Consistent downregulation of human lactoferrin gene, in the common eliminated region 1 on 3p21.3, following tumor growth in severe combined immunodeficient (SCID) mice

Cancer Letters 191 (2003) 155–164 www.elsevier.com/locate/canlet

Consistent downregulation of human lactoferrin gene, in the common eliminated region 1 on 3p21.3, following tumor growth in severe combined immunodeficient (SCID) mice Ying Yang, Jingfeng Li, Anna Szeles, Marta P. Imreh, Maria Kost-Alimova, Hajnalka Kiss, Irina Kholodnyuk, Ludmilla Fedorova, Eva Darai, George Klein, Stefan Imreh* Microbiology and Tumor Biology Center, Karolinska Institute, Nobelsva¨gen 16, 171 77 Stockholm, Sweden Received 10 March 2002; received in revised form 10 September 2002; accepted 11 November 2002

Abstract Lactoferrin (LF ) is one of 19 active genes in the common eliminated region 1 at 3p21.3 identified by us. LF was transfected into mouse fibrosarcoma A9. Fourteen severe combined immunodeficient (SCID) derived tumors from two PI based artificial chromosome (PAC)-transfectants containing the entire LF gene and two LF-cDNA transfectants were analyzed by real time polymerase chain reaction at the DNA and RNA level. Following SCID tumor passage, LF expression was decreased or eclipsed, in all tumors although DNA levels did not change considerably. Promoter methylation and/or rearrangement of the insertion site may be responsible for human LF downregulation in mouse fibrosarcoma derived tumors. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Human lactoferrin gene; Elimination test; Real time-polymerase chain reaction; Common eliminated region 1 (CER1); 3p21.3

1. Introduction Using human chromosome 3 (chr 3)/monochromosomal A9 fibrosarcoma hybrids, we previously reported that an , 1 Mb large fragment designated as the common eliminated region 1 (CER1) at 3p21.3 was regularly eliminated from tumors that have grown in severe combined immunodeficient (SCID) mice. The same region is affected by deletions and loss of heterozygosity (LOH) in many tumors [1 –3]. We have bridged CER1 by a PI based artificial chromosome (PAC) contig. It was found to contain 19 genes * Corresponding author. Tel.: þ46-8-728-6770; fax: þ 46-8-330498. E-mail address: [email protected] (S. Imreh).

by sequencing [4 – 6]. One of them is the lactoferrin (LF ) gene. To test the possibility that LF is a tumor antagonizing gene that is responsible for or contributes to CER1 elimination, we transfected a PAC containing LF and an LF-cDNA, respectively, into mouse A9 fibrosarcoma cells. Transfectants that expressed LF were inoculated into SCID mice. The presence of the LF gene and its expression was assessed by real time polymerse chain reaction (PCR). The promoter region was examined for the methylation status of its CpG sites. All tumors decreased or eclipsed LF expression following SCID passage, while the original transfectants regularly expressed the gene. Six and eight CpG

0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. doi: 1 0 . 1 0 1 6 / S 0 3 0 4 - 3 8 3 5 ( 0 2 ) 0 0 6 7 7 - 8

156

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

sites (out of 14) were methylated in two tumors tested (six sites were common). None of the 14 sites were methylated in the in vitro maintained transfectants.

software Image-Pro Plus (Media Cybernetics) and Adobe Photoshop. 2.3. DNA preparation

2. Materials and methods 2.1. Cell lines and transfections The mouse fibrosarcoma cell line A9, used as recipient of human chr 3 in microcell hybrids in the previous Et studies [1], was maintained in Iscove’s modified Eagle’s medium containing 10% fetal calf serum. For transfections the following probes were used: † RP6-PAC 19p19 containing full length LF gene in a blasticidin selectable pPAC4 vector [7]; † LF-cDNA, neomycin selectable, in pcDNA3 Invitrogen vector, kindly sent by Philip Furmanski (Department of Biology, New York University); † control RP6-PAC 188g11 that locates , 330 kb telomeric to PAC 19p19 and contains the full chemokine receptor 1 (CCR1 ) gene [4]. Fifty to 70% confluent monolayer A9 cells were transfected with a mixture of 1 – 2 mg PAC DNA or 1 mg LF cDNA and 10 ml Lipofectamine (Life Technologies) per well, in six well plates according to standard protocols. The transfectants were selected on blasticidin (1 – 2 mg/ml) and neomycin (500 mg/ml). The mean transfection efficiency was 4/5 £ 104 and 8/5 £ 104, respectively. Seven and five wells with three to four clones each were chosen and expanded in vitro for chromosome and DNA analyzes and inoculation into SCID mice. All transfectants were analyzed by fluorescent in situ hybridization (FISH). 2.2. FISH FISH with PAC 19p19, LF cDNA and PAC188g11 as probes was performed as described [1]. FISH and image analyzes were performed using a fluorescent microscope (LEITZ-DMRB, Leica, Heidelberg) equipped with a Hamamatsu 4800 cooled CCD camera (Hamamatsu, Herrsching) and processing

Genomic DNA was isolated by proteinase K digestion and followed by phenol/chloroform extraction [8]. DNA from PACs was prepared using CsCl ultra-centrifugation [8]. 2.4. Tumor formation in SCID mice LF-PAC transfectants 19p19-1, 19p19-7, LFcDNA transfectants LF-1, LF-2 and CCR1-PAC transfectant 188g11-9 were chosen for mouse inoculation. One million LF-PAC and LF-cDNA transfectants were inoculated subcutaneously into 6 week-old SCID mice. The mice were observed for tumor formation once a week up to 6 weeks. Three, four, three, four and four tumors from the transfectants mentioned above were successfully explanted and expanded for chromosome, DNA and RNA preparations. 2.5. RNA extraction, purification and cDNA synthesis Total RNA was extracted from 5 £ 106 cells in each sample using TRIzol Reagent (GIBCO BRL) and treated with Deoxyribonuclease I, Amplification Grade (GIBCO BRL). Reverse transcription of RNA was done mainly according to the protocol of SuperScripte II. Briefly, the reaction was performed in a final volume of 40 ml containing 4 mg total RNA, 2 ml (0.2 mg/ml) random hexamers (Pharmacia Uppsala, Sweden), 2 ml Oligo (dT)14 (500 mg/ml). The samples were incubated at 258C for 10 min, 708C for 10 min and quickly chilled on ice. Eight ml 5 £ standard buffer, 4 ml 0.1 M 1,4-dithio-DL -threitol (DTT), 200 units of SuperScripte II (GIBCO BRL), 40 units RNaseOUTe Ribonuclease inhibitor (recombinant) (GIBCO BRL) and 2 ml of 10 mM dNTP were added, then incubated at 428C for 50 min and 708C for 15 min. 2.6. Primers, probes, and real time reverse transcription (RT)-PCR Primers and probes for human lactoferrin gene were designed with the assistance of Primer Express

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

(Perkin-Elmer Applied Biosystems, Foster City, CA). Primers and probes were purchased from Life Technologies (GIBCO BRL) and MedProbe AS (Oslo, Norway), respectively. The sequences of primers and probes for LF cDNA are forward: 50 AGGAGGATGCCATCTGGAAT-30 , reverse: 50 AACCCAATGGCAGAGTCCTT-30 , probe: 5 0 CTCCGCCAGGCACAGGAAAAG-30 ; for LF DNA, forward: 50 -TTCCAGATGGCATCCTCCTT30 , reverse: 50 -CCTGTCAGACGAGGCTGAAA-30 , probe 50 -ACTTCGTGCCACAACGGCATGAGA-30 ; for CCR1, forward: 50 -TTCCCTTCTGGATCGACTACAA-30 , reverse: 5 0 -CTGGTGATGACACCAAAAGTGA-30 , probe: 50 -AGGCAAACACGGCG TGGACGAT-30 . The rodent GAPDH probe (VICe) (Applied Biosystems) was used as internal control. Quantitative RT-PCR was carried out on ABI Prism 7700 Sequence Detection System (Applied Biosystems) with the final volume of 20 ml containing 1 £

157

TaqMan Universal PCR Master Mix (Applied Biosystems), 5 ml of each appropriately diluted cDNA (1 –2 ml reverse transcriptase sample) or DNA (50 ng) sample, and 2 ml (2 mM) of each forward, reverse primer and probe. The thermal cycling conditions comprised an initial step 508C for 2 min, denaturation of 958C for 10 min, 40 cycles at 958C for 15 s, and 608C for 1 min. Triplicates of each sample were assayed in the experiment. The relative kinetic method was applied in the study by using a standard curve, which was constructed with 10-fold serial dilutions of each PCR products from LF-DNA, LFcDNA, CCR1-DNA, CCR1-cDNA and mouse GAPDH-DNA, GAPDH-cDNA. The quantities of LF-DNA, LF-cDNA, CCR1-DNA, CCR1-cDNA and mouse GAPDH-DNA, GAPDH-cDNA were determined by their standard curves from each run. The relative DNA and cDNA quantities of genes were normalized in each sample by dividing them with the

Table 1 Genomic DNA and cDNA levels as measured by real time PCR, and the in vivo/in vitro DNA and RNA ratios in LF and CCR1 transfectants and their derived SCID tumors Panels

DNA level

cDNA level

DNA-ratio tumors/in vitro line

cDNA-ratio tumors/in vitro line

19p19-1 19p19-1t1 19p19-1t4a 19p19-1t4b

549 659 488 599

357 0.003 98 16

1.2 0.888 1.02

8.4 £ 1026 0.27 0.0448

19p19-7 19p19-7t2 19p19-7t3 19p19-7t4 19p19-7t5

127 128 114 264 95

495 255 368 17 14

1.00 0.897 2.01 0.75

0.52 0.743 0.034 0.028

Lf-1 Lf-1t5 Lf-1t6 Lf-1t8

523 303 438 144

308 40 4.7 5.7

0.58 0.837 0.27

0.129 0.015 0.018

Lf-2 Lf-2t9 Lf-2t10 Lf-2t11 Lf-2t12

1221 1069 1364 1066 1066

0.87 0.117 0.87 1.78

0.011 0.009 0.145 0.206

1.56 1.85 0.96 3.01

1.61 1.83 0.82 3.15

CCR1-9 CCR1-9t1 CCR1-9t2 CCR1-9t3 CCR1-9t4

88 138 163 85 265

82.5 0.9 0.75 12 17 158 255 290 130 499

158

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

quantities of internal control mouse GAPDH-DNA or -cDNA. The results of DNA and cDNA values in Table 1 followed this calculation. 2.7. Bisulfite modification, PCR amplification and genomic sequencing for methylation study DNAs (1 mg) were denatured by NaOH (final concentration, 0.2 M) for 10 min at 378C in a volume of 50 ml. Thirty ml of 10 mM hydroquinone (Sigma) and 520 ml of 3 M sodium bisulfite (Sigma) at pH 5, both freshly prepared, were added and mixed, and samples were incubated under mineral oil at 508C for 16 h. Modified DNA was purified using PCR purification kit JETquick according to the manufacturer (Saveen, Bad Oeynhausen, Germany) and eluted into 50 ml of water. Modification was completed by NaOH (final concentration, 0.3 M) treatment for 5 min at room temperature, followed by ethanol precipitation. DNA was resuspended in water and used immediately or stored at 2 208C [9]. By combination of bisulfite modification and Table 2 Frequency of different LF-PAC insertion signals in the in vitro cultivated lines 19p19-1 (A1–4) and 19p19-7 (B1–3) and their SCID derived tumors (A5–8 and B4 –12)a FISH signals on chromosomes A1 A2 A3 A4 A5 A6 A7 A8 B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11 B12 a

19p19-1

19p19-1 T1

67 11 11 11 0 0 0 0 19p19-7 63 25 12 0 0 0 0 0 0 0 0 0

48 0 0 0 19 14 14 5 19p19-7 T5 0 0 7 21 21 14 4 4 4 7 7 11

The codes in the first column correspond to those used on Fig. 5.

PCR amplification, unmethylated cytosine residues can be converted to thymines (C to T) while methylated cytosine will remain unaltered (C to C) [10,11]. For examining the completion of bisulfite modification, a pair of primers was designed inside the LF gene that can amplify a 449 pb PCR product containing three Msp I sites. The sequences were: forward, 50 -TGTTTTTTGTGTTTTGTTTGTT AGGTG-30 , reverse, 50 -CTCCTCACAACCCTATAAACCCATAAT-30 . The PCR conditions were 958C for 5 min, then 35 cycles of 958C 30 s, 568C 1 min and 728C 1 min, followed by 728C 7 min. Msp I is a methylation insensitive restriction endonuclease, its recognition site is 50 -CCGG-30 . Bisulfite treatment and PCR amplification can convert the Msp I recognition site, unmethylated CCGG to TTGG and methylated CmCGG to TCGG. By incomplete treatment, part of CCGG remains unchanged. The digestion of 449 bp PCR products by Msp I will then result in 94, 12, 59 and 284 bp DNA fragments. One pair of primer forward 50 -TTGAGATTAGAGTTGGGATAGGG-30 and reverse 50 -CCCCCAAACACCTACACTCA-30 were purchased for methylation screening of LF gene promoter. The PCR product contains 14 CpG dinucleotides (Fig. 1). For each PCR reaction 50 ng bisulfite-modified DNA was used. PCR was performed under the same conditions as for Msp I site amplification. The PCR fragments were cloned using TOPO TA Cloning Kit for sequencing according to instruction manual (Invitrogen, USA). Genomic sequencing of cloned PCR products was accomplished using BigDye-terminator cycle sequencing kit (Applied Biosystems) on ABI310 (Applied Biosystems).

3. Results 3.1. Maintenance of genomic LF-DNA in transfectant derived SCID tumors The relative level of LF-DNA was determined as described under M & M. Table 1 lists the results. The DNA levels in most 19p19-1, LF-2 and 19p19-7 derived tumors were similar to the in vitro transfectants, but 19p19-7t4 and LF-2t12 had higher levels. Only two of the 14 tumors tested (LF-1t5 and LF-1t8)

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

159

Fig. 1. Potential CpG methylation sites (bold and numbered) in the human lactoferrin promoter (unconverted normal sequences). The forward and reverse primers for complementing the bisulfite converted DNA sequences are shown in bold and underlined.

had reduced DNA levels (58 and 27% of the in vitro level, respectively).

3.2. Downregulation of LF expression in transfectant derived SCID tumors LF mRNA expression was strongly reduced or completely eclipsed in 12 of 14 examined tumors in contrast to the in vitro transfectants. Only two of 14 tumors showed expressions corresponding to 48 and 26% of the in vitro level, respectively (19p19-7t2, 19p19-7t3) (Table 1). Figs. 2a,b illustrates the DNA ratio (tumor DNA/in vitro transfectant DNA) and cDNA ratio (tumor cDNA/in vitro transfectant cDNA) for each tumor. In all cases the reduction of expression was not related to losses at the DNA level. Two tumors 19p19t4 and LF-2t12 had higher levels of DNA than the original in vitro transfectant, but even in those the expression levels were strongly diminished during in vivo growth. LF-t5 and t8 had 4.5 and 15 times less cDNA values than the corresponding DNA levels.

3.3. No significant loss of CCR1 expression in transfectant derived SCID tumors Another CER1 localized gene, chemokine receptor 1 (CCR1 ) was tested in parallel. Genomic DNA and cDNA analysis was performed on tumors, derived from A9 cells transfected with genomic CCR1 DNA (in a PAC). The real time PCR amounts of CCR1 DNA and cDNA in vitro transfectant and in tumors are listed in Table 1. The DNA and cDNA ratios for each tumor versus in vitro clone are shown in Fig. 2c. They were not affected by the in vivo passage. 3.4. Methylation of the human LF promoter region in LF-PAC transfectant derived SCID tumors Two tumors from two LF-PAC transfectants, 19p19-1t1 (no expression) and 19p19-7t5 (low expression) were chosen. The PCR products were tested by Map I digestion (see Section 2). Fig. 3 shows the complete bisulfite modification. Eight, ten, seven and ten subclones from 19p19-1, 19p19-1t1, 19p19-7 and 19p19-7t5 were sequenced (Fig. 4). No CpG methylation was detected in any of the subclones from

160

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

Fig. 2. DNA and cDNA levels of LF and CCR1 gene in tumors derived from LF-PAC transfectants (a); LF-cDNA transfectants (b); and the CCR1-PAC transfectant (c). The grey and black bars represent DNA and cDNA levels for each tumor (Y axis calculated as: tumor DNA and RNA levels divided by the DNA and RNA levels of their own in vitro transfectants, see Section 2).

two in vitro transfectants 19p19-1 and 19p19-7. In all 20 subclones from 19p19-1t1 and 19p19-7t5 methylated CpG sites were found in the promoter region. The methylation patterns were different in these two tumors but six CpG sites: 4, 5, 9, 11, 12, 13, were common. 3.5. In vivo diversification of insertions in PACtransfectant derived tumors Fig. 5 and Table 2 shows the metaphase FISH analysis of the two LF-PAC transfectant lines and their tumors. The size of PAC signal and the morphology of host mouse chromosomes made it possible to follow the different subpopulations. In the

in vitro maintained 19p19-1 line a dominant (67%, A1 on Fig. 5 and Table 2) and three minority subpopulations were found. In the tumor (19p19-1 T1) the dominant subpopulation persisted, although somewhat diminished in frequency. Interestingly half of the tumor cell population consisted of four additional different clones not seen in vitro (A5– 8). Three subpopulations have been recorded in the 19p19-7 in vitro line (B1 –3 on Fig. 5 and Table 2). The in vitro dominant subpopulation has disappeared in the tumor. In vivo, 97% of the cells have rearranged the transfected LF-PAC carrying chromosomes, nine additional subpopulations were recorded (B4 – 12). Interphase FISH analysis of additional eight tumors

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

161

Fig. 3. Digestion of PCR products from bisulfite treated DNA by Msp I. Lanes 1, 3, 5 and 7 are PCR products from 19p19-1, 19p191 t1, 19p19-7 and 19p19-7 t5 bisulfite treated DNA. The size is 449 bp. Lanes 2, 4, 6 and 8 are their Msp I digested PCR products, showing the completion of bisulfite treatment due to its unchanged size (incomplete digestion should result in 94, 12, 59 and 284 bp products). Lane 9 is the control PCR fragment with size 293 bp (amplified by RFLP marker D3S30) and lane 10 are its Msp I digestion products with sizes 224 and 69 bp indicating the efficiency of Msp I.

suggest that the in vivo diversification may be a general phenomenon (data not shown).

4. Discussion LF is one of the 19 CER1 mapped genes regularly eliminated from SCID derived tumors of human chr 3/mouse fibrosarcoma microcell hybrids [2,4]. In human chr 3/human Renal cell carcinoma (RCC) microcell hybrid tumors LF was either eliminated or interrupted by a chromosome break (inversion) [3]. In the present study we have examined PAC transfectants that contained the full LF gene in normal regulatory context or LF-cDNA transfectants. Real time PCR has shown, that although in vitro both

Fig. 4. Methylation pattern of 14 CpG sites in LF promoter in subclones of LF-PAC transfectants 19p19-1, 19p19-7 and their tumors 19p19-1 t1 and 19p19-7 t5. White and black boxes indicate unmethylated and methylated CpG sites, respectively. Note that no methylation was found in the transfectants. Six common methylation sites were found (4, 5, 9, 11, 12, 13) in tumors.

categories expressed LF, during SCID passage this was either eclipsed or significantly diminished in 12/14 transfectant derived tumors. We excluded the possibility that the expression losses may be due to in vivo eliminations at DNA level. In contrast, CCR1 – another gene located in CER1 – maintained similar DNA and cDNA levels after in vivo tumor growth. Earlier it was demonstrated that human LF inhibits the growth of transplanted, v-ras driven solid tumors and a methylcholanthrene-induced fibrosarcoma. It reduced lung metastases of B16 – F10 melanoma cells in syngeneic mice [12]. The alternative form of human LF mRNA (DLF ) was expressed in normal tissues but not in 14 tumor cell lines [13]. Bovine LF can inhibit intestinal polyposis, colon, oesophagus and lung carcinogenesis and metastasis in rodents [14 – 17]. Bovine lactoferrin derived peptide, lactoferricin-B induced apoptosis and G1 arrest in a Tcell leukemia cell line [18]. The oral administration of bovine lactoferrin can inhibit Vascular endothelial growth factor (VEGF) (165)-mediated angiogenesis in the rat [19]. We have attempted to examine human LF promoter methylation by bisulfite treatment and sequencing. We have chosen only one of each tumor which showed either lost or highly reduced expression of LF from the two independent LF-PAC

162

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

Fig. 5. Examples of metaphase FISH analysis of the two LF-PAC transfectant lines and their tumors using PAC 19p19 DNA as probe. Green or red signals represent LF-PAC sequences. A1–4: insertions in the in vitro line 19p19-1; A5– 8: insertions in the SCID derived tumor 19p19-1 t1; B1 –3: insertions in the in vitro line 19p19-7; and B4–12: insertions in the SCID derived tumor 19p19-7 t5. Inset (A1) shows the same insertion site on the mouse chromosome commonly found both in transfectant (19p19-1) and in a derived tumor (19p19-1 t1). Inset (B3) shows the same insertion both in PAC transfectant (19p19-7) and in a derived tumor (19p19-7 t5). The codes correspond to those used on Table 2.

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

transfectants. Analysis of 20 subclones of two LFPAC transfectant derived tumors showed that six common promoter CpG sites were consistently methylated during tumor growth. In contrast, no methylated CpG site was found in 15 subclones of the two in vitro LF-PACs. This results indicate that promoter methylation may be one of the mechanisms responsible for the LF downregulation in LF-PAC derived SCID tumors. Cytogenetic (FISH) analysis of LF-PAC integration sites has shown a remarkable diversification in vivo. Between the multiple new subpopulations many pericentromeric localizations could be observed suggesting that downregulation or silencing could be a consequence of a position effect also. Downregulation of LF in transfectants suggests additional mechanisms of lactoferrin inactivation in tumors. Irrespective of the mechanism it strengthens the evidence that loss of LF provides the tumor cell with selective growth advantage. The gene remains a tumor suppressor gene candidate in CER1.

[4]

[5]

[6]

[7]

[8]

[9]

Acknowledgements [10]

Human lactoferrin cDNA was kindly provided by Dr Philip Furmanski NY University. This work was supported by: the Swedish Cancer Society and by the Cancer Research Institute/Concern Foundation, NY, LA.

[11]

[12]

References [13] [1] S. Imreh, I. Kholodnyuk, R. Allikmetts, E.J. Stanbridge, E.R. Zabarovsky, G. Klein, Non-random loss of human chromosome 3 fragments from mouse-human microcell hybrids following progressive growth in SCID mice, Genes Chromosomes Cancer 11 (1994) 237–245. [2] A. Szeles, Y. Yang, A.M. Sandlund, I. Kholodnyuk, H. Kiss, M. Kost-Alimova, E.R. Zabarovsky, E. Stanbridge, G. Klein, S. Imreh, Human/mouse microcell hybrid based elimination test reduces the putative tumor suppressor region at 3p21.3 to 1.6 cM, Genes Chromosomes Cancer 20 (1997) 329–336. [3] Y. Yang, M. Kost-Alimova, S. Ingvarsson, Q. Qianhui, H. Kiss, A. Szeles, I. Kholodnyuk, A. Cuthbert, G. Klein, S. Imreh, Similar regions of human chromosome 3 are eliminated from or retained in human/human and human/mouse microcell hybrids during tumor growth in severe combined immunode-

[14]

[15]

[16]

163

ficient (SCID) mice, Proc. Natl. Acad. Sci. USA 98 (2001) 1136–1141. Y. Yang, H. Kiss, M. Kost-Alimova, D. Kedra, I. Fransson, E. Seroussi, J. Li, A. Szeles, I. Kholodnyuk, M.P. Imreh, K. Fodor, G. Hadlaczky, G. Klein, J.P. Dumanski, S. Imreh, A 1Mb PAC contig spanning the common eliminated region 1 (CER1) in microcell hybrid-derived SCID tumors, Genomics 62 (1999) 147–155. H. Kiss, D. Kedra, C. Kiss, M. Kost-Alimova, Y. Yang, G. Klein, S. Imreh, J.P. Dumanski, The LZTFL1 gene is a part of a transcriptional map covering 250 kb within the common eliminated region 1 (C3CER1) in 3p21.3, Genomics 73 (2001) 10– 19. H. Kiss, D. Kedra, Y. Yang, M. Kost-Alimova, C. Kiss, K.P. O’Brien, I. Fransson, G. Klein, S. Imreh, J.P. Dumanski, A novel gene containing LIM domains (LIMD1) is located within the common eliminated region 1 (C3CER1) in 3p21.3, Hum. Genet. 105 (1999) 552 –559. P.A. Ioannou, C.T. Amemiya, J. Garnes, P.M. Kroisel, H. Shizuya, C. Chen, M.A. Batzer, P.J. de Jong, A new bacteriophage P1-derived vector for the propagation of large human DNA fragments, Nat. Genet. 6 (1994) 84 –89. J. Sambroook, E.F. Fritsch, T. Maniatis, Molecular Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1989. J.G. Herman, J.R. Graff, S. Myohanen, B.D. Nelkin, S.B. Baylin, Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands, Proc. Natl. Acad. Sci. USA 93 (1996) 9821–9826. R.Y. Wang, C.W. Gehrke, M. Ehrlich, Comparison of bisulfite modification of 5-methyldeoxycytidine and deoxycytidine residues, Nucleic Acids Res. 8 (1980) 4777–4790. M. Frommer, L.E. McDonald, D.S. Millar, C.M. Collis, F. Watt, G.W. Grigg, P.L. Molloy, C.L. Paul, A genomic sequencing protocol that yields a positive display of 5methylcytosine residues in individual DNA strands, Proc. Natl. Acad. Sci. USA 89 (1992) 1827–1831. J. Bezault, R. Bhimani, J. Wiprovnick, P. Furmanski, Human lactoferrin inhibits growth of solid tumors and development of experimental metastases in mice, Cancer Res. 54 (1994) 2310–2312. P.D. Siebert, B.C. Huang, Identification of an alternative form of human lactoferrin mRNA that is expressed differentially in normal tissues and tumor-derived cell lines, Proc. Natl. Acad. Sci. USA 94 (1997) 2198–2203. Y. Ushida, K. Sekine, T. Kuhara, N. Takasuka, M. Iigo, H. Tsuda, Inhibitory effects of bovine lactoferrin on intestinal polyposis in the Apc(Min) mouse, Cancer Lett. 134 (1998) 141 –145. Y. Ushida, K. Sekine, T. Kuhara, N. Takasuka, M. Iigo, M. Maeda, H. Tsuda, Possible chemopreventive effects of bovine lactoferrin on esophagus and lung carcinogenesis in the rat, Jpn. J. Cancer Res. 90 (1999) 262–267. Y.C. Yoo, S. Watanabe, R. Watanabe, K. Hata, K. Shimazaki, I. Azuma, Bovine lactoferrin and lactoferricin, a peptide derived from bovine lactoferrin, inhibit tumor metastasis in mice, Jpn. J. Cancer Res. 88 (1997) 184 –190.

164

Y. Yang et al. / Cancer Letters 191 (2003) 155–164

[17] M. Iigo, T. Kuhara, Y. Ushida, K. Sekine, M.A. Moore, H. Tsuda, Inhibitory effects of bovine lactoferrin on colon carcinoma 26 lung metastasis in mice, Clin. Exp. Metastasis (1999) 35–40. [18] Y.C. Yoo, R. Watanabe, Y. Koike, M. Mitobe, K. Shimazaki, S. Watanabe, I. Azuma, Apoptosis in human leukemic cells induced by lactoferricin, a bovine milk protein-derived

peptide: involvement of reactive oxygen species, Biochem. Biophys. Res. Commun. 237 (1997) 624–628. [19] K. Norrby, I. Mattsby-Baltzer, M. Innocenti, S. Tuneberg, Orally administered bovine lactoferrin systemically inhibits VEGF(165)-mediated angiogenesis in the rat, Int. J. Cancer 91 (2001) 236 –240.