics
(for pBUDCE4.1 transfectants) or 1000 |ig/ml G418 antibiotics (for
ALDH7Al-shRNA transfectants). Cells were selected for about 2 weeks until stable
clones were obtained. Stable transfectants were kept and passed in culture medium
containing 125 ^ig/ml Zeocin antibiotics or 500 ).ig/ml G418 antibiotics.
3.2.10 2D cell culture and ultra-low attachment cell culture
All cell lines were obtained from ATCC except 0SE7, 0SE9, OVCA432.
SK0V3 and RMGl cells lines which are gifts from Prof. Alex Shu-Wing Ng
(Brigham and Women's Hospital, Harvard Medical School). WRL68, HT29 (Human
colonic adenocarcinoma type II cell line), DU145 (Human prostate cancer cell line),
MCF7 (Human breast adenocarcinoma cell line), HepG2 (Human hepatoma cell line) cells were cultured in RPMI supplemented with 10 % fetal bovine serum (v/v) and 1 o/o penicillin-streptomycin (v/v). HEK293 cells were cultured in complete DMEM supplemented with 10 % heat-inactivated horse serum, 1 % non-essential amino acids and 1 o/o penicillin-streptomycin. SHSY5Y (Human neuroblastma cell line), MDA
(Human breast adencarcinoma cell line) and T84 (Human colonic adenocarcinoma cell line) cells were cultured in F12/DMEM (1:1, v/v) supplemented with 10 % fetal bovine serum (v/v) and 1 % penicillin-streptomycin (v/v). 0SE7 (Human ovarian surfacc epithelial cell line), OSE9 (Human ovarian surlacc epithelial ccll lino).
so 0\'C\\432 (1 liinian ovarian carcinoma ccll line), SK()V3 (Human mariaii adcnocarcinonia ccll line) and RMCi 1 (1 liiiiian ovarian dear carcinoma ccll linuj were cultured ill MCl)B/M199 (1:1, v/v) supplement with 10 % fetal bovine scrum fv/v) and 1 % pcnicillin-streplomycin (v/v). All the cells were cultured at 37 "C in a humidified incubator containing 5 % CO：. All the cell culture materials used were purchased from Invilrogen.
For the culture of stem-like spheroids, 1 ml of 50X B27 Supplement without vitamin A (Invitrogen), 1 ml of glutamax (Invitrogen), 1 ^il of 207 mg/ml heparin
(Sigma), epidermal growth factor (EGF) (5000X stock) and (3-fibroblast growth factor
(P-FGF) (5000X stock) to final concentrations of 20 ng/ml EGF and 10 ng/ml p-FGF were added to every 50 ml of plain F12/DMEM. Before seeding, cells growing in monolayer were trypsinized. The action of trypsin was stopped with complete media containing serum, and the cells were well-resuspended using 1-ml pipette tip. Cells were spun down and washed with EGF/ (3-FGF-free and serum-free medium. Cells were resuspended in a small volume of EGF/ (3-FGF containing serum-free medium using a 1-ml pipette tip. Cells were passed through a 40 jam cell strainer and observed under a light microscope to make sure that at least 90% of the cells are single cells.
Cells were counted so as to adjust the cell number to be seeded at 1x10"^ /ml in 3 ml per well in a six-well ultra-low attachment plate (Hyclone). Medium was changed once a week unless the cells showed excessive growth.
3.2.11 Collection of total cell lysates
The cells cultured in 2D culture flasks, ultra-low attachment plates and the stable transfectants were lysed and collected to obtain the 'total protein’ as described
(Section 2.2.2).
3.2.12 Western blot analysis
The protein concentration of the lysates was determined by the BC/\ protein
00 qiiaiUitlcalion method with BSA as slandard. The ccll lysalc with 40 fig ol" prolciii lor each lane were resolved in SDS-PAClli followed by Western blotting as described
(Scclion 2.2.3).
3.2.13 CiIroAvth analysis
Stable Iransfeclants for ALDH7A1 knockdown (empty vector control and
ALDH7Al-shRNA) and ALDH7A1 overexpression (empty, FL, FL-C302S and
SPl-FL) were seeded in 6-we 11 plates. Cells were trypsinized on alternate days and counted using trypan blue staining observed under a light microscope.
3.2.14 Aldefluor assay
The assay was performed according to the manufacturer's (Aldagen) protocol.
Briefly, for each sample of the cell line to be tested, one 'test' and one ‘control，12 X
75 mm tube were labeled. One milliliter of cell suspension containing 1X10^ cells/ml was added to each ‘test’ sample tube. Five fil of diethylaminobenzaldehyde
(DEAB) solution was added to the empty ‘control, tube. Five of activated
Aldefluor substrate was added per milliliter of sample to the first 'test' sample tube.
The sample was mixed and 0.5 ml o「the mixture was added immediately to the
DEAB "contror tube. Samples were incubated for 30 to 60 min at 37°C. Following
incubation, all lubes were centrifuged for 5 min at 250 x g to remove supernatant. Cell
pellets were resuspended in 0.5 ml of Aldefluor assay buffer to perform flow
cytometric analysis.
M Results
3.3.1 Fxprcssioii level of ALDl 17AI at different phases of the cell cycle
Cclls were first synchronized to di lie rent phases of the ccll cycle by incubating the cclls with diffcrenl inhibitors. The success of the synchronization was checked by How cytometric analysis which showed that the relative amount of DNA, as stained by propidium iodide, of the synchronized cells was consistent with their anticipated phases, i.e. cells in the G(), Gi and Gi-S phases are diploid; cells in the S phase are between diploid and tetraploid; whereas those in the G2-M and M phases are tetraploid (Figure 3.1 A).
Cyclin A and Cyclin D expression levels confirmed the success of cell synchronization (Chen et cd., 2002; Yang et al., 2006). Cyclin A was up-regulated during S phase while cyclin D was up-regulated during G! and G2-M phases (Figure
3.IB).
In both HEK293 and WRL68 cell lines. Western blot analysis showed that
ALDH7A1 was expressed constitutively throughout the cell cycle. However, the expression of ALDH7A1 was not the same in different phases. A more careful inspection revealed that there was an up-regulation during the G\-S phase transition
(Figures 3.1C and 3.ID). The data shown were obtained from WRL68 cells. Similar results were obtained for Western blot analysis of the HEK293 cells (data not shown).
0： Un Go Gi Gi-S S G.-M M
个 i A 个 • A A ^ A
Un Go Gi Gi-S S G2-M M (B) Cyclin D 一 ^ — - — Un Go Gi Gi-S S G^-M M
(C) WRL68 一
/3-actin
Antiquitin (D) HEK293 _
/3-actin l^i—^ipumip^m^^^
Figure 3.1 Analysis of ALDH7A1 in cells synchronized at different phases of the cell cycle. (A) Flow cytometric analysis of the relative amount of DNA in WRL68 cells after staining with propidium iodide. (B) Cyclin A and cyclin D levels in synchronized WRL68 cells. (C and D) Change in ALDH7A1 expression level during cell cycle progression.
Flow cytometric analysis showed the relative amount of DNA in synchronized cells to confirm the success of synchronization. X-axis represents relative fluorescence and y-axis represents counts. Cyclin A and D were also checked to confirm the success of synchronzation. Western blot analyses showed the level of ALDH7A1 in the whole ccll lysates of WRL68 and HEK293 cells synchronized at different phases of ihc ccll cyclc. f]-actin was used as loading control. ‘‘Un,’ refers to the unsynchronizcd cclls. 3.3.2 SII h c c 11111 a r cl is t li b ii t i o ii of AIJ)II7A1 in syiicliroiii/ed cells
In addition lo ihc changes in whole ccll AI.DI 17A1 expression level, iiiinuinotluorcsccncc staining was used to monitor Ihc subcellular distribution of
ALDI 17A1 during ccll cycle progression, or more specifically, to see whether the iip-regulalion (as shown by Western blot analysis) in Gi-S phase is specific lo any subcellular localization (Figure 3.2). It could be observed that during Go phase, the
ALDH7A1 expression level was relatively low and the protein was mainly localized in the cytosol. As the cells proceeded to Gi phase, the expression of ALDH7A1 increased in both cytosol and nucleus. During GpS phase transition, the protein accumulated in the nucleus. When the cells entered the S phase, nucleus ALDH7A1 decreased and the ALDH7A1 signal scattered in both the cytosol and the nucleus. The expression level of ALDH7A1 declined during G2-M transition with appreciable decrease in the nuclear region. Finally, in the M phase, the protein was exclusively present in the cytosolic region. Similar results were obtained for both WRL68 and
HEK293 cells.
(、I Go Gi Gi_S S G2-M M
(A) T^邏••••• Go Gi Gi_S S G2-M M
(B) I'ijiurc 3.2 I 111 111 11II011 u o res c e ii ce a ii a ly s es of ALI)H7A1 suhcelluhir distribution in
(A) \\ RL68 am! (B) MEK293 cells synchronized at different phases of the cell cycle.
ALOl 17 A1 was first stained with anli-ALDI I7A1 antibody, loll owed by secondary antibody conjugated with Alexa Fluor 647 (green signal). Nuclei and mitochondria were stained with SYTOX Green (blue signal) and MTR (red signal) respectively. The fluorescence signals were detected by confocal microscopy.
07 人3 ( hallocs ill the expression level of key ccll cycle rcj^uliitors and (he growth
rate after ALDH?A1 knockdown
lo furlher explore the role of ALDI17A1 in cell cycle progression, a preliminary study was carried out using cells with ALDII7A1 knocked down by shRNA technology. The results indicated that in WRL68 cells, the protein levels of cyclin E, cyclin D1 and B2F-1 increased while the level of cyclin A decreased after transfection with ALDH7A1-shRNA (Figure 3.3). However, the growth rate of cells with ALDH7A1 knockdown showed no significant difference compared with the empty vector control (Figure 3.4).
3.3.4 Absence of catalytic activity in the purified ALDH7A1 mutant C302S
The wild type ALDH7A1 and catalytic mutant C302S were expressed in the soluble fractions of the transformed E. coli cells as shown in SDS-PAGE (Figure 3.5).
The soluble fraction was then applied to the Affi-Gel Blue Gel for further purification.
SDS-PAGE analysis showed that the eluates from the Affi-Gel Blue Gel of wild type and mutants were quite homogenous (Figure 3.5).
a-AASA and acetaldehyde were used to study the oxidizing activity of the enzyme preparations. The specific activity of wild type ALDH7A1 towards a-AASA and acetaldehyde were 4.74 U/mg and 1.33 U/mg respectively while the catalytic mutant ALDH7A1-C302S showed no detectable activity (Table 3.2).
OS Control ALDH7A1- shRNA Antiquitin 丨丨丨丨丨丨丨丨丨丨丨__ , i -•
Cyclin E —
Cyclin D1 ’^染 丨•"丨丨丨丨丨丨丨丨丨
Cyclin A
E2F-1 I I [wiiir"-
/3-actin | 麵 ‘.、| [MI^^
Figure 3.3 Western blot analyses on cells transfected with ALDH7Al-shRNA.
The protein expression levels of various cell cycle regulatory proteins, including cyclin E, cyclin Dl, cyclin A and E2F-1 were analyzed by using the corresponding primary antibodies, followed by HRP-conjugated anti-rabbit or anti-mouse IgG secondary antibody. The signal was detected by using the ECL Western blotting reagents. (3-actin was used to ensure equal loading. The result shown is a set of representative data of three independent trials. 3MX)()00
3000000 丨Hmpty Vcctor Z
c 2S00000 : -»-ALDH7Al-shRNA
I 2000000 ; /
£ 1500000 : /J i
< 1000000 /x
500000
0 Day 0 Day 2 Day 4 Day 6 Day 8 Day after cell seeding
Figure 3.4 Growth analysis of transfected WRL68 cells by counting the cell number.
WRL68 cells were either transfected with empty vector or ALDH7A1 -shRNA.
Knockdown of ALDH7A1 was confirmed by Western blotting (Figure 3.3). Cells were trypsinized and the cell numbers were counted using trypan blue staining.
Results were the mean 士 S.E.M. (n=3).
70 BL21(DE3)pLysS lysates After Affi-eel Blue 汉 1 Marker (Soluble fiactions) size (kDn) M lA IB IC 2A 2B
94
67 ,..，〜:.；
mmmrn wm
30 I
Figure 3.5 Expression of ALDH7A1 in BL21(DE3)pLvsS cells and isolation of the over-expressed proteins.
The soluble fractions of the whole cell lysate obtained after ultra-centrifugation were analyzed by SDS-PAGE. Lane M is the molecular weight marker. Lanes lA, IB and
IC are the soluble fractions of cells transformed with wild type ALDH7A1. cells transformed with catalytic mutant C302S and untransformed cells, respectively, after overnight IPTG induction. The soluble fractions of the whole cell lysate were subjected to purification in Affi-gel Blue Gel system. The eluates were loaded to
SDS-PAGE for analysis. Lane 2A and 2B are the eluates of wild type ALDH7A1 and mutant C302S respectively.
71 I able 3.2 Activity ol recombinant Al.1)117A1 and the catalytic mutant ( 302S.
一 _ _
Specific Activity (U / mg) — ot-AASA Acclaldchydc ALDH7A1 4.74 1.33 ALD117A1-C302S Undetected Undetected
Enzyme activity with a-AASA (10 f.iM) was determined using 0.1 M MOPS, pH 7.8,
2.5 mM NAD+ while the activity with acetaldehyde (35 mM) was determined using
0.1 M sodium pyrophosphate, pH 9.5, 2.5 mM NAD+. One U is defined as the conversion of 1 |imol of NADH per min.
7： 3.3.5 Ovcr-cxprcssioii of AI.DIITAl variants in in:K293 cells.
1 lie siicccss of trans feet ion and ovcr-cxprcssion of ALDI 17AI variants was conilrnicd by Western blot analysis on the total cell lysates of stable Iranslcctants.
The protein levels of full length ALDH7A1 (FL), catalytic mutant (FL-C302S) and milochondrial-largcling ALD117A1 (SP-FL) showed up-regulation as compared to ihc control transfected with empty vector (Empty) (Figure 3.6).
3.3.6 Growth rates of cells over-expressing different ALDH7A1 variants
The effect of ALDH7A1 on cell growth was examined using cell counting.
Compared to the empty vector control, cells over-expressing either cytosolic (FL) or the mitochondrial form (SP-FL) of ALDH7A1 had growth advantage as indicated by a greater average cell number (Figure 3.7). Interestingly, cells over-expressing the catalytic mutant (FL-C302S) also showed growth advantage as compared to the control.
1:、 i:ni|)tv I L I L-C 3()2S SIM L
ALDU7A1 fllllHIIPMHHi^-xoxMMO"^
Figure 3.6 Over-expression of various ALDH7A1 constructs in HEK293 cells.
Western blot analysis of HEK293 cells transfected with plasmids containing full-length ALDH7A1 (FL), catalytic mutant with catalytic cysteine C302 mutated to serine (FL-C302S) and full length ALDH7A1 attached to the MTS (SP-FL). Cells were selected using 250 |_ig/ml Zeocin antibiotics and protein lysate was harvested from stable transfectants. Protein levels were detected using 1:1000 anti-ALDH7Al antibody and 1:10000 anti-actin antibodies with 1:1000 anti-rabbit HRP using ECL reagents.
l.\ •lOOOOOO
侧)0() ： +E_y *
3000000 FL # ^ FL-C302S / I 2500000 / < I SP-FL / /
1 2000000 : //
I 1500000 • /
< 1000000 J^
500000 ^^^^^^^^
Day 0 Day 2 Day 4 Day 6 Day 8 Day after cell seeding + _• • •• - - — - -- •冊 +_ _ ...
Figure 3.7 Growth analysis of transfected HEK293 cells by counting the cell number.
HEK293 cells were either transfected with empty vector or plasmids containing FL,
FL-C302S and SP-FL. Cells were trypsinized and the cell numbers were counted using trypan blue staining. Results were the mean 士 S.E.M. (n=3). *P < 0.05, when
FL, FL-C302S and SP-FL were compared to the empty vector control by student's
/-test. 3.3.7 l,:\i)rcssi("i ol AIJ)II7A1 in various 21) cell types and stem-like cclls
The expression level of AU)117A1 was compared in the total ccll lysalus ol
21) colls and ultra-low attachment cullurcd cclls (I^giirc 3.8). Cclls IVom clifTcrcnl ccll lines were first analysed for their ALDII7A1 expression level. IIHK293 and WRI.68 are normal cclls while the rest are canccr cells from various origins. The expression level of ALDH7A1 was found to be higher in I IEK293, WRL68, HepG2 and T84 cells (Figure 3.8A). For the other cell lines SHSY5Y, DU145, MDA, MCF7 and
HT29. the ALDH7A1 expression level was comparatively lower. Therefore, it could be implied that 2D cancer cells do not express higher level of ALDH7A1 in an obligatory manner as compared to normal cells.
The expression of ALDH7A1 in stem-like cells obtained in the ultra-low attachment culture system showed changes compared to their 2D counterparts.
SHSY5Y and DU145 showed appreciable up-regulations of ALDH7A1 expression while the others showed slight down-regulation (Figure 3.8B). Therefore, ALDH7A1 expression is up-regulated only in specific types of stem cells.
Using a specific normal-cancer cell pair, ovarian cancer cells OVCA432,
MCAS and SK0V3 showed ALDH7A1 up-regulation as compared to their normal counterparts 0SE7 and 0SE9 (Figure 3.9A). The 2D culture of ovarian cancer cell
RMGl showed lower ALDH7A1 expression as compared to those cultured in ultra-low attachment plate (Figure 3.9B).
7(、 (、） HEK293 WRL68 HepG2 SHSY5Y DU145 MDA MCF7 HT29 ISA
ALDH7A1 ^^^m^rnmrntm-mmm ——' —^ _ 丨,^ _ ,
m
-ACTIN IMMP^IIPI^^••I^^HMHII^^WH^^IMIMHMP^
(^) HEK293 WRL68 HepG2 SHSY5Y DU145 MDA MCF7 HT29 T84
niP 明 p -Actin iimill• “ 111 TT-T
Figure 3.8 Expression levels of ALDH7A1 protein in normal and cancer cell lines in 2D and ultra-low attachment culture.
Expression level of ALDH7A1 in cells cultured in (A) 2D culture system, and (B) ultra-low attachment culture system. Total cell lysates were collected from various types of normal and cancer cells. The target proteins were detected using 1:1000 primary antibody (Actin and ALDH7A1, both raised in rabbit) followed by 1:1000 anti-rabbit secondary antibody conjugated with HRP. The protein bands were visualized by film development using ECL.
77 ()SI:7 ()SI:() MC'AS ()V(’A SK()V3
Al-l)丨丨7A1 一…“
Ultra-low 2D attachment culture
ALDH7A1 一，儿 ^gggam
Figure 3.9 ALDH7A1 expression level in 21) and stem-like ovarian cancer cells.
(A) Normal cells 0SE7, 0SE9 and cancer cells MCAS, OVCA, SK0V3 2D cultures were used to check the differential ALDH7A1 expression level in normal and cancer ovarian cells. (B) RMGl cells were cultured in 2D culture system as well as in ultra-low attachment plate. The result shown is a set of representative data of three independent trials.
7S Aldetliior assay on cells ()ver-ex|)rcssinj» different AIJ)II7A1 variants
rdls o\'cr-c\prcssing dilTcrcnl 八U)117八 1 variants were tested lor ihc rcacti \ ily towards the Aldclluor substrate. The empty vcclor control provided ihc background lor endogenous catalytic activity of all ALDH isozymes towards the substrate (Figure 3.1 OA). Cells over-expressing FL (Figure 3.1 OB) and SP-FL (Figure
3.10D) showed a slightly greater peak shift as compared to the empty vector control, indicating that ALDH7A1 has some activity towards the Aldefluor substrate. For the cells over-expressing FL-C302S, the peak shift was almost the same as the empty vector control (Figure 3.IOC).
7‘） (A) o
^：, Empty
， A
10。 io' 10= T^ 10' ’。. 二 '。’ ’。. FSC-A (B) o
。 FL
» 丨 A
sj . . -I .-….. • ' V,. 10' ^ To" 10' ’。 ’。. FSC-A (C) b
FL-C302S o， .
‘ \ i R. w \ k
。。 I 一A 一 "--10I 。— 10' 10_- . 10I ' 10 ‘ ,0. 10' '"(10-、 IV 7o' FSC-A (D) o „ SP-FL o R
: _ � A
： ‘. “A
.. o • I r^ffr^'VHhH'Wiffr ”丨 W 10' '。• ’ FSC-A SO I ijiiirt�3.1 0lU)y\ cvtomctric analysis on cells over-expressing AIJ)II7AI variants
staiiuii hy Aide flu or rca^ciit.
cells were Iranstccled lo over-cxprcss ALI)I17A1 constructs. Stable iransfcctanls were stained by the Aldelluor reagent Ibr How cytometric analysis. A greater peak shift lo the right indicates higher fluorescence intensity. (A) Empty, (B)
1:L, (C) FL-C302S and (D) SP-FL. The dot plots shown on the left column represent the chosen cell population in each sample.
SI 3.4 Discussion
3.4.1 Nuclear locali/atioii of ALDHTAI
While ihc roles of ALDI 17A1 in osmoregulation / delox ill cation (in cytosol)
(Tsai c( cil.. 1974) and lysine calabolism (in inner mitochondrial matrix) (Mills ct a I.,
-006; Blocker cf c//., 2010) are well-characterized, its potential role in regulating cell cycle progression is relatively unexplored. The present study represents such an effort.
This functional role of ALDH7A1 in growth and cell cycle was first studied in two aspects: the change in the expression level and the change in the subcellular localization during cell cycle progression. In the previous chapter, transfection of imsynchronzied cells using different ALDH7A1 variants and subcellular fracationation demonstrated the presence of ALDH7A1 in the cytosolic and mitochondrial regions. However, nuclear localization was observed only in unsynchronized cells' immunofluorescence staining. Therefore, it is speculated that the nuclear presence of ALDH7A1 happens exclusively in a specific phase during cell cycle or cell division. In a cell population, cells are in different cell cycle phases and therefore the nuclear localization of ALDH7A1 may not be observed in every cell. In addition, the EGFP-conjugation of ALDH7A1 may exceed the nuclear pore size limit or lead to masking of targeting signals on the native protein and thus fail to go into the nucleus. In the study of subcellular localization of ALDH7A1 in this chapter, special focus was put onto the nuclear localization which was recently demonstrated both by using subcellular fractionation with differential centrifugation (Brocker et al.’ 2010) and by immunofluorescence staining mentioned in the previous chapter. In the present study, it was also observed that ALDH7A1 exists in the nucleus in WRL68 and
MEK293 cells. Together with the demonstration of the presence of ALDH7A1 in the nucleus of other cell lines such as vein endothelial cell, cardiomyocytes (Brockcr ci a/., 2010), the nuclear prcscncc of ALDI 17A1 could be concludcd as a general s： plwsiological phenomenon without ccll-lypc spccilicily, suggesting thai the iiiiclccir
1)117A1 is iniporlanl Ibr normal functioning and growth in living cclls.
l:mm our experiment, Ihc expression of ALDI I7A1 during Ihc ccll cyclc was conslilutivc with an obvious up-regulalion during Gi-S phase transition. The constitutive expression of ALDH7A1 is expected as lysine catabolism in protein turnover and detoxification of aldehydes occur continuously during cell division.
However, il is possible that the increase in total ALDH7A1 expression during G|-S phase transition is contributed by a single subcellular localization. Subcellular fractionation was performed to investigate the distribution of ALDH7A1 in the cytosol. mitochondria and nucleus during cell cycle progression. However, the results do not provide a clear-cut conclusion because of the incomplete separation of the organelles (data not shown). Immuno-fluorescence staining was thus performed as an alternative. It was shown that ALDH7A1 accumulated in the nucleus during G\-S phase transition and then in the S phase, its level in the nucleus decreased (Chan et cd.,
2011).
The presence of ALDH7A1 in the nucleus is believed to be regulated by the nuclear localization signal (NLS, NLStradamiis - http://www.bar.utoronto.ca/~anguyenba/) and the nuclear export signal (NES,
NetNES 1.1 - http://www.cbs.dtu.dkyservices/NetNES/) in its amino acid sequence
(Figure 3.11). The NLS and NES of ALDH7A1 are located at amino acid sequences
65-83 and 281-292, respectively. The NLS is a lysine-rich region while the NES is a leucine-rich region. These two signal regions are both a-helices with their side chains highly accessible to external environment. It is believed that the expression of ALDH increases and the proteins are then imported into the nucleus during G\-S phase by the prcscncc of the NLS. After the phase transition, the proteins arc exported out ol、the nuclcus by virtue of the NES. However, the exact signal and machinery leading to the tLlll~pprt reln~llll to be elucidated. 1'\lrthcnnorc, thc kc y signal s indu cin g th e l':xpression or /\LDI17 i\ 1 also rClnain largely unknown as various trcatnlcnts such as dehydration. heat shock. ionIzing irradiation and challengc with Iron, r-butylhydropero:xid e as well as glucocorticoids all fail to alter its mRNA level in l-lepG2 and J-IEK293 cells (Wong el af., 2005). Whether Illitogenic signals can trigger any increase in ALDH7 Al IllRNA and/or protein expression remains to be explored.
~ --l (A) Hpi^jllllH
Figure 3.11 Nuclear localization signal (NLS) and nuclear export signal (NES) of
ALDH7A1
(A) Monomer of ALDH7A1 (B) NLS with side chains shown in light-blue (C) NES
1II d 10 vv •
The structure information (PDB: 2J6L) was adapted from Protein Data Bank. The figure was prepared with PyMOL
(http://www.pymol.org). 3.4.2 Potential role of ALDIITAl in cell cycle
The potential role of ALDI 17A1 in ccll cyclc may be related lo its capability in
oxidizing \arioiis aldehydes. Oxidative stress results in the production ol' cxccssivc
ROS which subsequently leads to lipid peroxidation and finally the generation of
aldehydes. These aldehydes will react with DNA and proteins, leading lo inaclivalion of macromolecules and mutation (Comporli, 1998). ALDH7A1 can oxidize efficiently a number of lipid peroxidation-derived aldehydes, including the medium-chain saturated aldehydes as well as the unsaturated aldehyde //Y/m--2-nonenal (Brocker et
2010). The constitutive expression of ALDI-17A1 during cell cycle is important to protect DNA by oxidizing the reactive aldehydes, thus minimizing iindersiable mutations of genes in the daughter cells. Among the various phases of the cell cycle, oxidative stress is known to be most severe during the GpS phase transition
(Takahashi et 2004; Havens et cd., 2006; Mitra et al., 2009). This may also be related to the up-regulation of ALDH7A1 during this phase transition.
3.4.3 Non-catalytic role of ALDH in cell growth and development
The importance of ALDH in regulating cell growth and proliferation was first reported in ALDH3A1 (Pappa ef cd., 2005). Human ALDH3A1 was shown to inhibit proliferation and promote survival of human corneal epithelial cells against UV- and
4-hydroxynonenal-induced cellular damage mainly by metabolizing toxic lipid peroxidation aldehydes and acting as a negative cell cycle regulator. Reduction in
ALDH3A1 gene expression was observed in actively proliferating primary human corneal epithelium explant cultures, indicating that ALDH3A1 expression is inversely correlated with replication. More recent finding demonstrated an inverse correlation between ALDH3A1 and PPARy. PPAR is a category of orphan nuclear hormone rcccptors involved in cell proliferation (Oraldi ct al.. 2011). In addition, ihc nuclcar prcscncc of ALDI I3A1 also suggested its potential involvcmcnl in ccll c>'clc
So rcj^iiLumii (Slagos ci (".，2010). The involvcnicnl of dehydrogenases in regulating ccll
CYCIL、progression by I heir niiclcar prcscncc was first reported on the glycolytic nictabolic enzyme glyccraldchydc-3-phosphalc dehydrogenase and lactalc dcln drogcnasc. In ihc ccll cyclc, S-phasc transcription of the historic 2B (11213) gene is dependent on Oclamer-binding faclor 1 (Oct-1) and Oct-1 Co-Activator in S-phasc
(OCA-S), a protein complex comprising glyceraldehyde-3-phosphate dehydrogenase and lactate dehydrogenase (p38/GAPDH and p36/LDH) along with other components
(Zheng et al, 2003: Dai et uL 2008). All these give a new prospect to further study the physiological role of ALDH7A1 as a potential regulator in cell growth and cell cycle progression by its novel nuclear localization.
The potential involvement of ALDH7A1 in cell growth was demonstrated by our over-expression experiment. Cells with FL ALDH7A1 overexpression showed enhanced growth rate. A catalytic mutant FL-C302S was included to examine if a catalytically *active’ protein is essential to enhance growth rate. The catalytic thiol of
ALDH is essential for the nucleophilic attack of the carbonyl carbon on the aldehyde substrate. Without the thiol -SH group, the catalytic mutant showed undetectable enzyme activity (Table 3.2). The absence of activity of this catalytic mutant is essential for subsequent transfection into the cell lines for investigation of the non-catalytic functional role of ALDH7A1. Our growth analysis using cells over-expressing FL-C302S demonstrated an increase in growth rate similar to cells with the catalytically-active form being over-expressed. The potential function of
ALDH7A1 in cell growth is likely to bind to DNA or control the expression of growth-related gene transcription but not in the lysine catabolic pathway which occurs in the mitochondrial matrix.
ALDM7A1 was first reported lo interact with RecQL5 protein in vitro in a systematic study aiming at setting up a human protein-protein interaction network
87 (Sk、l/1 a ‘".，2005). RccQLS is a protein important for maintaining genome slability through inlciacling with RN八 polymerase II and ilsclf acting as a hclicasc (Islam cl
(//., 2010: Schwcndcncr c/ al.. 2010). To study this potential interaction partner oC
ALI)M7A1, we have performed pull-down assay using the NI IS column with purified
ALDH7A1 as bail. Unfortunately, our preliminary experiment failed to show any specific proteins which interact with ALDH7A1 (data not shown). Therefore, the exact mechanism leading to the enhanced growth rate by over-expressing ALDH7A1 needs further investigation. To conclude, cells expressing constitutive ALDH7A1 is important to their metabolic processes, namely lysine catabolism and detoxification of reactive aldehydes. However, iip-regulation of ALDH7A1 in both cancer and CSCs may confer growth advantage such as higher drug detoxification power similar to the over-expression of ALDHl A1/ALDH3A1 in some CSCs as well as increasing their growth rate by potentially promoting the transcription of growth-related genes.
To study the physiological roles of a protein, reducing endogenous expression level by transfecting with specific siRNA or shRNA will be useful. However, the effect of siRNA is temporary and is very susceptible to complete degradation.
Therefore, long-term knockdown of ALDH7A1 using cells continously expressing
ALDH7A1 -shRNA was adopted to investigate the change in cyclins’ expression. change in cell cycle profile as well as change in the growth rate. The levels of cell cycle regulatory proteins change and exhibit distinct expression patterns during the cell cycle. Cyclin D is induced by certain growth factors (Lee et aL, 2001). Cyclin D1 associates with some Cdk proteins for the entry into Gi phase (Sherr, 1994), inducing the release of transcription factor E2F-1 and expression of cyclin E (Brehm c( al.,
]998). The increased expression of cyclin E is essential for Gi-S phase transition.
Cyclin A is important during the S-phase and the entry of G2-M phase. Therefore, any ccll cycle arrest could result in, or from the changes in expression of ccll c>clc
SS rcguLuory proteins.丨：or example, the knockdown of chloridc channcI will
inhibit cyclin 1)1 and cyclin \i to arrest ccll cycle at (i|-S phase (Tang c/ 2008b).
Both lll:lv293 and WR1.68 cell lines were Iransfccled in this study. 1 lovvcvcr,
llHlv293 could not survive during antibiotic selection. It may be due lo the
importance of ALDH7A1 in maintaining HEK293 cellular osmoregulation and detoxification. Depriving HEK293 cells of its endogenous ALDH7A1 will therefore lead to cell death. In the present study, cyclin E and E2F-1 expression level increased after ALDH7A1 knockdown while cyclin Dl and cyclin A expression level decreased.
From this expression pattern, a cell population with an increase in Gi or Gi-S phase and a corresponding decrease in S and G2-M phase was expected. However, flow cytometric analysis of cells stably overexpressing ALDH7A1-shRNA did not show any significant change in cell cycle population (data not shown) and cell proliferation rate (Figure 3.4). Based on these results, no conclusion on the exact molecular mechanism can be drawn. It may be due to the endogenous compensation of function loss by other ALDHs or even other proteins since the effect of ALDH7A1 on growth may not simply involve one part of the cell cycle but a generalized delay in the progression. A longer time of observation may be essential to show the effect of
ALDH7A1 in cell growth. For example, the ectopic expression of CTCF, a multifunctional zinc-finger factor, causes growth inhibition without altering cell cycle profile (Rasko et al., 2001). However, the possibility that ALDH7A1 control the expression of growth-related genes during cell division in a direct and/or indirect manner cannot be ruled out. The participation of ALDH7A1 in cell growth may also help to explain the highly conserved nature of the protein from plant and animal during evolution.
3.4.4 Relationship between ultra-low attachment culture and stem-like cells
S^) I he use of ultra-low altachnicnl plates in culturc bccomcs more popular iii
generating rsr spheroids (Marsden cl ul., 2009; Krishnamurlhy cl d., 2010). I hu
absoncc of vitamin A and the anchorage-indcpendcnl condition prohibit ccll
dilTcrcntialion and maintain the \stemness' properties of cells. Thus, floating yd
proliferating spheres that can be serially passaged and remained undiffercnlialed will
be obtained. This method provides an effective way to obtain CSCs which are difficult to obtain in reasonable amount from clinical samples. Cells obtained from ultra-low attachment culture demonstrated many CSC properties. For example, ‘prostaspheres’ obtained from low-attachment culture of immortalized human prostate cancer cells showed a much higher CD 133 expression level (a marker of CSCs from many different origins) compared to normal cells. In addition, those CD 133+ CSC
•prostaspheres，showed a high proliferative potential, the ability to undergo induced-phenotypic and morphologic differentiation as well as a high metastatic ability (Miki et cd.’ 2007). In another study using ‘mammospheres，developed from breast cancer cells, the cells exhibited telomerase activity which cannot be detected in primary breast tissue (Dey et cd., 2009). In addition, the ‘mammospheres’ habor cells which can efflux Hoechst33342 and Rliodaminel32, a property that has been widely explored to identify and isolate stem cells from different tissues such as blood, brain, liver, muscles and others (Challen and Little, 2006; Kim et al, 2002). The rationale behind is that many stem cells and CSCs are endowed with the ability to efflux certain lipophilic drugs due to their cell surface expression of ABC family of membrane transporter proteins (Lin et al., 2006).
Ultra-low attachment plate was used in this study to obtain stem-like cells. In the cell lines used in this study, all of them showed clustering when cultured using scrum- and vitamin A- deprived medium. However, not all of I hem showed spheroid
formation. Only IIBK293, MCF7, T84 and I IepG2 cclls dcnioiislralcd sphcrical
()o morphology in a ccll duster. This may be due lo the iialurc oflhc ccll lypc used and ihc necessity of linc-tiining ihc unique culliirc conditions in clilTcrcnl ccll types lor spheroid tbrnialion (Figure 3.12).
‘)1 sm bph 厂，
|m丨丨丨丨
m^m mmtm i
^^HB^H^H^H I— 一—』
Figure 3.12 Morphology of cells cultured in ultra-low attachment plate.
Cells from 2D culture were trypsinized and introduced into ultra-low attachment plate for stem-like cell culture using specially formulated medium. After several weeks of culture, the cells morphology changed from linear shape to spheroid shape with aggregation. 3.4.5 l'p-re<:iilatioii of Al.DHs in cancer and ( S( s and the evaluation ol
applicability of Aldetluor assay in CSC isolation
In order lo isolate cancer stem cclls for biochemical rcscarch, there is a need
for a universal marker or set of、markers that allow accurate identification of, this rare
cell population. Currenlly, ALDH coupled with the immuno-slaining of
tissue-specific cluster of differentiation protein (CD) is most commonly used for such purpose. Within the human ALDH siiperfamily, ALDHlAl and ALDH3A1 are
important for normal stem cells as well as CSCs (Patel et cd., 2008; Ucar et aL, 2009).
ALDHlAl and ALDH3A1 are important for cancer stem cells in terms of detoxification, differentiation and cellular expansion. In particular, ALDHlAl has been widely accepted as a protein marker to identify and isolate normal and CSCs due to its substrate specificity. ALDHlAl is an enzyme responsible for converting retinal to retinoic acid which serves as a ligand to RXR. Ligand-bound RXR couples with
PPAR-y to control the expression of genes involved in growth and differentiation
(Ziouzenkova and Plutzky, 2008; Desvergne, 2007). The over-expression of
ALDHlAl is believed to be essential to maintain the 'sternness' property of both normal and CSCs by inhibiting differentiation. In addition, its over-expression was found to be important in conferring resistance to cancer cells against common anti-cancer drugs such as cyclophosphamide-based and oxazaphosphorine-based chemotherapeutics (Sladek et al., 2002). The toxic metabolic intermediates of these drugs can be effectively oxidized by ALDHlAl, leading to an increased resistance of cancer cells.
The success of CSCs isolation should be credited to the invention of the
Aldclluor assay kit. High ALDH activity, as detected by the Aldetluor assay, can be used as a functional marker to isolate CSCs in several types of epithelial canccrs, including those of、breast, lung, and colon, ovary and prostate (1 luang ct 200(): Umcstmcr cf al., 2007: I Icar c( al.. 2009). The principle behind is llic use ol' an
Al Dll-spccific subslmlc BODIl^Y-aminoacctaldchydc diethyl acclal (BAAA-I)A)
(l.igua、3.13). The subslmlc is activated in the prcscncc of conccntratcd hydrochloric
acid. Appropriate amount of activated BAAA is Ihcn introduced lo the ccll sample.
Since stem cells are believed to highly express ALDH, they will produce a large
amount of BODlPY-aminoacetale (BAA) from BAAA. The presence of an
ABC-transporter inhibitor in the incubation buffer will lead to the retention of BAA product in cells. Sample with both BAAA and DEAB added serves as a control as
DEAB is an inhibitor of ALDH enzyme. After incubation, cells will be subjected to tlcnv cytometric analysis to detect for their relative amount of BAA product.
Increasing number of studies reported the isolation of CSCs usin" this assav kit and demonstrated the "sternness' properties for the isolated cells (denoted as
ALDH— cells) as well as their indispensible roles in carcinogenesis and tumorigenesis.
For example, only 11 ALDH+/ CD133+ (CD 133 is a reported marker for ovarian CSC) dual positive CSCs could initate a tumor in immunocompromized mice whereas
50.000 ALDH7 CDl33+cells failed to initiate tumors (Silva et al.’ 2011). Increased angiogenesis was also observed in the ALDH+/ CD133+ cells. Similar results were obtained in a study using isolated ALDH+ prostate CSCs (van den Hoogen et aL.
2010). The ALDH+ cells showed greater tumorigenicity as well as increased metastatic growth (van den Hoogen e/ al.. 2010). Activated ALDEFLUOR ...-“• •. BAAA I . ’ I BAAA ALDH DEAB ALDH NO PRODUCT ,z baa ABC TRANSPORTERS BAA Assay Buffer Ice or cold (2-8"X) (A) H :: • .. • 、 • ...N N .」. .0:〕H. :H. / .B, ’ ：―NH- H-C F - 。：: h:,::H BODI?V:^:::i:ioacei.ilJc;.;.de dirrhvl aceral ： 3AAA-DA; H 1 H . L： 。：, NH. 0 F 厂 .. . H B(:)DIP\'-3iiuii03ceTnldehvde ！ 3AAA: ALDH H. C ‘、r飞- \\ I I 、？.：: V、日•,�气’w.、::::： H:u r。： 卜八「 \__/ (B) B0DI?Y-atnui03cer3ie iBAA'i Figure 3.13 A schematic diagram for the Aldefluor assay. (A) Cells were incubated with activated-BAAA. A higher ALDH expression level would produce a higher amount of BAA product. With appropriate excitation during How cytometric analysis, cells with higher level of BAA would be delcclcd as a OS popiilatmii with higher endogenous green lliiorcsccncc. (H) Molecular structures ()[ lk)l)ll)Y-aminoacctaklchycie diethyl acclal (BAAA-DA), the inactive lorm of Aldelluor reagent: BIDOPY-aniinoacclaldchydc (BAAA), the IICl-aclivatcd ALDIl siibslralc: BODIPY-aminoacelale (BAA), the ALDl 1-catalyzed end product. Alllimigh a high ALDIl aclivily could be observed in the CSCs, il is unclcar \\hcthcr the high Aldclluor activity is functionally involved in the \slcmncss^ (normal and canccr) or whether it is a use fill hiomarkcr to identify CSC’s. Increasing reports showed the over-expression of other members of、ALDH in canccr, and thai the observed high Aldefluor fluorescence may not be due only to ALDHlAl o\er-exprcssion. For example, a recent study lound that the 八U)II activity patient breast uiiiior CSCs and ccll lines correlates hcsl with ihc expression ofAU)! fl A3 but noi ALDHlAl (Marcarto d a I.. 201 lb). The increase in ALDH expression in cancer cells was not restricted to ALDHlAl and ALDH 1 A3 but includes other ALDHs such as ALDH2, ALDH4A1, ALDH7A1 and even ALDH18A1 (Marcarto et "/.，2011b). Thus, it is possible that Aldefluor activity is not contributed by a single ALDH but is the collective contribution of many members of the ALDH superfamily. Attempts were made in this study using an over-expresion of ALDH7A1 for investigating the reactivity of the enzyme towards the Aldefluor substrate. The activity, as observed from the flow cytometric analysis, showed that ALDH7A1 is slightly reactive towards the substrate. The Mow' activity may be due to the size limitation or absence of favorable interaction of the substrate-binding pocket to the Aldefluor substrate. In addition, the activity of ALDH7A1 towards Aldefluor substrate can be confirmed using pure enzyme preparation. However, it cannot be done due to limitation on the sensitivity of the instrument. Apart from the non-specificity of the Aldefluor substrate, the degree of inhibition by DEAB on various ALDHs may also be different. DEAB is an inhibitor towards ALDHlAl. However, its inhibitory effect on other ALDH isozymes remains questionable. Purified ALDH7A1 was tested for its inhibition by DEAB using acctaldchydc and (x-AASA as substrate. Our results showed that the activity of ALI)I17A1 towards acctaldchydc (35 mM) in the presence of suggested concentration 1)7 ol ni-AB (5 [iM) in the Aklcllum assay kit and 5 limes the suggested conccnlratioii :u\、85%) and 62% compared lo the noii-inhibitcd control. The corresponding activities towards a-AASA (K) mM) arc 85% and 52%, respectively (data not shown). Although we were not able lo obtain all the purified human ALDH isozymes, it can be speculated from our results that DEAB does not inhibit all the ALDH lo the same extern. Therefore, those isozymes without appreciable inhibition by DEAB but reactive towards the Aldefluor substrate could contribute significantly to the observed activity and give high background fluorescence. Those CSCs without high ALDHlAl expression may not be able to be isolated by using Aldefluor assay kit. Further research is warranted to identify the isozymes, besides ALDHlAl and ALDH7A1, which contribute to Aldefluor activity in highly tumorigenic and metastatic CSCs. In addition, the pattern for ALDH isozymes over-expression may vary in different cancer types or stages of cancer progression. 3.4.6 Comparison on ALDH7A1 expression level in 2D and stem-like cells Cells cultured with special formulation of medium in ultra-low attachment plates are becoming popular nowadays to study the physiology of CSCs which are difficult to obtain in significant amount using traditional culture method (Marsden et al., 2009; Krishnamurthy et al, 2010). In this study, the expression of ALDH7A1 in 2D culture and ultra-low attachment cell culture was compared. Similar to the recent finding of van den Hoogen et al. (2010), the expression of ALDH7A1 in prostate cancer stem-like cells was higher than the DU145 cells in 2D culture. However, there is no trend showing the up-regulation of ALDH7A1 in all types of cancer. The normal cell lines HEK293 and WRL68 showed higher expression than the cancer cells. It may be due to the necessity of high ALDH7A1 expression for normal funclions in kidney and liver, namely osmoregulation and detoxification respectively. Another possihnit〉 is the ()\LM-c.\prcssi()n of oIIkt AiJ)I I is()/.ymc(s) clumig (he iranstornuuion iVoiii normal to canccr cclls. AU)117/\1 is believed to play a role in CSC growth. This is supported by ihc reccm finding that ALL)H7A1 is highly expressed in Ihc tumor-initialing and metastasis-initialing cclls in human prostate cancer (van den Hoogen e( 2010). More recent finding showed that ALDH7A1 is required for the acquisition of a metastatic slem/progenilor cell phenolype in human prostate cancer and is functionally involved in prostate cancer bone metastasis. For example, Knockdown of ALDH7A1 resulted in a decrease of the a2'^'/av'^7CD44"' stem/progenitor cell subpopulation in the human prostate cancer cells. In addition, ALDH7A1 knockdown significantly inhibited the clonogenic and migratory ability of human prostate cancer cells in vitro. Moreover, a number of genes/factors involved in migration, invasion and metastasis were affected including transcription factors (snail snail2, and twist) and osteopontin, an ECM molecule involved in metastasis (van den Hoogen et al.’ 2011). However, research on ALDH7A1 and cancers only gave evidence for the functional involvement of the protein in CSCs. It is still questionable whether ALDH7A1 over-expression initiates cancer development. For example, it is possible that normal cells/stem cells initially increase ALDH7A1 expression, transforming into cancer-like cells and finally maintain a high ALDH7A1 expression level. It is also possible that ALDH7A1 is not the cause of cancer, but is an essential enzyme for the CSCs to counteract high oxidative stress and to augment their exceptionally high growth rale. Strikingly, our preliminary results demonstrated an increased expression of ALDII7A1 in Ihe ovarian cancer cells MCAS, OVCA432 and SK0V3 compared to Ihc normal cells 0SE7 and 0SE9 (Figure 3.9A). In addition, the expression level of AU)II7 八 1 in RM(j1 ovarian cancer ccll culturc resembles I lie ease in prostate canccr 1)0 LL'lb (l :igurL' ~.l)n). It should be noted that ALI)II enl.ynle acti vity is f'r equentl y rL'quircd tl)}, dclo:\itication or chClllothcrapcutic agent and stress-induced aldeh yde s. Thus 1hL' o\"cr-c:\ preSs ion serves to protect (stcnl) cell sand nlai ntai n cellular inte gri ty. In addition. ALDH enzynles l11ay also play inlportant roles in retin oic acid l11etabolis111 and sex hornlone binding (Pereira el al., 1991), which are both involved in se:\ organ developrllent, genital cancer progression and metastasis fonnation. These t\\"o pieces of data provide clues for further investigation of ALDH7 A 1 in genital oncology. as a potential lllarker for genital cancer as well as a hallmark for metastasis. l OO (liaptcr 4 I iitiirc Prospects Ihe discovery of aiiliquilin in 1990 initiated the investigations on this c\oliitionarily conserved protein. Recently, Ihc focus shifts from traditional purificalion and characteri/alion lo its physiological and clinical significance. In plants, the protein is strongly suggested to be related lo osmoregulation and/or detoxification due lo its inducibility under osmotic and oxidative stresses. However, the protein seems to provide more functions in animals. Apart from protecting animal cells in osmotic and oxidative stresses, it has been shown in this study for its potential involvement in cell growth and its significance in cancer development (van den Hoogen et a!., 2010; 2011). In terms of cell growth there is a change in the expression level and nuclear accumulation of ALDH7A1 in GpS phase transition of the cell cycle. Yet, the machinery leading to increased ALDH7A1 expression needs further evaluation. For example, further characterization of the promoter region of ALDH7A1 gene is a workable direction. The comparison between the promoter region of plant ALDH7B and animal ALDH7A may also provide clues to explain the marked differences between their inducibility under various stresses. In terms of the protein's clinical significance, it has been reported that ALDH7A1 is highly expressed in prostate CSCs (van den Hoogen et al.’ 2010; 2011). Yet, the possibility of its over-expression in other cancer cell types and CSCs from different origins cannot be ruled out. Preliminary results obtained in this study showed that an over-expression of this protein may confer growth advantage to the cancer cells. Further investigation is therefore essential to elucidate the exact mechanism leading to this advantage. For example, the over-expression of ALDH7A1 may increase the ability of cancer cells to detoxify anti-cancer drug or ROS-induced aldehydes and also increase the turnover of protein by speeding up lysine catabolisni. in addition to its catalytic role, it is likely that a non-calalylic pathway involving 101 Al 1)117A1 takes placc in the iiiiclcus. I'lirlhcr investigation on the rolu ol nuclear Al 1)117A1 nuiN also be one OIMIIL、directions lo understand this protein in the control 。1 ccll grow th such as its inlcraclions with other proteins or even its interactions with Uk、promoter regions of growth-relalcd genes. As the nuclear prcscncc of ALDI17A1 is of special interest, ihc confirmalion of NLS and NliS in Ihc protein is essential and this can be done by systematic mutagenesis lo examine mutants' nuclear-importing and -exporting abilities, and their effects of on cell growth. The preliminary data showing the over-expression of ALDH7A1 in ovarian CSCs in this study, together with the reported ALDH7A1 over-expression in prostate CSCs, provides possibility for the generalization of ALDH7A1 as a biomarker for genital carcinoma and metastasis in clinical diagnosis. Yet, further investigation should be conducted to find out the roles of the protein in cancer progression and metastasis. More clinical trials should be performed for the demonstration of up-regulation of ALDH7A1 in genital cancers. 120 Rcfcrciices AkMinia, M., Pain, T., Algar \LM. and Holmes, R.S. (1990) Bovine corncal aldehyde dclndrogcnasc: the major soluble corncal protein with a possible dual protective role tor the eye. l:xp. IZye Res. 51, 419-426. Andrews. J.C. and Hounrubia, V. (1996) Meniere^s disease. In: Disorders of the Vestibular System, Baloh, R.W., Halmagyi, G.M. (eds.), pp 300-317. Oxford University Press, New York. Ashibe’ B., Hirai, T., Higashi, K., Sekimizu, K. and Motojima, K. (2007) Dual subcellular localization in the endoplasmic reticulum and peroxisomes and a vital role in protecting against oxidative stress of fatty aldehyde dehydrogenase are achieved by alternative splicing. J. Biol. Chem. 282, 20763-20773. Awasthi, S.’ Srivatava, S.K., Piper, J.T., Singhal, S.S., Chaubey, M. and Awasthi, Y.C. (1996) Curcumin protects against 4-hydroxy-2-trans-nonenal-induced cataract formation in rat lenses. Am. J. Clin. Nutr. 64，761-766. Been, J.V., Bok, J.A., Andriessen, P. and Renier, W.O. (2005) Epidemiology of pyridoxine-dependent seizures in The Netherlands. Arch. Dis. Child. 90, 1293-1296. Bok, L.A., Struys, E., Wilemsen, M.A.A.P., Been, J.V. and Jakobs, C. (2007) Pyridoxine-dependent seizures in Dutch patients: diagnosis by elevated urinary alpha-aminoadipic semialdehyde levels. Arch. Dis. Child. 92, 687-689. l():、 Blcmmgs, k.l)., Crenshaw, T.I)., Swick, R.W. and Bcncvciiga. N.J. (”m) l.\sino-alpha-kclogliilaralc rcdiiclasc and saccharopiiic dehydrogenase arc localcd only in Ihc niilochondrial matrix in rat liver. J. Nutr. 124, 1215-1221. Brchm, A.’ Miska, E.A., McCance, D.J., Reid, J.L., Bannister, A.J. and Kouzaridcs, T. (l〔)c)8) Retinoblastoma protein recruits histone deacelylase lo repress transcription. Nature 391. 597-601. Brocker’ C., Lassen, N., Estey, T.’ Pappa, A., Cantore, M.’ Oiiova, V.V., Chavakis’ T.. Kavaiiagh, K.L., Oppermann, U. and Vasiliou V. (2010) Aldehyde dehydrogenase 7A1 (ALDH7A1) is a novel enzyme involved in cellular defense against hyperosmotic stress. J. Biol. Chem. 285, 18452-18463. Brocker, C.’ Cantore, M.’ Failli, P. and Vasiliou, V. (2011) Aldehyde dehydrogenase 7A1 (ALDH7A1) attenuates reactive aldehyde and oxidative stress induced cytotoxicity. Chem. Biol. Interact. 191, 269-277. Burg, M.B., Kwon, E.D. and Kultz, D. (1996) Osmotic regulation of gene expression. FASEB J. 10, 1598-1606. Challen, G.A. and Little, M.H. (2006) A side order of stem cells: the SP phenotype. Stem Cells 24, 3-12. 122 Uiaii, m Tang, W.K., Clicng, CMl.K. and l-ong, W.R (2007) Cloning of ihe black scabrcani [Accnif/iopch^rus sch/ci^c/i) anliquilin gene and (unctional characlcri/ation ()1 its promoter region. Mol. Cell. Biochem. 297, 151-160. Chan, C.L., Wong, J.W., Wong, C.P.’ Chan，M.K. and Fong WP. (2011) Human antiquitin: Structural and functional studies. Chem. Biol. Interact. 191, 165-170. Chan，W.M., Tang, W.K., Cheng, C.H.K. and Fong, W.R (2003) Purification, N-terminal sequence determination and enzymatic characterization of antiquitin from the liver of grass carp. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 136, 443-450. Chang’ Y.F., Ghosh, P. and Rao, V.V. (1990) L-Pipecolic acid metabolism in human liver: L-alpha-aminoadipate delta-semialdehyde oxidoreductase. Biochem. Biophys. Acta. 1038, 300-305. Chen, T.H. and Miirata, N. (2002) Enhancement of tolerance of abiotic stress by metabolic engineering of betaines and other compatible solutes. Curr. Opin. Plant Biol. 5,250-257. Chen, Z.Y., Shie, J.L. and Tseng, C.C. (2002) Gut-enriched Kruppel-like factor represses ornithine decarboxylase gene expression and functions as checkpoint regulator in colonic cancer cells. J Biol Chem. 277, 46831-46839. 105 Chen, N'., Oiiick) , 1).J., Malsiiiiiolo, A., Singh, S., Thompson, I).(\ and Vasiliou. V. (201 1) Aldch\dc dehydrogenase 11)1 (ALDI 11 B1) is a potential bioiiiarkcr (or luiman colon canccr. Biochcni. Biophys. Res. C'ommun. 405, 1 73-179. Chen. Z. and Slamler, J.S. (2006) Bioactivalion of nitroglycerin by the mitochondrial aldehyde dehvdro(二enase. Trends Cardiovasc. Med. 16, 259-265. Chertoff. M.E. and Brownell, W.E. (1994) Characterization of cochlear outer hair cell turgor. Am. J. Physiol. 266, 467-479. Chew, O.. Riidhe, C., Glaser, E. and Whelan, J. (2003) Characterization of the targeting signal of dual-targeted pea glutathione reductase. Plant Mol. Biol. 53. 341-356. Christ, O., Lucke, K., Imren, S., Leung, K., Hamilton, M., Eaves, A., Smith, C. and Eaves, C. (2007) Improved purification of hematopoietic stem cells based on their elevated aldehyde dehydrogenase activity. Haemotologica 92, 1165-1172. Comporti, M. (1998) Lipid peroxidation and biogenic aldehydes: From the identification of 4-hydroxynonenal to further achievements in biopathology. Free Radic. Res. 28, 623-635. Dai, R.P., Yu, F.X., Goh, S.R., Chng, H.W., Tan, Y.L., Fii, J丄.’ Zheng, L. and Luo.Y. (2008) Historic 2B (H2B) expression is confined lo a proper NAD+/NADH redox status. J. Biol. (:hcm. 283, 26894-26901. io(、 Danpure, C\J. (11 low can ihc products of a single gene be locali/.cd lo more than one intracellular comparlmcnl? Trends Cell Biol. 5, 230-238. Dcs\ergnc, B. (2007) RXR: from partnership to leadership in metabolic regulations. VitaiiL Horm. 75, 1-32. Dey, D.. Saxena, M.’ Paranjape, A.N., Krishnan, V., Giraddi, R. Kumar, M.V., Mukherjee, G. and Rangarajan, A. (2009) Phenotypic and functional characterization of human mammary stem/progenitor cells in long term culture. PLoS ONE 4，e5329. Dockham. P.A., Lee, M.O. and Sladek, N.E. (1992) Identification of human liver aldehyde dehydrogenases that catalyze the oxidation of aldophosphamide and retinaldehyde. Biochem. Pharmacol. 43, 2453-2469. Doolittle, R.F. (1992) Reconstructing history with amino acid sequences. Protein Sci. 1, 191-200. Duester, G. (2000) Families of retinoid dehydrogenases regulating vitamin A function, production of visual pigment and retinoic acid. Eur. J. Biochem. 267, 4315-4324. Ellederova, Z., Halada, P., Man, P., Kubelka, M., Motlik, J. and Kovarova, H. (2004) Protein patterns of pig oocytes during in vitro maturation. Biol. Reprod. 71, 1533-1539. 107 l\lasscr\, S.L., Shalavi, S.V., Kunkcl, J.P., 1 kuicr, C\, iiruckcr, (,., Pciuimalsa, K.. Yu. 丫.，J.A. and 1 Aiborsky, j.L. (2010)八utoantigciis in ovarian auloimmiiiiily associated with unexplained infertility and premature ovarian failure. I'crtil. Slcril. 94. 2636-2641. Fan-am, R.D., Walker, V., Mills, G.A., Mel lor, J.M. and Langley, G.J. (2001) Pyridoxal phosphate de-activation by pyiToline-5-carboxylic acid. Increased risk of vitamin B6 deficiency and seizures in hyperprolinemia type II. J. Biol. Chem. 276, 15107-15116. Gaucher, S.P., Taylor, S.W., Fahy, E., Zhang, B., Warnock, D.E., Ghosh, S.S. and Gibson. B.W. (2004) Expanded coverage of the human heart mitochondrial proteome using multidimensional liquid chromatography coupled with tandem mass spectrometry. J. Proteome Res. 3, 495-505. Geraghty, M.T., Vaughn, D., Nicholson, A.J.’ Lin’ W.W., Jimenez-Sanchez, G.’ Obie. C., Flynn, M.R, Valle, D. and Hu, C.A. (1998) Mutations in the Deltal-pyrroline-5-carboxylate dehydrogenase gene cause type II hyperprolinemia. Hum. Mol. Genet. 7, 1411-1415. Gibson, K.M., Christensen, E., Jakobs, C., Fowler, B., Clarke, M.A., Hammersen, G., Raab, K., Kobori, J., Moosa, A., Vollmer, B., Rossier, E., lafolla, A.K., Matern. D.，Brouwer, O.F., Finkelstein, J.，Aksu, R, Weber, H.R, Bakkeren, J.A., Gabreels. F.J., Bluestone, D., Barron, T.F., Beauvais, P., Rabier, D.，Santos, C. and Lehnert, W. (1997) The clinical phenotype of succinic semialdehyde dehydrogenase deficiency (4-hydroxybutyric aciduria): case reports of 23 new patients. Pediatrics 99, 567-574. lOS Cullman, SI usher, L.B., Martin, N.C\ and I loppcr, A.K. (1991) M()l)5 translation inilialion sites determine N6-isopcnlcnyladcn()sinc modil'ication ol mitochondrial and cytoplasmic IRNA. Mol. Ccll. Biol. 11, 2382-2390. Gineslier, C., Hiir, M.H., Charal^-Jauffret, E., Monville, F., Dutcher, J., Brown, M., Jacquemier, J., Viens, P., Kleer, C.G., Liu, S., Schott, A., Hayes, D., Birnbaum, D.. Wicha, M.S. and Dontu, G. (2007) ALDHl is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 1. 555-567. Gruez, A.，Roig-Zamboni, V.’ Grisel, S., Salomoni, A., Valencia, C.’ Campanacci, V., Tegoni, M. and Cambillau, C. (2004) Crystal structure and kinetics identify Escherichia coli YdcW gene product as a medium-chain aldehyde dehydrogenase. J. Mol. Biol. 343,29-41. Grunewald, R.W. and Eckstein, A. (1995) Osmotic regulation of the betaine metabolism in immortalized renal cells. Kidney Int. 48, 1714-1720. Guerrero, F.D., Jones, J.T. and Mullet, J.E. (1990) Turgor-responsive gene transcription and RNA levels increase rapidly when pea shoots are wilted. Sequence and expression of three inducible genes. Plant Mol. Biol. 15, 11-26. 127 Cilio, Y., I'an, L.J., l.ci, S.l’.，Yang, 1 .L., Chcii, X.I)., Zhang, 1’.’ Chen, Y., Pan. I.., Yaii. 11., Liu, I"ian, Q., /hang, Z.X., Zhou, Q., Qiu, C., Dong, S.S., Xu, X.I I., (mo. Y.l:.,/hu, X.Z., Liu, S.I.., Wang, X.L., Li, X., I.uo, Y., Zhang, L.S., IJ, M., Wang- J.T., Won, r.. Drees, B.’ Hamilton, J., Papasian, C.J., Recker, R.R., Song, X.P., Cheng, J. and Deng, H.W. (2010) Genomc-wide association study identifies ALDII7A1 as a novel susceptibility gene for osteoporosis. PLoS Genet. 6, el000806. Havens. C.G., Ho, A., Yoshioka, N. and Dowdy, S.F. (2006) Regulation of late Gl/S phase transition and APC Cdhl by reactive oxygen species. Mol. Cell. Biol. 26, 4701-4711. Horton. P.. Park, K.J., Obayashi, T., Fiijita, N., Harada, H., Adams-Collier, C.J. and Nakai. K. (2007) WoLF PSORT: Protein localization predictor. Nucleic Acids Res. 35, 585-587. Huang, E.H., Hynes, M.J., Zhang, T.’ Ginestier, C., Dontu, G., Appelman, H., Fields. J.Z., Wicha, M.S. and Boman, B.M. (2009) Aldehyde dehydrogenase 1 is a marker for normal and malignant human colonic stem cells (SC) and tracks SC overpopulation during colon tumorigenesis. Cancer Res. 69, 3382-3389. Hunt, A.D., Strokes, J., McGorry, W.W. and Stroud, H.H. (1954) Pyridoxine dependency. Report of a case of intractable convulsions in an infant controlled with pyridoxine. Pediatrics 13, 140-145. 110 Islam, M.N., 1\)\ 1)., Ciuo, R., Mnoiiiolo, T. and Wang, W. (2010) R。c(儿5 promotes genome slabili/alion through two parallel mcchanisms-inlcracling with RNA polymerase 11 and acting as a hclicasc. Mol. Ccll Biol. 30, 2460-2472. Jester, J.v., Moller-Pedersen, T.’ Huang, J., Sax, C.M., Kays, W.T., Cavangh, H.D., PelrolK W.M. and Piatigorsky, J (1999) The cellular basis of corneal transparency: evidence for 'corneal crystallins'. J. Cell Sci. 112, 613-622. Johansson, K.. El-Ahmad, M., Ramasvvamy, S., Hjelmqvist, L., Jornvall, H. and Ekhind, H. (1998) Structure of betaine aldehyde dehydrogenase at 2.1 人 resolution. Protein Sci. 7,2106-2117. Jones, R.J.. Barber, J.P., Vala, M.S., Collector, M.I., Kaufmann, S.H., Ludeman, S.M., Colvin, O.M. and Hilton, J. (1995) Assessment of aldehyde dehydrogenase in viable cells. Blood 85, 2742-2746. Kanai, N., Vreeke, T.M. and Parker, C.J. (1999) Paroxysmal nocturnal hemoglobinuria: analysis of the effects of mutant PIG-A on gene expression. Am. J. Hematol. 61, 221-231. Kemper, K., Grandela, C. and Medema, J.P. (2010) Molecular identification and targeting of colorectal cancer stem cells. Oncotarget 6, 387-395. Klyosov, A.A., Rashkovelsky, L.G., Tahir, M.K. and Keiing, W.M. (1996) Possible role of liver cytosolic and mitochondrial aldehdye dehydrogenases in acclaldchydc metabolism. Biochemistry 35, 4445-4456. 111 King, Cl. and 1 lolnics, R.S.(卜)()3) I Iiinian conical aldehyde dehydrogenase: purificalion, kinclic characlcrisalion and phcnolypic variation. I^iochcm. Mol. Biol. Int. 31,49-63. Kim, M., Turnquist, H., Jackson, J., Sgagias, M., Yan, Y., Gong, M.’ Dean, M.’ Sharp, J.G. and Cowan, K. (2002) The multidrug resistance transporter ABCG2 (breast cancer resistance protein 1) effluxes Hoechst 33342 and is overexpressed in hematopoietic stem cells. Clin. Cancer Res. 8, 22-28. Kirch, H.H., Schlingensiepen, S., Kotchoni, S., Siinkar, R. and Bartels, D. (2005) Detailed expression analysis of selected genes of the aldehyde dehydrogenase (ALDH) gene superfamily in Arabidopsis thalkma. Plant Mol. Biol. 57, 315-332. Krishnamiirthy, S., Dong, Z.’ Vodopyanov, D.’ Imai, A., Helman, J.I., Prince, M.E., Wicha, M.S. and Nor, J.E. (2010) Endothelial cell-initiated signaling promotes the survival and self-renewal of cancer stem cells. Cancer Res. 70, 9969-9978. Ladner, R.D. and Caradonna, S.J. (1997) The human dUTPase gene encodes both nuclear and mitochondrial isoforms. Differential expression of the isoforms and characterization of a cDNA encoding the mitochondrial species. J. Biol. Chem. 272, 19072-19080. Lee, J.S., Liu, J.J., Hong, J.W. and Wilson, S.E. (2001) Differential expression analysis by gene array of cell cycle modulators in human corneal epithelial cells stimulated with epidermal growth factor (EGF), hepalocyte growth factor (IIGF). or kcratinocytc growth 1 actor (KGF)- Curr. Eye Res. 23, 69-76. 130 I 00. 1), IviihL W., Ciclbarl, T., Kamimura, T., West, C. and licullcr, 1". (1()(>>4) 1 loniolog) bclwccn a human protein and a prole in of the green garden pea. (icnomics 21, 371-378. Lescuyer, P., Strub, J.M., Liiche, S., Diemer, H., Martinez, P., van Dorsselaer, A.’ Lunardi, j. and Rabilloud, T. (2003) Progress in the definition of a reference human mitochondrial proteome. Proteomics 3, 157-167. Lin, T., Islam, O. and Heese, K. (2006) ABC transporters, neural stem cells and neurogenesis — a different perspective. Cell Res. 16, 857-871. Liu. Z.J.’ Sun, Y.J., Rose’ J.，Chung, Y.J., Hsiao, C.D., Chang, W.R., Kuo, I., Perozich. J., Lindahl, R., Hempel, J. and Wang, B.C. (1997) The first structure of an aldehyde dehydrogenase reveals novel interactions between NAD and the Rossman fold. Nat. Struct. Biol. 4, 317-326. Lopez-Gomez, R. and Gomez-Lim, M.A. (1992) Changes in mRNA and protein synthesis during ripening in mango fruit. J. Plant Physiol. 141, 82-87. Lynch, M., Cameron, T.L., Knight, M., Kwok, T.Y., Thomas, P., Forrest, S.M.’ Giersch, A.B., Briggs, R.J. and Pyman, B.C. (2002) Structural and mutational analysis of antiquitin as a candidate gene for Meniere's disease. Am. J. Med. Genet. 110. 397-399. 113 Man/or, R., Qaniar, L., l{slc>', T., Pappa, A., Petersen, D.R. and Vasiliou, V. (2003) Molccular cloning and baculovirus expression of Ihc rabbit corneal aldehyde dchNcirogcnasc (ALl)l 11A1) cDNA. DNA Cell Biol. 22, 329-338. Marcalo, P., Dean, C.A., Giacomantonio, C.A. and Lee, P.W. (201 la) Aldehyde dehydrogenase: Its role as a cancer stem cell marker comes down to the specific isoform. Cell Cycle 10, 1378-1384. Marcato, P., Dean, C.A., Pan, D.’ Araslanova, R., Gillis, M., Joshi, M., Helyer, L., Pan, L., LeidaL A.. Giijar, S., Giacomantonio, C.A. and Lee, P.W. (2011b) Aldehyde dehydrogenase activity of breast cancer stem cells is primarily due to isoform ALDHl A3 and its expression is predictive of metastasis. Stem Cells 29, 32-45. Marchitti, S.A., Orlicky, D.J. and Vasiliou, V. (2007) Expression and initial characterization of human ALDH3B1. Biochem. Biophys. Res. Commun. 356, 792-798. Marchitti, S.A., Brocker, C., Stagos, D. and Vasiliou, V. (2008) Non-P450 aldehyde oxidizing enzymes, the aldehyde dehydrogenase superfamily. Expert Opin. Drug Metab. Toxicol. 4, 697-720. Marchitti, S.A, Brocker, C., Orlicky, D.J. and Vasiliou, V. (2010) Molecular characterization, expression analysis, and role of ALDH3B1 in the cellular protection against oxidative stress. Free Radic. Biol. Med. 49, 1432-1443. Ill Mars丄、11, C\Ci., Wriohi, M.j., Pochaiiipally, R. and Rowan, (2009) Bruasl tuinor-iniliating cclls isolated from patient corc biopsies for study of hormone action. Methods Mol. Biol. 590, 363-375. Matsui，W.. Wang, Q., Barber, J.P., Brennan, S., Smith, B.D., Borrello, I., McNiece, 1., Lin, L., Ambinder, R.F., Peacock, C., Watkins, D.N., Huff^ C.A. and Jones, R.J. (2008) Clonogenic multiple myeloma progenitors, stem cell properties, and drug resistance. Cancer Res. 68, 190-197. Matthews, G.D., Gould, R.M. and Vardimon, L. (2005) A single glutamine synthetase gene produces tissue-specific subcellular localization by alternative splicing. FEBS Lett. 579, 5527-5534. Miki, J., Furusato, B., Li, H., Gu, Y., Takahashi, H., Egawa, S., Sesterhenn, I.A.’ McLeod, D.G., Srivastava, S. and Rliim, J.S. (2007) Identification of putative stem cell markers, CD 133 and CXCR4, in hTERT-immortalized primary nonmalignant and malignant tumor-derived human prostate epithelial cell lines and in prostate cancer specimens. Cancer Res. 67, 3153-3161. Millet A., Salomons, G.S., Cneude, F., Come, C., Debillon, T., Jakobs, C.’ Stmys, E. and Hamelin, S. (2011) Novel mutations in pyridoxine-dependent epilepsy. Eur. J. Paediatr. Neurol. 15, 74-77. 115 Milk Struys, 1':., Jakobs, C., IMecko, K, Baxter, P., Baunigarlncr, M., Willuniscii. Nl.A.A.I)，Oniran, II., I'ackc, II., Uhlcnbcrg, B., Wcschkc, B. and Clayton, P.T. (2006) Mutations in anliquitin in individuals with pyridoxinc-clependent seizures. Nat. Med. 12, 307-309. Mills, P.B., Foolitt, E.J.’ Mills, K.A., Tiischl, K., Aylett, S., Varadkar, S., Hemingway, C., Marlow. N., Rennie, J.’ Baxter, P., Dulac, O.’ Nabbout, R.,Craigen, W.J., Schmitt, B., Feillet, R, Christensen, E., De Lonlay, P., Pike, M.G., Hughes, M.I., Struys, E.A., Jakobs, C.，Zuberi, S.M. and Clayton, P.T. (2011) Genotypic and phenotypic spectrum of pyridoxine-dependent epilepsy (ALDH7A1 deficiency). Brain 133, 2148-2159. Mitra, K., Wunder, C., Roysam, B., Lin, G. and Lippincott-Schwartz, J. (2009) A hyperfused mitochondrial state achieved at Gi-S regulates cyclin E buildup and entry into S phase, Proc. Natl. Acad. Sci. USA 106, 11960-11965. Moore, S.A., Baker, H.M., Blythe, T.J., Kitson, K.E., Kitson, T.M. and Baker, E.N. (1998) Sheep liver cytosolic aldehyde dehydrogenase: the structure reveals the basis for the retinal specificity of class 1 aldehyde dehydrogenases. Structure 6, 1541-1551. Motlik, J., Pavlok, A., Lapathitis, G. and Kiibelka, M. (2000) Impact of two-step in vitro culture systems on developmental potency of oocytes. Reprod. Domest. Anim. 35,267-271. 11(> Mum, A., PiiUiis, 1:., Mcdda, R., Moris, ()., Rinakli,八•(’. and l)adiglia, A. (2007) CmalasL、and aiiliqiiitin from luiplwrhia characias: two proteins involved in plant defense'.) Biochcmislry Mosc. 72, 501-508. Neben, D.W. and Wain, H.M. (2003) Update on human genome completion and annotations, gene nomenclature. Hum. Genomics 1, 66-71. Oraldi, M.. Saracino, S., Maggiora, M., Chiaravalloti, A., Buemi, C., Martinasso, G, Paiuzzi, E., Thompson, D.，Vasiliou，V. and Canuto, R.A. (2011) Importance of inverse correlation between ALDH3A1 and PPARy in tumor cells and tissue regeneration. Chem. Biol. Interact. 191, 171-176. Oztur, Z.N., Talame, V., Deyholos, M.，Michalowski, C.B., Galbraith, D.W., Gozukirmizi, N., Tuberosa, R. and Bohnert, H.J. (2002) Monitoring large-scale changes in transcript abundance in drought- and salt-stressed barley. Plant Mol. Biol. 48, 551-573. Pappa, A., Estery, T.’ Manzer, R., Brown, D. and Vasiliou, V. (2003) Human aldehyde dehydrogenase 3A1 (ALDH3A1), biochemical characterization and immunohistochemical localization in the cornea. Biochem. J. 376, 615-623. Pappa, A., Brown, D., Koutalos, Y.’ DeGregori, J., White, C. and Vasiliou, V. (2005) Human aldehyde dehydrogenase 3A1 inhibits proliferation and promotes survival of human corneal epithelial cells. J. Biol. Chem. 280, 27998-28006. 117 Paid, M., Lil L., Zander, D.S., Srccrama, 1 …Coco, I), and Morcb, J.S. (2()()X) ALDlllAl and AL1)113A1 expression in lung canccrs: correlation with histologic l> pc and potential precursors. Lung Cancer 59, 340-349. Pereim, F.’ Rosenmann, E., Nylen, E., Kaufman, M., Pinsky, L. and Wrogemann, K. (1991) The 56 kDa androgen binding protein is an aldehyde dehydrogenase. Biochem. Biophys. Res. Commun. 175, 831-838. Plecko, B., Paul, K., Paschke, E., Stoeckler-Ipsiroglu, S.’ Struys, E., Jakobs, C.. Hartmann, H.. Luecke, T., di Capua, M., Korenke, C., Hikel, C., Reutershahn, E.， Freilinger, M., Baumeister, R, Bosch, F. and Erwa, W. (2007) Biochemical and molecular characterization of 18 patients with pyridoxine-dependent epilepsy and mutations of the antiquitin (ALDH7A1) gene. Hum. Mutat. 28, 19-26. Piatigorsky, J. (2001) Enigma of the abundant water-soluble cytoplasmic proteins of the cornea, the ‘refraction, hypothesis. Cornea 20, 853-858. Rao, V.V., Tsai, M.J.’ Pan, X. and Chang, Y.F. (1993) L-Pipecolic acid oxidation in rat: Subcellular localization and developmental study. Biochim. Biophys. Acta. 1164. 29-35. Rasko, J.E., Klenova, E.M., Leon, J., Filippova, G.N., Loukinov, D.I., Vatolin, S.. Robinson, A.F., Hu, Y.J., Ulmer, J., Ward, M.D., Pugacheva, E.M., Neiman, P.E., Morse, H.C. Ill, Collins, S.J. and Lobanenkov, V.V. (2001) Cell growth inhibition by the multifunctional multivalent zinc-finger factor CTCF. Cancer Res. 61, 6002-6007. 1 IS Rcgc\ -Riul/ki, N. and Pines, (). (2007) Ixlipscd clislribulion: a phenomenon of dual largcling of pro loin and its significance. Biocssays 29, 772-782. Rcntcro, C. and Puigdomenech, P. (2006) Specific use of start codons and cellular localization of splice variants of human phosphodiesterase 9A gene. BMC Mol. Biol. 7, 39. Reya, T.. Morrison, S.J., Clarke, M.F. and Weissman, I.L. (2001) Stem cells, cancer, and cancer stem cells. Nature 414. 105-111. Rizzo, W.B. and Carney, G. (2005) Sjogren-Larsson syndrome, diversity of mutations and polymorphisms in the fatty aldehyde dehydrogenase gene (ALDH3A2). Hum. Mutat. 26, 1-10. Robin, M.A., Anandatheerthavarada, H.K., Biswas, G., Sepuri, N.B., Gordon, D.M., Pain, D. and Avadhani, N.G. (2002) Bimodal targeting of microsomal CYP2E1 to mitochondria through activation of an N-terminal chimeric signal by cAMP-mediated phosphorylation. J. Biol. Chem. 277, 40583-40593. Rodrigues, S.M., Andrage, M.O., Gomes, A.P., Damatta, F.M., Baracat-Pereira, M.C. and Pontes, E.P. (2006) Arabidopsis and tobacco plants ectopically expressing the soybean antiquitin-like ALDH7 gene display enhanced tolerance to drought, salinity, and oxidative stress. J. Exp. Bot. 57, 1909-1918. 1卜） Ruiubcro, A., Martin, J.l:, Liinihrcras, M.A., Liras, \\ and l^sniahan, (’.（1()。5) dicniical synthesis of al lysine ethylene acclal and conversion in si hi into l-piperidcinc-6-carboxYlic acid: key intermediate of the alpha-aminoadipic acid lor bcta-laclam antibiotics biosynthesis. Bioorg. Med. Chem. 3, 1237-1240. Salomons, G.S., Bok, L.A., Slruys, E.A., Pope, L.L., Darmin, P.S., Mills, P.B., Clayton, P.T., Willemsen, M.A. and Jakobs, C. (2007) An intriguing "silent,, mutation and a founder effect in antiquitin (ALDH7A1). Ann. Neurol. 62, 414-418. Sass, E.. Karniely, S. and Pines, O. (2003) Folding of fumarase during mitochondrial import determines its dual targeting in yeast. J. Biol. Chem. 278, 45109-45116. Scharer, G, Brocker, C.，Vasilioii, V., Creadon-Swindell, G.，Gallagher, R.C., Spector, E. and van Hove, J.L. (2010) The genotypic and phenotypic spectrum of pyridoxine-dependent epilepsy due to mutations in ALDH7A1. J. Inherit. Metab. Dis. 3, 571-581. Schmitt, B., Baumgartner, M., Mills, P.B., Clayton, P.T.’ Jakobs, C., Keller, E. and Wohlrab, G. (2010) Seizures and paroxysmal events: symptoms pointing to the diagnosis of pyridoxine-dependent epilepsy and pyridoxine phosphate oxidase deficiency. Dev. Med. Child. Neurol. 52, 133-142. Schwendener, S., Raynard, S., Paliwal, S., Cheng, A., Kanagaraj, R., Shevdev, I., Stark, J.M., Sung, P. and Janscak, P. (2010) Physical interaction of RecQ5 with RAI)51 facilitates its anti-recombinase activity. J. Biol. Chem. 285, 15739-15745. i:o SheiT, C\J.(卜)()4) Cll phase progression: cycling on CLIC. ( L'11 79, 551-555. Shin, J.11., Kim, S.R. and An, G. (2()()9) Ricc aldehyde dehydrogenase? is needed lor seed maturation and viability. Plant Physiol. 149, 905-915. Silva, l.A.’ Bai, S., McLean, K., Yang, K., Griffith, K.A., Dafydd, T., Ginestier, C., Johnston, C., Kueck, A., Reynolds, K., Wicha, M.S. and Buckanovich. R.J. (2011) Aldehyde dehydrogenase and CD 133 define angiogenic ovarian cancer stem cells that portend poor patient survival. Cancer Res. 71, 3991-4001. Skvorak, A.B., Robertson, N.G., Yin, Y., Weremowicz, S.，Her, H., Bieber, F.R., Beisel, K.W., Lynch. E.D., Beier, D.R. and Morton, C.C. (1997) An ancient conserved gene expressed in the human inner ear: identification, expression analysis, and chromosomal mapping of human and mouse antiquitin {ATOl). Genomics 46, 191-199. Sladek, N.E. (1999) Aldehyde dehydrogenase-mediated cellular relative insensitivity to the oxazaphosphorines. Curr. Pharm. Des. 5, 607-625. Sladek, N.E., Kollander, R., Sreerama, L. and Kiang, D.T. (2002) Cellular levels of aldehyde dehydrogenases (ALDHl Al and ALDH3A1) as predictors of therapeutic responses to cyclophosphamide-based chemotherapy of breast cancer: a retrospective study. Rational individualization of oxazaphosphorine-based cancer chemotherapeutic regimens. Cancer Chemother. Pharmacol. 49，309-321. 121 Snull, L Willi/., 11.，Akashi, K. and Mircaii, II. (199S) Two birds with one sloiic: genes thai cncode products targeted lo two or more compartmcnls. Hant Mol. I^iol. 38. 26S-277. Soda, K., Misono, H. and Yamamolo, T. (1968) L-Lysine: alpha-ketoglutarate aminotransferase. 1. Identification of a product, delta-1-piperideine-6-carboxylic acid, Biochemistry 7, 4102-4109. Sorolla. M.A., Rodn'guez-Colman, M.J., Tamarit, J., Ortega, Z., Lucas, J.J., Ferrer, I.. Ros, J. and Cabiscol, E. (2010) Protein oxidation in Huntington disease affects energy production and vitamin B6 metabolism. Free Radic. Biol. Med. 49, 612-621. Stagos, D., Chen, Y., Cantore, M., Jester, J.V. and Vasiliou, V. (2009) Corneal aldehyde dehydrogenases: multiple functions and novel nuclear localization. Brain Res. Bull. 81,211-218. Steinmetz, C.G., Xie, P., Weiner, H. and Hurley, T.D. (1997) Structure of mitochondrial aldehyde dehydrogenase, the genetic component of ethanol conversion. Structure 5, 701-711. 122 Stcl/1’ II., Worm, 11., Lalovvski, M., Ilacnig, (’.，Brcmbcck, l-.Il., (iochlcr. 11., Strocdicko, M., /cnkncr, M., Schocnhcrr, A., Koepp⑶，S., l imm, J., Mint/lalT. S.’ Abraham, C., Bock, N.’ Kielzmann, S., Gocdde, A., loksoz, 1:., Drocgc, A., Krobilsch, S., Korn, B., Birchmeier, W., Lehrach, 11, and Wanker, E.E. (2005) A human protein-protein interaction network: a resource for annotating the proteome. Cell 122. 957-968. Stephen, A.G., Trausch-Azar, J.S.，Ciechanover, A. and Schwartz, A.L. (1996) The iibiquitin-activating enzyme El is phosphorylated and localized to the nucleus in a cell cycle-dependent manner. J. Biol. Chem. 271, 15608-15614. Striano, P., Battaglia, S., Giordano, L., Capovilla, G., Beccaria, R, Struys, E.A., Salomons, G.S. and Jakobs, C. (2008) Two novel ALDH7A1 (antiquitin) splicing mutations associated with pyridoxine-dependent seizures. Epilepsia 50, 933-936. Stroeher, V.L., Boothe, J.G. and Good, A.G. (1995) Molecular cloning and expression of a turgor-responsive gene in Brassica nap us. Plant Mol. Biol. 27, 541-551. Takahashi, Y., Ogra, Y. and Suzuki, K.T. (2004) Synchronized generation of reactive oxygen species with the cell cycle. Life Sci. 75, 301-311. Tang, W.K., Cheng, C.H.K and Fong, W.P. (2002) First purification of the antiquitin protein and demonstration of enzymatic activity. FEBS Lett. 516, 183-186. |：3 lang, W.K., riiaii, C\B., Clicng, CMl.K. and I'ong, W.l). (2005) Scahrcaiii aiiliquiliii: niolccular cloning, tissue distribution, subcellular localization and liinclional expression. FliBS Lett. 579, 3759-3764. Tang, W.K., Wong, K.B., Lam Y.M., Cha’ S.S., Cheng, C.H.K. and Fong, W.P. (2008a) The crystal structure of seabream antiquitin reveals the structural basis of its substrate specificity. FEBS Lett. 582, 3090-3096. Tang. Y.B., Liu, Y.J.’ Zhou, J.G., Wang, G丄.，Qiu, Q.Y. and Giian, Y.Y. (2008b) Silence of ClC-3 chloride channel inhibits cell proliferation and the cell cycle via Gl/S phase arrest in rat basilar arterial smooth muscle cells. Cell Prolif. 41，775-785. Tsai, C.H. and Henderson, L.M. (1974) Degradation of o-phosphohydroxylysine in rat liver. Purification and properties of 2-aminoadipic semialdehyde dehydrogenase. J. Biol. Chem. 249, 5790-5792. Tsybovsky, Y., Donate, H.’ Krupenko, N.I., Davies, C. and Krupenko, S.A. (2007) Crystal structures of the carboxyl terminal domain of rat 10-formyltetrahydrofolate dehydrogenase: implications for the catalytic mechanism of aldehyde dehydrogenases. Biochemistry 46，2917-2929. Ucar, D., Cogle, C.R., Zucali, J.R., Ostmark, B., Scott, E.W.’ Zori, R., Gray, B.A. and Moreb, J.S. (2009) Aldehyde dehydrogenase activity as a functional marker for lung cancer. Chem. Biol. Interact. 178, 48-55. 1：4 Uma, 1.., Ilariharan, j, Shamia, Y. and Ikilasubramanian, I). (19%) Conical aldehyde dehydrogenase displays anlioxidanl properties. \ixp. l{yc Res. 63, 117-120. van den Hoogen, C., van der Horst, G, Cheung, H., Buijs, J.T., Lippilt, J.M., Guzman-Ramirez, N., Hamdy, F.C., Eaton, C.L., Thalmann, G.N., Cecchini, M.G.. Pelgen R.C. and van der Pluijm, G. (2010) High aldehyde dehydrogenase activity identifies tumor-initiating and metastasis-initiating cells in human prostate cancer. Cancer Res. 70, 5163-5173. van den Hoogen, C.’ van der Horst, G., Cheung, H., Buijs, J.T., Pelger, R.C. and van der Pluijm, G. (2011) The aldehyde dehydrogenase enzyme 7A1 is functionally involved in prostate cancer bone metastasis. Clin. Exp. Metastasis DOI 10.1007/S10585-011-9395-7 van Loon, A.P. and Young, E.T. (1986) Intracellular sorting of alcohol dehydrogenase isoenzymes in yeast: a cytosolic location reflects absence of an amino-terminal targeting sequence for the mitochondrion. EMBO J. 5, 161-165. Vasiliou, V., Bairoch, A., Tripton, K.F. and Nebert, D.W. (1999) Eukaryotic aldehyde dehydrogenase (ALDH) genes, human polymorphisms, and recommended nomenclature based on divergent evolution and chromosomal mapping. Pharmacogenetics 9, 421-434. Vasiliou, V., Pappa, A. and Estey, T. (2004) Role of human aldehyde dehydrogenases in cndobiotic and xcnobiotic metabolism. Drug Metab. Rev. 36, 279-299. 125 Yasilimi, \\ and Nobcrl, D.W. (2005) Analysis and update of ihc hiiniaii aldclulyc deh\drogcnasc (ALl)ll) gene family. Hum. Genomics 2, 138-143. W'ilzlum, J.L. (1994) The oxidation hypothesis of atherosclerosis. I.ancct 344, 793-795. Wong, W.Y. (2005) Inducibility and overexpression studies of antiquitin in HEK293 and HepG2 cells. MPhil Thesis. The Chinese University of Hong Kong. Wong, W.Y., Cheng, C.H.K. and Fong, W.P. (2006) Lack of inducibility of antiquitin (ALDH7A1) in cultured human embryonic kidney (HEK293) cell under osmotic or oxidative stress. In: Enzymology and Molecular Biology of Carbonyl Metabolism, Vol.12: Weiner, H., Plapp, B., Lindahl, R. and Maser, E. (eds.), pp. 96-103, Purdue University Press, Indiana. Wong, J.W., Chan, CI., Tang, W.K., Cheng, C.H. and Fong, W.P. (2010) Is antiquitin a mitochondrial enzyme? J. Cell Biochem. 109, 74-81. Wyn Jones, R.G and Storey, R. (1981) Betaines. In: The Physiology and Biochemistry of Drought Resistance in Plants, Paleg, L.G., Aspinall, D. (eds.), pp. 171-204. Academic Press, Sydney. Yang. K., ! litomi. M. and Staccy. D.W. (2006) Variations in cyclin Dl levels through the cell cycle determine the proliferative fate of a cell. Ccll l )iv, 1, 32. Yam:kla, K., Mori, 11. and Yamaki, S. (1999) Idcntilicalion and cDNA cloning ol a protein abiindanlly expressed during apple Iruil development. Plant Ccll Physiol. 40. 1^)8-204. Zheng，L., Roeder, R.G. and Luo’ Y. (2003) S phase activation of the histone H2B promoter by OCA-S, a coactivator complex that contains GAPDH as a key component. Cell 114, 255-266. Zhu, J.K. (2001) Plant salt tolerance. Trends in Plant Sci. 6, 66-71. Zhou, J.B., Yang, J.K., Zhao, L. and Xin, Z. (2010) Variants in KCNQl, AP3S1, MAN2A1, and ALDH7A1 and the risk of type 2 diabetes in the Chinese Northern Han population: a case-control study and meta-analysis. Med. Sci. Monit. 16， 179-183. Zinovieva, R.D., Tomarev, S.I. and Piatigorsky, J. (1993) Aldehyde dehydrogenase derived omega-crystallins of squid and octopus. Specialization for lens expression. J. Biol. Chem. 268, 11449-11455. Ziouzenkova, O. and Plutzky, J. (2008) Retinoid metabolism and nuclear receptor responses: New insights into coordinated regul ation of the PPAR-RXR complex. FEBS Lett. 582, 32-38. 127 CUHK Libraries _____ 004806873