INTRODUCTION  

The  CD151  gene  is  a  transmembrane molecule that has been characterized as a  member  of  the  tetraspanin  family.  The CD151  transcript  consists  of  253  amino acids that encode a 28 kDa protein. This gene  is  mapped  to  the  human  chromo- some 11p15.1. It is implicated in cell mo- tility,  cell  adhesion,  and  stability  and  for- mation of hemidesmosomes (Yauch et al. 1998).  The  role  of  CD151  gene  in  cell motility  highlighted  its  function  as  a  pro- moter  in  tumour  progression  and  metas- tasis  events  (Kohno  et  al.  2002).  Over- expression  of  CD151  was  found  to  be associated  with  poor  prognosis  in  non- small  cell  lung  cancer,  colon  cancer  and prostate  cancer  (Tokuhara  et  al.  2001; Hashida et al. 2003; Ang et al. 2004). In breast  cancer,  CD151  is  elevated  in  ap- proximately  31%  of  human  breast  can- cers  and  is  significantly  related  to  the high-grade  (40%)  and  estrogen  receptor negative (45%)  subtypes  (Yang  et  al. 2008). The CD151 gene is also a poten- tial  marker  in  predicting  histological grading  in  colon  and  prostate  cancer (Hashida et al. 2003; Ang et al. 2004). 

Study on the CD151 mRNA expression in  correlation  with  clinicopathological features of breast cancer is rare. Herein, we investigated the expression of CD151 and  correlated  its  expression  with  the histological  features,  ER  status,  PR  sta- tus,  c-erbB-2  expression  and tumour grades in breast cancer patients.

 MATERIALS AND METHODS  Patients and biopsy specimens   A  total  of  45  breast  carcinomas  were collected  from  fresh  surgical  resections from  Kuala  Lumpur  Hospital,  Universiti Kebangsaan  Malaysia  Medical  Centre and  Putrajaya  Hospital.  The  biopsies were  snap-frozen  in  liquid  nitrogen  prior to  sample  preparation.  Tissues  were touched  on  poly-L-Lysine  coated  slides and  subjected  for  Diff  Quick  Staining (Imeb  Inc.).  Microscopic  examination  of touch  preparations  verified  the  presence of at least 70% of cancer cells in all sam- ples.  Infiltrating  ductal  carcinoma  (IDC) was  diagnosed  in  84.4%  of  the  patients (N=39). Other tumours were identified as infiltrating  lobular  carcinoma  (N=1),  duc- tal  carcinoma  in-situ  (DCIS)  (N=2), mucinous  carcinoma  (N=1),  mixed  mu- cinous  &  ductal  carcinoma  (N=1),  and IDC  with  DCIS  (N=1).  Majority  of  the cases  were  from  postmenopausal  pa- tients. The mean and median ages of the patients are 50.8 and 52.0 years old, re- spectively.  Tumour  grade  was  evaluated using  the  Bloom-Richardson  grading system.  Nine,  24  and  12  tumours  were classified  as  Grade  1,  Grade  2  and Grade  3  tumours,  respectively.  The  ER, PR and c-erbB-2 expression status for all samples  were  determined  by  immuno- histochemistry (IHC) and the results were as  evaluated  by  the  reporting  patholo- gists.  Tumours  were  regarded  as  ER  or PR  positive  if  >10%  of  tumour  cells  ex- pressed  nuclear  positivity.  On  the  other hand, strong membrane staining in >10% cells  was considered positive  for  c-erbB- 2. Breast carcinomas with known ER, PR and  c-erbB-2  expression  were  used  as controls.  The  clones  for  ER,  PR  and  c- erbB-2  were  1DS  (DAKO),  PgR  636 (DAKO)  and  SP3  (Neomarker),  respec- tively.  All  IHC  procedures  were  per- formed  according  to  the  manufacturer’s recommendation.  The  clinical  and  histo- pathological  data  for  the  patients  are listed in Table 1. 

Ethics  approval  was  obtained  from Medical  Research  &  Ethics  Committee, Ministry  of  Health  Malaysia  to  perform this study.  RNA isolation   Minced  tissues  were  placed  directly  in Trizol  Reagent  (Invitrogen),  homoge- nized,  and total RNAs  were isolated and purified  through  RNeasy  columns  (QIA- GEN)  according  to  the  manufacturer’s recommendation. The integrity of the pu- rified total RNA was assessed by visuali- zation  of  the  28S/18S  ribosomal  RNA ratio  on  1%  agarose  denaturing  gel  and the quantity was assessed based on ab- sorbance at 260nm and 280nm in Nano- drop.   cDNA synthesis  Total RNA was converted to cDNA using AMV  reverse  transcriptase  First  strand cDNA  synthesis  kit  for  RT-PCR  (AMV), Roche Diagnostics]. The cDNA synthesis reactions  were  performed  in  20µL  vo- lumes  containing  200ng  of  total  RNAs, 1X  reaction  buffer,  5mM  MgCl2,  1mM dNTPs,  3.2µg  of  random  primer  p(dN)6 , 50 units RNase inhibitor and ≥ 20 units of AMV  reverse  transcriptase.  Total  RNAs with  the  appropriate  volume  of  nuclease free  water  were  first  incubated  at  65°C for  15min.  The  master  mix  was  then added to the tube and incubated at 25°C for  10min,  and  then  at  42°C  for  1h.  The reaction  was  then  incubated  at  99°C  for 5min  to  denature  the  enzymes.  Finally, the reaction was cooled to 4°C for 5min. 

Polymerase chain reaction (PCR)  

Polymerase  chain  reaction  was  per- formed  to  amplify  GAPDH  and  CD151 genes. The PCR reaction was carried out in  25µL  volumes  containing  4-6µg  of cDNA  generated  from  breast  tumour, 0.4pmole/µL  of  primers  and  1X  PCR master  mix  (Panomics).  The  reaction mixture  was  pre-incubated  at  95°C  for  5 min,  followed  by  40-45  cycles  of  amplifi- cation at 94°C for 30s, 54°C for 30s and 72°C  for  30s.  A  negative  control  was  in- cluded in each PCR to exclude contami- nation.  PCR  products  for  GAPDH  and CD151 genes were purified (Qiagen PCR purification  kit).  Then,  10X  serial  dilution were prepared for GAPDH and CD151 to generate standard curves in qRT-PCR. 

Real-time RT-PCR (qRT-PCR) 

Real-time  RT-PCR  was  performed  using LightCycler®  LC480  Probe  Master (Roche  Diagnostics).  Probes  for  CD151 (cat.  no.:  04688660001)  and  house- keeping  gene,  GAPDH  (cat.  no.: 04688589001) were purchased from Uni- versal  ProbeLibrary  (Roche  Diagnostics) and  the  primers  were  synthesized  from Bioneer.  The  primer  sequences  were CD151-F:  5’-CTG  CGC  CTG  TAC  TTC ATC  G-3’  and  CD151-R:  5’-TTC  TCC TTG  AGC  TCC  GTG  TT-3’;  GAPDH-F: 5’-CTC  TGC  TCC TCC  TGT TCG  AC-3’ and  GAPDH-R:  5’-ACG  ACC  AAA  TCC GTT  GAC  TC-3’.  qRT-PCR  was  carried out  in 20µL  volumes  containing  4µg  of cDNA,  1X  LightCycler®  LC480  Probe Master,  2µL  of  probes  and  0.8–1.0pmol of each primer. qRT-PCR was carried out on  LightCycler®   LC480  using  a  96  well plate  (Roche  Diagnostics).  All  reactions were  run  in  duplicates.  The  amplification program  consisted  of  pre-incubation  at 95°C  with  a  10min  hold,  denaturation  at 95°C  with  a  10s  hold,  followed  by  an- nealing  at  54°C  with  a  10s  hold  (40 cycles)  and  extension  at  72°C  with  a  1s hold.  The  sizes  of  the  amplicons  were 107  bp  and  112  bp  for  CD151  and GAPDH, respectively.  

Statistical analysis 

Standard curves for GAPDH and CD151 were  optimized  by  using  diluted  PCR products.  Gene  expressions  of  each sample  were  obtained  by  comparing  the crossing  points  (Ct)  with  the  standard curve.  The  obtained  gene  expression  or concentration  of  CD151  gene  was  then normalized  with  GAPDH.  Mean  expres- sions  of  CD151  for  each  case  were grouped  according  to  the  ER  status,  PR status,  c-erbB-2  expression  and  tumour grades.  The  significant  association  of CD151  expression  with  the  ER  status, PR status and c-erbB-2 expression were determined  by  t-test  unequal  variances while  the  relation  of  CD151  and  tumour status was determined by ANOVA.

RESULTS

 In our study, PCR efficiencies for CD151 and  GAPDH  were  1.674  and  1.758,  re- spectively.  The  normalized  mRNA  ex- pressions  of  CD151  for  all  the  samples examined in this study are listed in Table 1. The mean CD151 expression in differ- ent  histological  groups  are  illustrated  in Figure 1.

Our  preliminary  real  time  RT-PCR  re- sults  revealed  that  there  was  significant difference  between  CD151  expression and  the  ER  status  (p=0.012)  and  PR status (p=0.009) of breast tumour. Higher CD151  mRNA  level  was  detected  in  ER positive and PR positive breast tumours. Conversely,  CD151  expression  was  not associated  with  the  tumour  grade (p=0.057)  and  c-erbB-2  expression (p=0.060) in our present study. 

DISCUSSION

 Several  members  of  the  tetraspanin  su- perfamily  have  been  identified  as  me- tastasis  suppressor  genes  in  cancer pathways.  The  CD151  gene  was  re- ported  to  be  the  first  member  of  the  te- traspanin superfamily, which plays a cru-cial  role  as  promoter  in  cancer  metasta-sis  (Testa  et  al.  1999).  Notably,  CD151 gene  mediates  cell  migration  and  facili-tates invasion by regulating laminin (LN)- binding  intergrins  such  as  α3β1,  α6β1, α6β4,  and  α7β1  through  formation  of CD151-integrin  complexes  called  tetras- panin-enriched microdomains (TEM) (Liu et al. 2007). Recent study on knockdown CD151  expression  in  breast  cancer  cell showed  that  hepatocyte  growth  factor (HGF)  receptor/c-met  signaling  pathway might  be  changed  by  the  decreased  Akt phosphorylation  in  cells  lacking  CD151 (Klosek  et  al.  2009).  C-met  signaling pathway plays a crucial role in mitogenic, proliferative,  morphogenic  and  angi- ogenic activities. It correlates with breast cancer progression and metastasis (Park et al. 2005). The ability of CD151 in me- diating  cell  migration  suggested  that CD151  is  possibly  involved  in  a  molecu-lar  mechanism  that  could  lead  to  metas-tatic  progression  of  cancerous  cells  (Ha-shida  et  al.  2003).  This  hypothesis  is proven  by  Novitskaya  et  al.  (2010)  and Sadej  et  al.  (2010).  They  reported  that CD151  expression  is  increased  in  inva-sive  breast  carcinomas  and  higher  tu-mour  grade  and  node  metastasis.  Addi- tionally,  CD151  is  confirmed  as  a   prognostic marker of outcome in lung and prostate  cancers  (Tokuhara  et  al.  2001; Ang et al. 2004). 

Although  previous  studies  have  con- firmed  the  ability  of  CD151  in  facilitating cell migration and spreading, very limited information  is  available  on  the  relation- ship  between  CD151  expression  and hormone  profile  such  as  estrogen  re- ceptor  and  progesterone  receptor  status in  breast  cancer  patients.  The  present study  revealed  that  there  was  significant association  between  CD151  expression and ER positive (p<0.05) and PR positive subtypes  (p<0.01).  CD151  expression was  higher  in  ER  positive  compared  to ER negative breast tumours. It is also ex- pressed  at  a  higher  level  in  PR  positive relative  to  the  PR  negative  breast  tu- mours.  Since  ER  positive  breast  cancer patients  have  better  survival  than  ER negative  patients,  we  suggest  that  the expression  of  CD151  could  be  another predictor  for  a  better  survival  in  breast cancer  patients.  Our  findings  is  contrary with previous findings by Tokuhara et al. (2001),  Ang  et  al.  (2004)  and  Zijlstra  et al. (2008) which said that CD151 expres- sion  is  increased  in  advanced  stage  of the  disease  and  shorter  overall  survival. However, our result is consistent with the findings  reported  by  Voss  et  al.  (2011). Voss et al. (2011) postulated that CD151 expression  may  prevent  the  transition  of endometrial cancer to a more aggressive phenotype.  They  found  that  high  CD151 expression  is  correlated  with  improved survival in the patients. How CD151 pre- vent  the tumour  cell  transition  is  un- known.  The  pathway  or  biochemical function of CD151 in estrogen dependent breast  cancer  is  still  unknown  and  re- mains to be investigated. 

Yang  et al. (2008)  had  suggested possible connection between high CD151 transcript  level  and c-erbB-2  expression in breast tumour (Yang  et al. 2008). Our study revealed that there is no significant difference  between  CD151  expression and  c-erbB-2  expression.  Nevertheless, our  findings  revealed  lower  CD151 mRNA  level in triple  negative  (ER,  PR and  c-erbB-2  negative)  breast  carcino- mas.  The  average  CD151  expression level  in triple  negative  breast  tumours (113,  147,  164, 168 and  173)  is  0.14, which  is  approximately  one  fold  lower that  other  types  of  tumours  (0.27).  This observation  suggests  the important  role of  CD151  in  the  pathogenesis  of  triple negative  breast  tumours.  Thus,  c-erbB-2 expression  may  not  link  directly  with CD151 expression but its expression de- pends on ER and PR status as well. 

Several  research  teams  have  found that elevated CD151 mRNA level is asso- ciated  with  a  more  advanced  and  ag- gressive stage of the disease. In prostate cancer,  CD151  expression  was  signifi- cantly  different  among  well,  moderately and  poorly  differentiated  tumours  (Ha- shida et al. 2003; Ang et al. 2004). How- ever, CD151 expression is not correlated with  breast  tumour  grades  in  our  study (p<0.05).  Our  result  is  unexpected.  The reason for inconsistency of our result with previous  findings  could  be  due  to  the sample size in our study. More than 70% of  the  subjects  used  in  this  study  were grade 1 and 2 breast tumours (N=33/45). Our  results  showed  that  elevated CD151  expression  may  be  associated be  another  important  prognostic  marker and  treatment  decision  in  estrogen  de- pendent  breast  cancer  patients.  Exten- sive  study  of  CD151  mRNA  level  in es- trogen-depe ndent  breast  tumours  is  cru- cial  to  delineate  its  role  in  breast  cancer pathogenesis. 

Our  study  however  has  limitations  due to the inter-observer variability of the pa- thologists in the histopathology examina- tion  of  ER,  PR,  c-erbB-2  and  tumour grade.  Therefore,  larger  sample  size  is needed  to  verify  the  correlation  between CD151 expression level and ER and PR status.  Alternatively,  the  role  of  CD151 gene  could  be  determined  by  studying the function of the gene in breast cancer cell line.

 ACKNOWLEDGEMENTS

 This  research  was  supported  by  a  grant from the Ministry of Science and Innova-tion  (MOSTI).  The  authors  would  like  to thank  the  Director  General  of  Health Malaysia for his permission to publish the study  findings. We  also  like  to  thank  the Director  of  the  Institute  for  Medical  Re-search for her support in the approval of publishing the research findings.