GASTROENTEROLOGY 1997;112:1163 -1168
`
`Epithelial Cell Folate Depletion Occurs in Neoplastic But Not
`Adjacent Normal Colon Mucosa
`
`JOHN MEENAN, * EILEEN O'HALLlNAN,'!' JOHN SCOTT,'!' and DONALD G. WEIR*
`* Department of Clinical Medicine, Trinity College and St. James's Hospital, Dublin; and "Department of Biochemistry, Trinity College, Dublin,
`Ireland
`
`Background & Aims: Restricted folate supply is associ(cid:173)
`ated with the development of carcinoma, and folate
`supplements have a protective effect in colorectal car(cid:173)
`cinoma. This effect may be mediated through correc(cid:173)
`tion of local folate deficiency. The aim of this study
`was to define the folate content of neoplastic colonic
`epithelial cells and its relation to that of adjacent nor(cid:173)
`mal tissue and circulating levels. Methods: Epithelial
`cells were isolated from endoscopic biopsy specimens
`of normal, adenocarcinoma, adenoma, and adjacent
`normal colonic mucosa by ion chelation. Intracellular
`folate levels were determined by microbiological assay.
`Results: Folate levels in carcinoma specimens were
`lower than in adjacent normal tissue (P < 0.02). Levels
`in adenoma epithelial cells were lower than in adjacent
`normal tissue, although this did not reach statistical
`significance (P < 0.06). Epithelial cells from normal
`tissue and mucosa adjacent to tumors and adenomata
`had similar folate contents. Blood folate and vitamin
`B12 indices for all groups were normal. Conclusions:
`Malignant colon epithelial cells show a relative local(cid:173)
`ized folate deficiency. However, there is no evidence
`for the occurrence of generalized mucosal folate defi(cid:173)
`ciency. This finding suggests that folate supplements
`do not inhibit carcinogenesis through correction of lo(cid:173)
`calized folate depletion.
`
`T he epidemiology of colorectal carcinoma (CRC) sug(cid:173)
`
`gests exogenous (dietary constituents) and endoge(cid:173)
`nous (bile acid metabolites) substances to influence mu(cid:173)
`cosal epithelial cell proliferation and progression to
`malignancy. A protective role has been ascribed to vita(cid:173)
`mins and micro nutrients in carcinogenesis. 1
`,2 Folate is of
`particular interest because this vitamin is required for
`all cellular one-carbon transfer reactions including DNA
`methylation and thymidine synthesis. 3 Folate supple(cid:173)
`mentation favorably influences epithelial dysplasia in hu(cid:173)
`mans,4,5 reduces colon tumor load in animals exposed to
`carcinogens,6 and protects against the development of
`colonic neoplasia in patients using sulfonamide-con(cid:173)
`taining drugs. 7
`
`The efficacy of folate in suppressing epithelial neoplasia
`may be related to the importance of DNA methylation in
`cell homeostasis. In vitro studies show that focal loss of
`methyl groups, which influences gene expression and gen(cid:173)
`eralized genomic hypomethylation, is a feature of several
`carcinomas. H Increased levels of mucosal genomic methyla(cid:173)
`tion after supplementation with supraphysiological doses
`of folate supports this hypothesis. 9 These and similar stud(cid:173)
`ies suggest a general theory that carcinogenesis is related
`to an epigenetic factor, altered patterns of genomic meth(cid:173)
`y lation,]() although, they do not show a direct link between
`localized folate depletion and neoplasia.
`Low folate levels occur in mixed cell homogenates of
`potentially premalignant colonic adenomatous polyps.ll
`However, folate deficiency has not been shown in cells
`that produce CRC, the colonic epithelial cell (colono(cid:173)
`cytes). Furthermore, the relation between the folate con(cid:173)
`tent of colon adenoma or tumor and adjacent normal
`epithelial cells remains to be defined. Consequently, there
`is little evidence to show that folate supplementation
`influences carcinogenesis through the correction of tissue
`folate deficiency.
`The aim of this study was to define the folate content of
`epithelial cells isolated from premalignant and malignant
`distal colonic lesions and to compare folate content in
`both adjacent normal colonocytes and tissue specimens
`taken from nontumor-bearing mucosa.
`
`Materials and Methods
`Tissue Collection
`
`Colonic biopsy specimens were obtained at endoscopy
`under a protocol approved by the Ethics Committee of the
`Federated Voluntary Dublin Hospitals.
`Patient details including medical history and current medi(cid:173)
`cation were recorded before endoscopy. Venous blood was
`drawn for the estimation of red cell folate, serum folate, and
`vitamin B12 levels. Pathology in enrolled patients included
`distal colonic carcinoma (within 30 em of the anal verge) (n
`
`Abbreviation used in this report: CRC, colorectal carcinoma.
`© 1997 by the American Gastroenterological Association
`0016-5085/97/$3.00
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0001
`
`

`
`1164 MEENAN ET AL.
`
`GASTROENTEROLOGY Vol. 112, No. 4
`
`= 12), distal colon adenomatous polyps (n = 7), and control
`patients (n = 8). Age-matched control patients had no history
`of colorectal neoplasia and underwent endoscopy for the inves(cid:173)
`tigation of altered bowel habit or rectal blood loss. They had
`normal endoscopic findings defined by normal macroscopic
`appearance and absence of inflammation on rectal biopsy speci(cid:173)
`mens. All tumors were moderately differentiated adenocarcino(cid:173)
`mas and diagnosed as Dukes' disease stage B after surgery.
`All adenomatous polyps showed high-grade dysplasia with no
`evidence of coexisting frank malignancy. No patients were
`taking folate supplements or antifolate medication.
`In patients with colonic tumors or adenomatous polyps,
`biopsy specimens were taken from both the lesion itself and
`from normal-appearing mucosa within 5 em of the lesion (adja(cid:173)
`cent). Biopsy specimens from lesions for folate content estima(cid:173)
`tion were taken before those for histopathology, and blood
`contamination was avoided. Biopsy specimens from control
`patients were taken at 30 em from the anal verge.
`
`Colonic Epithelial Cell Isolation
`
`Four endoscopic biopsy specimens from each area of
`interest were taken into calcium- and magnesium-free Hank's
`buffered saline solution (Life Technologies, Paisley, Scotland)
`supplemented with 0.3% bovine serum albumin, penicillin,
`and gentamicin containing 0.75 mmoliL dithiothreitol (Sigma,
`St. Louis, MO) and processed for epithelial cell isolation as
`previously described. 12 Briefly, biopsy specimens were allowed
`to stand in this solution for 3 hours at room temperature.
`Subsequently, the tissue was transferred to fresh medium (cal(cid:173)
`cium- and magnesium-free Hank's buffered salt solution sup(cid:173)
`plemented with 0.3% bovine serum albumin) containing 2
`mmoliL ethylenediaminetetraacetic acid and placed on an in(cid:173)
`clined (45°), rotating table at 37°C for 1 hour. The resulting
`cell suspension was washed and pelleted in Hank's buffered
`salt solution. Cells were counted and viability was assessed
`using ethidium bromide-acridine orange. Cells were saved in
`500 ).ll ascorbate 0.1 % in phosphate-buffered saline (PBS) at
`- 20°C until assayed.
`
`Flow Cytometry
`
`Aliquots of epithelial cell isolates were assessed for
`lamina propria contamination and intraepithelial lymphocyte
`content by flow cytometry (FACScan; Becton Dickinson, Ant(cid:173)
`werp, Belgium) using the following monoclonal antibodies
`and fluorescent conjugates: Ber-Ep-4 (epithelium) (Dako,
`Glostrup, Denmark); CD3 (T cells) (Becton Dickinson); CD14
`(monocytes) (Becton Dickinson); CD19 (B cells) (Becton Dick(cid:173)
`inson) and R1SS (dendritic cells, a kind gift from Dr. A.
`Whelan, St. James's Hospital, Dublin, Ireland); rabbit anti(cid:173)
`mouse fluorescein isothiocyanate (Dako); and rabbit anti-mouse
`phycoerythrin (Dako). Simultest immunoglobulin (Ig) Gland
`IgG2a (Becton Dickinson) were used as control antibodies.
`Lamina propria remnants from the epithelial compartment
`isolation were digested in collagenase type IV SO U/mL
`(Sigma) for 1 hour at 37°C and used as positive controls for
`the determination of leukocyte contamination.
`
`Cells (2.5 X 106
`) were suspended in fluorescence-activated
`cell sorter buffer (PBS, 0.1% bovine serum albumin, and
`0.01 % sodium azide) and washed. The cells were then incu(cid:173)
`bated with previously determined optimal concentrations of
`the various monoclonal antibodies for 30 minutes at 4°C. After
`two more washes, cells were incubated similarly with a second
`antibody or fluorescent conjugate and again washed twice.
`Flow cytometry was performed using a FACScan (Becton-Dick(cid:173)
`inson), and 10 X 103 events were saved for analysis. Data were
`analyzed using Lysys II sofrware (Becton Dickinson).
`
`Folate Assay
`
`Cell isolates were defrosted, sonicated (Cell Disruptor
`B1S; Branson, Utrecht, Netherlands), and deconjugated to
`permit microbiological assay. Three hundred microliters of
`sonicate was placed in a water bath at 100°C for 10 minutes.
`After cooling, 10 IlL of chicken pancreas conjugase (a gift from
`Dr. A. Molloy, Trinity College, Dublin, Ireland) was added
`and the suspension incubated at 37°C for 2 hours. After incuba(cid:173)
`tion, the suspension was again boiled. Aliquots were assayed
`for folate content using a microtiter plate method previously
`described. 13 Recovery of folate was estimated by addition of
`tritium-labeled pteroylmonoglutamate (Amersham). A yeast
`extract (0.1 %) control was used to ensure uniform conjugase
`activity between assays. Serum, red cell folate, and vitamin
`B12 levels were measured by standard microbiological assay.
`Folate results were standardized with respect to spectropho(cid:173)
`tometric determination of sonicate DNA content using bromo(cid:173)
`deoxyuridine. Isolate DNA content of tumor specimens corre(cid:173)
`lated closely (r = 0.85; P < 0.001) with protein content (BCA
`protein assay; Pierce, Rockford, IL).
`The ability of the microtiter assay to detect folate levels in
`neoplastic cells was verified by a colonic epithelial cell line
`cultured in varying folate concentrations.
`HT29 (cl.19A) cells, a colonic epithelium carcinoma line,14
`were grown to confluency in 5 days in 2S-cm2 culture flasks
`(Costar, Cambridge, MA) in Dulbecco's modified essential me(cid:173)
`dium (Life Technologies, Paisley, Scotland) supplemented with
`1 % nonessential amino acids (Sigma), 10% fetal calf serum,
`and gentamicin. Folic acid (Sigma) was added to this medium
`in the concentrations of 0 )lmollL, 10 )lmollL, 100 )lmollL,
`or 1 mmoliL (Sigma). Medium was refreshed at 24 hours and
`every 48 hours thereafter. The level of folate supplement did
`not influence rate of confluency. At day 7, medium was dis(cid:173)
`carded and the cells were washed three times with PBS. Cell
`monolayers were disrupted with trypsin-ethylenediaminetet(cid:173)
`raacetic acid (Life Technologies). The single cell isolate was
`washed three times in PBS, pelleted, resuspended in 1 mL
`of ascorbate buffer, and saved at -20°C. After thawing, cell
`suspensions were sonicated and assayed for folate content.
`These experiments were performed in quadruplicate. Dul(cid:173)
`becco's modified essential medium contains 2.5 )lg/L folate.
`
`Statistical Analysis
`
`Statistical analysis was performed using SPSS for Win(cid:173)
`dows 6.0 (SPSS Inc., New York, NY). Data were compared
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0002
`
`

`
`April 1997
`
`EPITHELIAL FOLATE IN COLORECTAL NEOPLASIA 1165
`
`controls were not significant. A similar relation between
`tissue folate concentrations was seen when results were
`calculated in terms of isolate protein content (data not
`shown).
`Serum folate levels for control and tumor or adenoma
`groups were 8.6 ± 1.1 and 6.77 ± 1.28 ng/mL, respec(cid:173)
`tively (P > 0.05). Red cell folate concentrations for these
`groups were 340 ± 39 and 361 ± 72.4 ng/mL, respec(cid:173)
`tively. Serum and red cell folate levels correlated (r =
`0.56, P < 0.01). Neither serum nor red cell folate levels
`correlated with tissue levels in any group. Serum vitamin
`Bl2 levels for all groups were within normal limits (data
`not shown).
`
`Discussion
`
`A significant difference has been identified be(cid:173)
`tween the folate content of colon tumor epithelial cells
`and that of adjacent normal cells. Because enriched epi(cid:173)
`thelial isolates were used, these results more accurately
`reflect the folate content of those cells that give rise to
`CRC than is permitted by the use of whole mucosal
`biopsy homogenates. 11
`Folate coenzymes are essential for cellular one-carbon
`transfer reactions, specifically protein and DNA methyla(cid:173)
`tion and nucleotide synthesis, IS which identify a potential
`role in carcinogenesis. 2,16 The biological effects of folate
`deficiency, including defective DNA synthesis and im(cid:173)
`paired cell replication, are well recognized 17: misincorpo(cid:173)
`ration of nucleotides into nascent DNA,IH increased sus(cid:173)
`ceptibility to genetic damage,19 alterations in folate
`pools,20 instability of folate binding protein messenger
`RNA/ I and altered gene expression caused by genomic
`hypomethylation contribute to this potentia1.22- 24 How(cid:173)
`ever, care must be used when extrapolating from cell
`culture studies to the clinical situation. The altered cell
`metabolism and association with carcinogenesis reported
`for in vitro and animal studies result from extreme folate
`depletion that is rarely seen in the clinical setting. Addi(cid:173)
`tionally, the use of hypermethylating agents (dimethyl(cid:173)
`hydrazine) in animal tumorigenesis studies and a report
`that hypomethylation may have a suppressive effect on
`intestinal neoplasia25 further complicate the issue.
`The relation between the degree of folate repletion and
`carcinogenesis may result from very different processes
`operating at low and high concentrations of this vitamin.
`Whereas in vitro studies indicate a potential direct caus(cid:173)
`ative relation at very low folate levels, intervention and
`epidemiological studies suggest that at higher folate lev(cid:173)
`els (supraphysiological) the beneficial effects of folate
`stem from overriding mechanisms driving neoplastic
`transformation other than simply deficiency.
`No evidence has been found for the occurrence of very
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0003
`
`Figure 1. Epithelial cell isolation from endoscopic biopsy specimens.
`Biopsy specimens were removed from ethylenediaminetetraacetic
`acid solution at 45 minutes. The epithelial cell layer is free of an
`intact basement membrane. There is retraction of the lamina propria
`(original magnification 64x).
`
`using Student's t test. Paired tests were used where appropriate.
`Correlation values are given as the Pearson correlation coeffi(cid:173)
`cient (r). Data are expressed as mean value ± SEM. P values
`of <0.05 are considered significant.
`
`Results
`
`Mean cell yields (viability) from adjacent normal
`tissue or polyps or tumors were 2.98 ± 0.34 X 106
`(83%) and 2.01 ± 0.08 X 106 (77%), respectively. Mean
`cell yield from control patient biopsy specimens was 3.1
`± 0.5 X 106 (80%). Light microscopy showed that the
`biopsy epithelial basement membrane was intact after
`ion chelation treatment (Figure 1). The proportion of
`intraepitheliallymphocytes present in isolates was simi(cid:173)
`lar between tumor (4.9% ± 0.5%) and control (4.7% ±
`0.6%) groups. Lamina propria contamination of all iso(cid:173)
`lates was < 1 % (Figure 2).
`The intracellular folate content of the HT29 cells var(cid:173)
`ied with medium folate content (Figure 3).
`Recovery of folate was in the range of 94%-110%.
`The interassay and intra-assay coefficients of variation for
`the folate assay were <7%. Folate was undetectable in
`Hank's buffered salt solution supplemented with 0.3%
`bovine serum albumin. Cell viability has been shown
`previously not to influence detectable folate content. 12
`The folate content of epithelial cells isolated from normal
`mucosa was 18.5 ± 1.56 pg//-lg DNA. Cellular folate
`levels for adenomas and tumors (adjacent normal cell
`level) were 15.5 ± 2.05 pg//-lg DNA (20.18 ± 2.18 pg/
`/-lg DNA; P < 0.06) and 15.07 ± 1.06 pg//-lg DNA
`(18.03 ± 1.10 pg//-lg DNA; P < 0.02), respectively
`(Figure 4). The differences between adjacent epithelial
`cell folate levels for adenomas or tumors and normal
`
`

`
`1166 MEENAN ET AL.
`
`GASTROENTEROLOGY Vol. 112, No. 4
`
`A
`
`B
`CD3+
`
`CD19+
`
`CDl9+
`
`10'
`
`"
`Figure 2. FACScan analysis of cell isolates from colonic biopsy specimens. Cells were labeled with anti·CD3 (T cells), anti·CD19 (B cells), and
`Ber·Ep·4 (epithelial cells). (A) Control cell isolate stained with IgG1 and IgG2a antibodies followed by rabbit anti·mouse phycoerythrin and rabbit
`anti·mouse fluorescein isothiocyanate conjugates. No difference in control staining pattern was observed between epithelial and lamina propria
`compartments. (8) Epithelial compartment isolated by ion chelation showing a predominantly epithelial cell population (x axis) with some
`intraepithelial lymphocytes (y axis). (C) Epithelial compartment is stained with anti·CD19 (x axis) and anti·CD3 (y axis) to detect lamina propria
`contamination. (0) Lamina propria compartment is isolated by collagenase digestion and labeled with anti·CD3 (x axis) and anti·CD19 (y axis).
`
`1 ::I
`
`1 4
`
`low folate levels in colon mucosa. Indeed, although levels
`were lower in carcinomatous epithelium, with a down(cid:173)
`ward trend in premalignant cells, they still remained
`within the range found in control patients. The close
`correlation between DNA and protein levels suggests
`that the low level of folate in adenoma and tumor epithe(cid:173)
`lium is not artifactual and results from abnormal neoplas-
`
`- - - - -
`
`700-
`
`600-
`
`t>oo-
`
`"E 400-
`E
`u 300-
`oc
`,
`Oi:
`-5
`u 200 ,
`
`" 2 100 '
`" d " ~
`
`.00
`
`O.01mM
`
`0.1r'11\1
`
`1mM
`
`Folate supplemenl
`
`Figure 3. The folate content of HT29 cells grown in varying concentra·
`tions of folic acid. Asterisk represents 25% of actual value.
`
`tic cell DNA content. These findings may mirror the
`effect of abnormal rates of cell turnover or abnormal
`cellular enzyme activity, such as methylene tetrahydrofo(cid:173)
`late reductase. However, neither of these possibilities ne(cid:173)
`gate any potential coeffects that low folate may have on
`cellular metabolism, regardless of how the deficiency is
`induced.
`Folate availability directly influences DNA methyla(cid:173)
`tion9
`,26; however, moderate folate deficiency identified in
`this study does not result in global gene or c-myc-specific
`hypomethy lation. l l This reflects a capacity to redirect
`methyl groups in times of shortage. 27 The results from
`the present study suggest that the genomic hypomethyla(cid:173)
`tion noted in CRC does not develop from a reduced
`supply of folate, but it does not exclude impaired use.
`However, the possibility cannot be excluded that tran(cid:173)
`sient episodes of more profound folate deficiency may
`occur over the long period of time associated with the
`development of colonic mucosal lesions.
`The requirements for folate may be higher in some
`tissues (digestive mucosa and cervix) and under certain
`conditions (neoplasia) than others, leading to localized
`deficiencies, despite blood levels being within the normal
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0004
`
`

`
`April 1997
`
`EPITHELIAL FOLATE IN COLORECTAL NEOPLASIA 1167
`
`30--
`
`---
`
`i A
`
`Cl
`
`'" '" '"
`
`; .,
`
`-< z: 24-
`
`Cl
`b'" 2~~
`u
`
`1': OJ; 20,
`Eo
`I
`c
`u 18-
`i"
`C
`u 16-
`~
`::s 14-
`co
`0
`-5
`"c 1 o~
`PJ
`
`12~
`
`- - - - - - - - - -
`
`B
`
`A
`
`All
`t
`A
`All
`
`A
`
`B __
`
`- - -
`
`-
`
`- - - - -
`
`All
`
`All
`
`..
`I
`
`A
`
`All
`All
`
`A
`A
`
`...
`
`t
`All
`t
`All
`
`,
`___________ J
`
`Control
`
`Adi
`
`Polyp
`
`Control
`
`Adj
`
`Tumor
`
`Figure 4. (A) Folate content of colonic epithelial cells isolated from normal colon (control, n = 8), colonic adenomatous polyps (polyp, n = 7),
`and normal-appearing tissue (adj) adjacent to the polyp. P = 0.06. (8) Folate content of colonic epithelial cells isolated from normal colon
`(control, n = 8), colonic adenocarcinoma (tumor, n = 12), and normal-appearing tissue (adj) adjacent to the tumor. P < 0.02.
`
`range. 1
`,2H-32 Consequently, moderate systemic folate
`depletion may be a cofactor in carcinogenesis, which has
`been described for cervical dysplasia. 4 The present study
`provides evidence of reduced folate levels in neoplastic
`colonic epithelial cells in the face of normal blood and
`adjacent tissue folate levels, but it does not indicate its
`relevance to malignant pathogenesis. Folate sequestration
`at the cell surface may confound these results. However,
`this is unlikely because the major cellular repositories
`for folate are cytosol, nucleus, and mitochondria. 33 Fur(cid:173)
`thermore, gut epithelium is devoid of folate receptors. 34
`Although such receptors have been noted in vitro, on
`some transformed colonic epithelial cells lines (Caco-2),
`such receptors are functional and rapidly internalize their
`ligand.
`The lack of correlation between measures of systemic
`folate and colonic epithelial cell concentrations is similar
`to previous studies for both folate replete individuals
`and those using sulfasalazine. 12
`,35 However, perhaps more
`importantly, the similarity in results between control
`and adjacent tissues provides no evidence to suggest a
`field-folate deficiency in the colonic epithelium of those
`with or susceptible to the development of malignancy.
`They are in keeping with a previous study using mainte(cid:173)
`nance sulfasalazine, which failed to find epithelial cell
`folate depletion in patients with ulcerative colitis, a con(cid:173)
`dition associated with an increased risk for the develop(cid:173)
`ment of CRc. 35 These results do not support the prophy(cid:173)
`lactic use of folate supplements prescribed with the
`intention of correcting any presumed epithelial defi(cid:173)
`ciency. However, the low serum folate levels noted in
`the tumor group may indicate a potential for folate sup(cid:173)
`plements to overcome an undefined alteration in folate
`metabolism.
`In conclusion, neoplastic colonic epithelial cells show
`a localized relative folate deficiency. However, adjacent
`
`normal-appearing epithelium shows no such depletion
`and has similar levels to those found in nontumor-bearing
`mucosa. This result suggests that any protective effect
`derived from dietary folate is mediated through mecha(cid:173)
`nisms other than correction of cellular folate depletion.
`
`References
`
`1. Giovannucci E, Stampfer MJ, Colditz GA, Rimm EB, Trichopoulos
`D, Rosner BA, Speizer FE, Willett WC. Folate, methionine and
`alcohol intake and risk of colorectal adenoma. J Natl Cancer Inst
`1993;85:846-848.
`2. Willett WC. Micronutrients and cancer risk. Am J Clin Nutr 1994;
`59:1162s-1165s.
`3. Adams RLP. Eukaryotic DNA methyltransferases-structure and
`function. Bioessays 1995; 17:139-145.
`4. Butterworth CE, Hatch KD, Macaluso M, Cole P, Sauberlich HE,
`Soong S, Borst M, BakerVV. Folate deficiency and cervical dyspla(cid:173)
`sia. J Am Med Assoc 1992; 267:528-533.
`5. Kamei T, Kohno T, Ohwada H, Takeuchi Y, Hayashi Y, Fukuma
`S. Experimental study of the therapeutic effects of folate, vitamin
`A and vitamin B12 on squamous metaplasia of the bronchial
`epithelium. Cancer 1993; 71:2477 -2483.
`6. Cravo ML, Mason JB, Dayal Y, Hutchinson M, Smith D, Selhub
`J, Rosenberg IH. Folate deficiency enhances the development of
`colonic neoplasia in dimethylhydrazine treated rats. Cancer Res
`1992; 52:5002-5006.
`7. Lashner BA, Heidenrich PA, Su GL, Kane SV, Hanauer SB. Effect
`of folate supplementation on the incidence of dysplasia and can(cid:173)
`cer in chronic ulcerative colitis: a case control study. Gastroenter(cid:173)
`ology 1989; 97:255-259.
`8. Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC,
`Leppert M, Nakamura Y, White R, Smits AMM, Bos JL. Genetic
`alterations during colorectal tumor development. N Engl J Med
`1988; 319:525-532.
`9. Cravo M, Fidalgo P, Pereira AD, Gouveia-Oliveira A, Nobre Leitao
`C, Costa Mira F. Folate supplementation increases the degree
`of DNA methylation in patients with colonic neoplasms (abstr).
`Gastroenterology 1993; 104:A394.
`10. Holliday R. The inheritance of epigenetic defects. Science 1987;
`231:163-170.
`11. Kim Y, Giuliano A, Hatch KD, Schneider A, Nour MA, Dallal GE,
`Selhub J, Mason JB. Global DNA hypomethylation increases pro(cid:173)
`gressively in cervical dysplasia and carcinoma. Cancer 1994; 74:
`893-899.
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0005
`
`

`
`1168 MEENAN ET AL.
`
`GASTROENTEROLOGY Vol. 112, No. 4
`
`12. Meenan J, O'Hallinan E, Lynch S, Molloy A, McPartlan J, Scott J,
`Weir DG. The folate status of gastrointestinal epithelial cells
`is not predicted by serum and red cell folate levels in replete
`individuals. Gut 1996; 38:410-413.
`13. O'Broin S, Kelleher B. Microbiological assay on microtitre plates
`of folate in serum and red cells. J Clin Pathol 1992; 45:344-
`347.
`14. Augeron C, Laboisse CL. Emergence of permanently differenti(cid:173)
`ated cell clones in a human colonic cancer cell line in culture
`after treatment with sodium butyrate. Cancer Res 1984;44:
`3961-3969.
`15. Verdine GL. The flip side of DNA methylation. Cell 1994; 76:197-
`200.
`16. Anonymous. Dietary fiber and colorectal cancer risk. Nutr Rev
`1993; 51:147 -155.
`17. Matthews JH, Shiels S, Wickremasinhe SN. The effects of folate
`deficiency on thymidylate synthetase activity, deoxyuridine sup(cid:173)
`pression, cell size and doubling time in a cultured human myeloid
`cell line. Eur J HaematoI1990;45:43-47.
`18. Wickremasinghe SN, Fida S. Misincorporation of uracil into the
`DNA of folate and B12 deficient HL60 cells. Eur J Haematol1993;
`50:127 -132.
`19. Branda RF, Blickensderfer DB. Folate deficiency increases ge(cid:173)
`netic damage caused by alkylating agents and gamma-irradiation
`in chinese hamster ovary cells. Cancer Res 1993; 53:5401-
`5408.
`20. James SJ, Cross DR, Miller BJ. Alterations in nucleotide pools
`in rats fed diets deficient in choline, methionine and/or folic acid.
`Carcinogenesis 1992; 13:2471-2474.
`21. Hsueh CT, Dolnick BJ. Altered folate binding protein mRNA stabil(cid:173)
`ity in KB cells grown in folate deficient medium. Biochem Pharma(cid:173)
`col 1993;45:2537-2545.
`22. Dizik M, Christman JK, Wainfan E. Alterations in expression and
`methylation of specific genes in livers of rats fed a cancer promot(cid:173)
`ing methyl deficient diet. Carcinogenesis 1991; 12:1307 -1312.
`23. Zapsiek WF, Cronin GM, Lyn-Cook BD, Poirier LA. The onset of
`oncogene hypomethylation in the livers of rats fed methyl defi(cid:173)
`cient amino acid defined diets. Carcinogenesis 1992; 13:1869-
`1872.
`24. Simile MM, Pascale R, De Miglio MR, Nufris A, Daino L, Seddaiu
`MA, Gaspa L, Feo F. Correlation between S-adenosyl-L-methionine
`content and production of c-myc, c-Ha-ras and c-Ki-ras mRNA
`
`transcripts in the early stages of rat liver carcinogenesis. Cancer
`Lett 1994; 79:9-16.
`25. Laird PW, Jackson-Grusby L, Fazell A, Dickinson SL, Jung WE, Li
`E, Weinberg RA, Jaenisch R. Suppression of intestinal neoplasia
`by DNA hypomethylation. Cell 1995; 81:197 -205.
`26. Balaghi M, Wagner C. DNA methylation in folate deficiency: use
`of CpG methylase. Biochem Biophys Res Commun 1993; 193:
`1184-1190.
`27. Perry J, Chanarin I, Deacon R, Lumb M. Methylation of DNA in
`megaloblastic anaemia. J Clin Pathol 1990;43:211-212.
`28. Potischman N, Brinton LA, Laiming VA, Reeves WC, Brenes MM,
`Herrero R, Tenorio F, de Britton RC, Gaitan E. A case control
`study of serum folate levels and invasive cervical cancer. Cancer
`Res 1991; 51:4785-4789.
`29. Liu T, Soong SJ, Wilson NP, Craig CB, Cole P, Macaluso M,
`Butterworth CEo A case control study of nutritional factors and
`cervical dysplasia. Cancer Epidemiol Biomarkers Prev 1993; 2:
`525-530.
`30. Piyathilake CJ, Hine RJ, Dasanayake AP, Richards EW, Freeberg
`LE, Vaughn WH, Krumdieck CL. Effect of smoking on folate levels
`in buccal mucosal cells. Int J Cancer 1992; 52:566-569.
`31. Freudenheim JL, Graham S, Marshall JR, Haughey BP, Cholewin(cid:173)
`ski S, Wilkinson G. Folate intake and carcinogenesis of the colon
`and rectum. Int J Epidemiol 1991; 20:368-374.
`32. Lashner BA. Red blood cell folate is associated with the develop(cid:173)
`ment of dysplasia and cancer in ulcerative colitis. J Cancer Res
`Clin OncoI1993;119:549-554.
`33. McGuire JJ, Hsieh P, Coward JK, Bertino JR. Enzymatic synthesis
`of folylpolyglutamates. J Bioi Chem 1980; 255:5776-5788.
`34. Weitman SD, Lark RH, Coney LR, Fort DW, Frasca V, Zurawski
`VR, Kamen BA. Distribution of the folate receptor GP38 in normal
`and malignant cell lines and tissues. Cancer Res 1980; 52:
`3396-3401.
`35. Meenan J, Hallinan E, Lynch S, Keeling PWN, Scott J, Weir DG.
`Therapy with sulfasalazine does not cause colon mucosal folate
`deficiency in ulcerative colitis. J Inflamm Bowel Dis 1996; 2:163-
`167.
`
`Received February 1, 1996. Accepted December 17, 1996.
`Address requests for reprints to: John Meenan, M.D., Department
`of Gastroenterology, Rayne Institute, St. Thomas's Hospital, London
`SE1 7EH, England. Fax: (44) 171-922-8394.
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1113-0006