Original Paper Ophthalmic Res 2005;37:270–278 DOI: 10.1159/000087372
Received: February 25, 2005 Accepted after revision: May 17, 2005 Published online: August 9, 2005
Lactoferrin Expression by Bovine Ocular Surface Epithelia: A Primary Cell Culture Model to Study Lactoferrin Gene Promoter Activity Maria Grazia Santagati Simona La Terra Mulè Carla Amico Matteo Pistone Dario Rusciano Vincenzo Enea Research Development and Innovation (RDI), SIFI SpA, Catania, Italy
Key Words Lactoferrin mRNA Promoter Conjunctiva Cornea
Abstract Tear lactoferrin, mainly secreted by the lachrymal glands, exerts a protective effect on the ocular surface, and an abnormal decrease of its production may lead to an increased risk of infection and pathological alterations of ocular surface epithelia. In this study we analyzed whether corneal and conjunctival epithelia could be an additional source of tear lactoferrin, and whether conjunctival epithelial cells in culture could be a suitable model system to address regulation of lactoferrin gene expression. Real-time PCR and Western immunoblotting showed that in bovines lactoferrin is indeed produced by these epithelia, and that the human lactoferrin promoter can direct the expression of a CAT reporter gene, thus indicating that these cells are a true source of lactoferrin, and may be used in vitro to study the regulation of lactoferrin expression. Copyright © 2005 S. Karger AG, Basel
Introduction
Lactoferrin (LF), a metal-binding glycoprotein closely related to the plasma iron transport protein transferrin, is synthesized by neutrophils and glandular epithelial
© 2005 S. Karger AG, Basel 0030–3747/05/0375–0270$22.00/0 Fax +41 61 306 12 34 E-Mail
[email protected] www.karger.com
Accessible online at: www.karger.com/ore
cells and thus is found in several exocrine secretions such as milk, tears and saliva. Due to its antimicrobial and anti-inflammatory activities in vitro, LF is thought to be involved in host defense against infection and severe inflammation, most notably at mucosal surfaces, through a mechanism mainly based on iron subtraction from the environment [1, 2]. Tear LF is synthesized in the lachrymal gland [3] and accounts for approximately 25% of total tear proteins [4]. In the eye, LF has been shown to contribute to the maintenance of tear film stability [5] and to protect the ocular epithelial surface from exogenous insults [6, 7]. Dry eye is a common disorder of the tear film that may result from decreased tear production, excessive tear evaporation, or abnormality in the protein or lipid components of the tear film [8]. The concentration of LF was found to be markedly lower in tears of patients affected by dry eye of different origins [9]. The level of tear LF in dry eye patients inversely correlated with the Rose Bengal staining score, an indicator of the severity of ocular surface damage in these patients [10], thus pointing to a possible protective role of LF. A similar indication came from a study conducted on a short-term dry-eye rabbit model, in which the extent of ocular surface damage was much reduced in eyes treated with LF solutions [11]. Therefore, it is of medical interest to understand the regulatory mechanisms governing the production of LF by epithelial cells, and in particular it is an interest of ophthalmology to see whether ocular surface epithe-
Dario Rusciano Research Development and Innovation, SIFI SpA Via Ercole Patti, 36 IT–95020 Lavinaio Catania (Italy) Tel. +39 0957922240, Fax +39 0957922414, E-Mail dario_rusciano@sifi.it
Table 1. Nucleotide sequence of primers and probe used for DNA amplification
Primers’ name
Primers’ sequence
Applications
LF-UNIF RUNIFAR
Sense Antisense
5-GGCTGGAAYATCCCCATGGGCCTG-3 5-CAGACACTCAGTGTTGTCATTG-3
RT-PCR
UNI-ACTIF UNI-ACTR
Sense Antisense
5-CATGTGCAAGGCCGGCTT-3 5-GTAGAAGGTGTGGTGCCAGA-3
RT-PCR
FL-FRT RL-FRT LF-PROBE
Sense Antisense Probe
5-CCTGGCAGAGAACCGGAAA-3 5-ACGGCA-AGGTACCCTTCCGTT-3 5-CCTCCAAACACAGTAGCCTAGATTGTGT-GCTGAGAC-3
Real-time PCR
LF-Fw LF-Rv
Sense Antisense
5-CGAGGATCATGGCTCACTG-3 5-AAACGAAGCCTCTGCCACT-3
PCR
Primers and probe sequence specificity is determined by BLAST Similarity Search provided by the National Center for Biotechnology Information and available at http://www.ncbi.nhm.nih.gov/blast.
lial cells have the potential to contribute to the tear LF content. Gene expression in eukaryotic cells is governed by a complex interaction between the cis-acting regulatory elements of the promoter region and specific transcription factors. Therefore, identification and characterization of each component represent the first step towards an understanding of such a complex mechanism. Recently, these types of studies have been conducted in primary cultures, which, being more similar to their in vivo counterpart, are to be preferred to immortalized cell lines [12, 13]. However, wide employment of primary cells is still limited because of the many technical difficulties encountered in growing and transfecting them in vitro, beside the difficulty in obtaining adequate biopsy specimens from normal donors, especially in humans. In our laboratory we have recently established a method for culturing bovine conjunctival epithelial cells (BCEC) that, as shown by morphological examination and specific marker detection, maintain structural and functional features of the original tissue [14, 15]. Thus, the purpose of this study was to investigate LF gene expression in conjunctival and corneal epithelial tissues as well as in BCEC. Moreover, in order to address the regulation of human LF promoter activity in conjunctival epithelial cells, we also developed a transient transfection system in primary cultures and tested a construct containing the CAT reporter gene under the control of the human LF promoter (LF/CAT).
Lactoferrin Expression in Ocular Tissues
Materials and Methods Tissue Collection and Cell Culture Corneal and conjunctival epithelial tissues and their derived primary cell cultures were prepared from bovine eyes provided by a local slaughterhouse, as previously described [14, 15]. Primary BCEC were either grown as monolayers on collagen type I precoated plastic dishes, or as multilayers on Transwell® culture chamber inserts (Corning, Milan, Italy) at 37 ° C and 5% CO2. Cells, when grown as multilayers, were used for RNA extraction when their transepithelial electrical resistance was calculated to be at plateau values, indicating the best physiological state of the culture [14, 15]. Sirc (from rabbit cornea, epithelial/fibroblastic mixed morphology) and clone 1-5c-4 (human epithelial cells, originally derived from conjunctiva) continuous cell lines (ATCC, Manassas, Va., USA) were cultured according to ATCC instructions. RNA Isolation and RT-PCR Total RNA was isolated from bovine lachrymal glands, epithelial tissues (pooling tissue fragments obtained from 5 eyes) and cell cultures (cells from 4 different transwells were pooled and extracted) following the method of Chomczynski and Sacchi [16]. One microgram of total RNA was then reverse transcribed with the SuperScript II RNaseH-Reverse Transcriptase (Invitrogen, Milan, Italy) and with either RUNIFAR (for LF cDNA) or UNI-ACTR (for -actin cDNA) specific reverse primers (table 1) according to the protocol of Invitrogen. Next, 2 l of each specific cDNA were amplified with the Advantage® 2 PCR polymerase mix (Clontech, Florence, Italy) and with either LF-UNIF and RUNIFAR, or the UNI-ACTF and UNIACTR couple of primers (table 1) as described [17] and following the manufacturer’s protocol. All primers were synthesized by MWG Biotech (Germany). Real-Time PCR For real-time PCR amplification, total RNA was reverse transcribed as described above, except that random hexamers (Invitro-
Ophthalmic Res 2005;37:270–278
271
gen) were used. Bovine-specific primers and probe, selected from an area containing an intronic segment, were designed with the assistance of Primer Express software (Applied Biosystems, Foster City, Calif., USA), or were purchased directly from Applied Biosystems (18S rRNA). The sequence specificity of the chosen oligonucleotides was confirmed by BLAST Similarity Search and by PCR assay. One microliter of cDNA was amplified with the ABIPrism7000 sequence detection system (Applied Biosystems) by using the nonfluorescent primers FL-FRT and RL-FRT, and the TaqMan LFPROBE (table 1) labeled at the 5-end with FAM (6-carboxy-fluorescein) and at the 3-end with TAMRA (6-carboxy-tetramethylrhodamine). Amplification reaction was performed in 50 l total volume containing 900 nM of each primer, 250 nM TaqMan fluorescent probe and 0.5! PCR TaqMan Universal PCR Master Mix (Applied Biosystems). The thermal cycling conditions were 95 ° C for 10 min, followed by 50 cycles at 95 ° C for 15 s, and 61 ° C for 1 min. To validate the use of these primers and probe for relative quantitation of mRNA, the efficiency of the LF gene amplification was compared with that of the 18S rRNA. The amount of specific mRNA was determined by using the comparative threshold cycle method (CT) [18]. The quantity of LF mRNA was normalized to the endogenous control 18S rRNA by subtracting the CT of 18S rRNA from that of the target messenger (CT). Each value reported is the mean of three independent experiments, and each, in turn, is performed in quadruplicate. Blood was selected as a tissue calibrator, and its LF value was set arbitrarily as 1. Immunoprecipitation and Immunoblotting Tissues and Cells. Lachrymal gland, corneal and conjunctival bovine epithelium tissues were homogenized at 4 ° C with PBS and protease inhibitors, and extracted with an equal volume of 2! concentrated buffer (10 mM Tris-HCl, pH 7.2, 1.5 M KCl, 0.5% Triton X-100, 5 mM EDTA, 25 g/ml aprotinin, 25 g/ml leupeptin, 1 mM AEBSF) for 20 min on ice. BCEC grown as monolayers for 3 days were directly extracted with the same buffer. Moreover, the residual insoluble material was further extracted with a urea buffer (40 mM Tris-HCl, pH 7.2, 8 mM urea, 5% -mercaptoethanol) for 20 min on ice. After centrifugation for 5 min at 14,000 rpm, 4 ° C, each supernatant was analyzed for the determination of the total protein quantity by the Bradford method. Conditioned Media. BCEC were grown to confluence with serum-containing medium and supplements. At confluence, monolayers were rinsed twice with PBS and incubated 4 h with serumfree medium. The medium was discarded, and fresh, serum-free medium was added for a further 24 or 48 h. This conditioned medium was collected, filtered through 0.22-m filters, and the protein content determined by the Bradford method. Biotin Labeling. Four hundred micrograms of proteins from tissues and 200 g of proteins from conditioned media or cell extract were incubated with biotin (250 g/mg of protein) for 4 h at 4 ° C. Biotin-treated samples were dialyzed for 3 days against PBS with two changes per day. Each sample was then sequentially incubated with affinity-purified rabbit anti-human LF polyclonal antibodies (Biodesign; overnight at 4 ° C with gentle agitation) and protein A/Sepharose (1 h at 4 ° C). After three rinses with PBS (16,000 g, 6 min, 4 ° C), samples were boiled for 2 min with reducing Laemmli’s sample buffer, loaded on a 7.5% SDS-PAGE and run on a Mini-PROTEAN® II (Bio-Rad, Milan, Italy). The gel was
272
Ophthalmic Res 2005;37:270–278
transferred (Sammy-Dry®, Schleicher & Schuell, Dassel, Germany) onto a PVDF membrane (Immobilon, Millipore, Bedford, Mass., USA). After the blotting, the membrane was incubated for 1 h in a blocking solution (0.5% Tween 20, 1% fetal calf serum, 1% bovine albumin serum, 1% Triton X-100, 100 mM MgCl2, 100 mM TrisHCl, pH 7.4), and avidin-peroxidase added for 30 min. After thorough rinsing with PBS, the filter was incubated with ECL™ (Amersham Pharmacia Biotech) and exposed to hyperfilm (Kodak). PCR Amplification and Cloning of the LF Promoter The human LF promoter gene was amplified from human blood (obtained in accordance with the principles of the Declaration of Helsinki) by PCR using LF-Fw and LF-Rv primers (table 1). The PCR product of 1,306 bp, which spans the whole sequence of the known LF promoter (locus S52659), was subcloned into pCRTOPO-blunt II vector (Invitrogen) and sequenced. The LF promoter was then excised with XhoI/HindIII enzymes (Promega, Milan, Italy) and ligated into similarly digested pCAT control vector (Promega) in place of the original SV40 promoter. To obtain the final LF recombinant vector (LF/CAT), the HindIII intron fragment, previously excised from the control vector, was then placed at the correct position and orientation as further confirmed by enzymatic digestion. Transient Transfections and CAT Assay BCEC primary cultures as well as Sirc and clone 1-5c-4 cell lines were transiently transfected by using the polycationic detergent Lipofectamine-Plus (Invitrogen) following the manufacturer’s recommendations. Semiconfluent monolayers were transfected with 1 g of LF/CAT recombinant vector (described above) or with 1 g of either control (SV40/CAT) or basic (0/CAT) pCAT vectors (Promega), used as positive and negative controls, respectively. After transfection, cells were incubated at 37 ° C in 5% CO2 for 48 h, then washed 3 times with PBS, and CAT activity was measured by using a Cat ELISA Kit (Boehringer-Mannheim, Milan, Italy). Transfection experiments were repeated at least 3 times, each time with a new preparation of cultured cells.
Results
Expression of LF mRNA in Corneal and Conjunctival Epithelial Tissues and in Derived Cell Culture Conventional RT-PCR was used to examine whether LF is expressed in corneal and conjunctival epithelial tissues as well as in derived cell cultures (BCEC). The homogeneity of BCEC was verified by fluorescent immunostaining with cytokeratin (as the marker of epithelial cells) and vimentin (as the marker of stromal cells) antibodies. Epithelial cultures were all positive for cytokeratin and negative for vimentin staining, thus indicating that the cultures were free of fibroblasts (data not shown). LF gene expression was also studied in the continuous cell lines Sirc and clone 1-5c-4, respectively derived from rabbit corneal fibroblasts and human conjunctival epithelium.
Santagati/La Terra Mulè/Amico/Pistone/ Rusciano/Enea
Fig. 1. Total RNA from lachrymal gland (LG), corneal epithelium (Cor), conjunctival epithelium (Conj), BCEC cultivated under either air-interface (AIC) or liquid-covered culture (LCC) conditions, Sirc and clone 1-5c-4 established cell lines was reverse transcribed and PCR amplified with either LF or -actin-specific primers. PCR products were electrophoretically separated on a 1% agarose gel. Positions of both LF (552 bp) and -actin (229 bp) fragments are indicated along with those of the closest molecular weight (MW) markers.
MW
LG
Cor
Conj
AIC
LCC
Sirc
Clone
H 2O
500 bp >
< LF (552 bp)
200 bp >
< β-Actin (229 bp)
1,000 260.29
by the real-time comparative CT method. The quantity of LF mRNA was normalized to the endogenous control 18S rRNA by subtracting the CT of 18S rRNA from that of the target messenger (CT). Blood was selected as a tissue calibrator and its normalized LF value (LFN) was arbitrarily set as 1. a Relative amounts of LFN mRNA in different tissues and cells. b Relative amounts of LFN mRNA in BCEC grown as monolayers for several passages on collagen I-coated dishes. The scale of the y-axes in both graphics is logarithmic. LG = Lachrymal gland; Conj = conjunctival epithelium; Cor = corneal epithelium; AIC = air-interface multilayer conjunctival epithelial cell cultures; LCC = liquid-covered multilayer conjunctival epithelial cell cultures. Asterisk indicates that the RNA sample was derived from cells cultivated in a duplicate well.
10
1
1.00 0.31
0.1
0.04
0.03
AIC
LCC
0.01 0.01
a
0.001
Blood
Conj
LG
Cor
10
LFN relative to blood
Fig. 2. Quantitation of LF transcripts made
LFN relative to blood
100
b
1
1.000
0.874
0.572 0.102
0.1 0.013 0.01
0.008 0.003
0.001 0.0001
Blood
P0
P0✽
P2
P3
P4
P4✽
LF RT-PCR products were resolved in agarose gels and visualized by ethidium bromide staining. Figure 1 shows that LF mRNA was detected in great abundance in the lachrymal gland (as expected), but also in corneal and conjunctival tissues and in their derived cell cultures. A semiquantitative evaluation of gel bands suggests a higher expression in conjunctival than in cor-
neal epithelium, comparable expression levels in BCEC cultured either as multilayers under air-lift conditions or as submerged cultures, very low levels in the human conjunctival continuous cell line clone, and undetectable levels in the rabbit corneal cell line Sirc, most likely due to the fact that rabbit tears contain transferrin instead of LF.
Lactoferrin Expression in Ocular Tissues
Ophthalmic Res 2005;37:270–278
273
LF
GL
CN
dependent measurements at P4 confirmed the lowest levels, 2 logs below P0.
CR
97 > 84 > 66 >
a 24 h
48 h
BSA
LF
Det. Urea
97 > 84 >
< LF
66 >
b
c
Fig. 3. Total proteins extracted from the indicated bovine tissues (LG = lachrymal gland; CN = conjunctiva; CR = cornea) (a), from
cell culture media conditioned 24 and 48 h by conjunctival cells in culture (b) and from bovine conjunctival cells grown in tissue culture for 3 days [extracted with detergent (Det.) and then with urea] (c) were biotin-labeled and immunoprecipitated with anti-LF antibodies. Each immunoprecipitate was recovered in 30 l of reducing Laemmli’s sample buffer and loaded on a 7.5% SDS-PAGE, blotted to a PVDF membrane, and revealed with avidin-peroxidase. A 1,000! dilution of the lachrymal gland sample (a) was loaded to allow a good resolution of the LF protein, which thus appears as a fainter band. Molecular weight markers are indicated on the left, and a standard of LF is also run as a reference.
Quantitation of LF transcripts in native tissues and cell cultures was done by real-time PCR, and the obtained values graphed on a logarithmic scale (fig. 2). In conjunctival epithelium, LF transcripts were approximately 800fold lower than in the lachrymal gland (fig. 2a), whereas in BCEC grown as multilayers either under air-lift or submerged conditions LF mRNA was a further 7- to 10-fold lower than in native epithelium. Corneal epithelium expressed the lowest amounts of LF mRNA, 30 times less than conjunctival epithelium, and 2.4 ! 104-fold lower than the lachrymal gland. Next, we analyzed the persistence of LF mRNA expression in BCEC upon repeated subculturing. BCEC were cultured as monolayers onto collagen I-coated dishes and the cultures were monitored for four passages, from P0 (primary culture) until P4 (4th passage) (fig. 2b). At P0, in two parallel cultures, we observed LF mRNA amounts which were even higher than those detected in the corresponding epithelium (fig. 2a, b), and much higher than those observed in BCEC cultured in transwells. However, a consistent trend towards decreased expression was evident with further subculturing, and two in-
274
Ophthalmic Res 2005;37:270–278
LF Protein Expression in Ocular Epithelial Tissues and Cells Protein analysis by specific immunoprecipitation with affinity-purified anti-LF polyclonal antibodies and Western blotting was performed to investigate whether LF mRNA transcripts were translated into a detectable amount of protein. Figure 3a shows that, consistent with transcription data, LF protein was expressed at higher levels in conjunctival than in corneal epithelial tissue. LF was also detected in BCEC cultured as monolayers on collagen-coated plates. The protein was evident both in cell extracts (fig. 3c) and conditioned medium (fig. 3b), thus indicating that it is endogenously produced and secreted. LF Promoter Activity in Conjunctival and Corneal Epithelial Cells In order to see whether BCEC would be permissive for human LF gene expression, primary BCEC at various passages were transiently transfected with an LF/CAT recombinant vector. Parallel cultures were transfected with positive and negative control vectors under the same conditions. The corneal and conjunctival cell lines Sirc and clone 1-5c-4 were also transfected, and used to control the transfection efficiency of the different constructs. Both the SV40/CAT-positive control vector, and the LF/ CAT vector were expressed with similar efficiency in these cell lines (fig. 4a). Consistent with mRNA data (fig. 1, 2) Sirc were somewhat less efficient than clone in expressing the LF/CAT vector. No activity was observed upon transfection of either cell line with the promoterless vector, implying that there was no spurious CAT expression. Similarly to established cell lines, BCEC up to 4 passages in vitro (P0–P4) also allowed expression of the transfected reporter gene (fig. 4b). The transfection efficiency measured on the expression of the CAT gene under the control of the SV40 promoter (right side of fig. 4b), although variable, remained on average close to the values observed for continuous cell lines (compare bars on the right in fig. 4a and b). Intriguingly, we found that expression of the CAT reporter gene under the control of LF promoter (fig. 4b, left side) was very low at P0, peaked at P2, being 6-fold higher than the initial value, and decreased at P4 to values similar to those seen at P0.
Santagati/La Terra Mulè/Amico/Pistone/ Rusciano/Enea
1.6
LF/CAT
SV40/CAT
Protein (ng/µg)
1.2
0.8
0.4
a
0
Sirc
1.6
Sirc
LF/CAT
Clone SV40/CAT
1.2 Protein (ng/µg)
Fig. 4. Levels of CAT expression in cell lines and primary cells transfected with either the SV40/CAT (right side) or the LF/ CAT (left side) vectors. CAT activity measured in each well of transfected cells was determined 48 h after transfection, and transformed into ng of CAT per g of total proteins. a Sirc and clone 1-5c-4 cell lines. b BCEC transfected at consecutive in vitro passages: P0, P2, P4. Each bar represents the mean (8 SD) of the CAT activity obtained from three to five independent experiments with different cell preparations. * p = 0.02 (P0 vs. P2); + p = 0.07 (P2 vs. P4).
Clone
+ 0.8
0.4
0 P0
b
P2
P4
P0
P2
P4
BCEC
In the present study we have shown that the lachrymal gland is not the only source of LF in the eye (although it remains the major source), but ocular surface epithelial cells also produce detectable amounts of LF. The expression at both the protein (fig. 3) and the mRNA (fig. 1, 2) levels appears to be much higher in conjunctival than in corneal epithelial tissue, which only expresses tiny amounts of the protein. LF expression is also retained in conjunctival epithelial cells when they are cultured in vitro, even though the precise values appear to depend both on culture conditions (much higher in cells grown as monolayers on coated plastic than in cells grown as multilayer on transwell filters; fig. 2a, b) and timing; the expression appears to be upregulated as soon as the cells are explanted in tissue culture and tends to decrease in further passages (fig. 2b). What factors regulate the expres-
sion in the different ocular tissues cannot yet be established from this study, which only addresses the presence of LF in ocular surface epithelial cells, and their suitability as a model system to study the regulation of LF gene expression. However, it is tempting to speculate that since conjunctiva, contrary to cornea, is a tissue rich in blood vessels, some serum nutrients, such as growth factors and vitamin A, may permeate the tissue and activate specific cellular receptors, ultimately modulating transcription of the LF gene above basal levels. In this respect, it has been shown that retinoic acid, the biologically active form of vitamin A, stimulates LF expression in murine embryonic stem cells [19], and that human conjunctival epithelial cells express functional retinoic acid receptors [20]. Our study is the first to show that LF, similarly to other antimicrobial molecules such as the defensins [21, 22], is produced by ocular surface epithelial cells, which also represent an integral part of the mucosal immune system.
Lactoferrin Expression in Ocular Tissues
Ophthalmic Res 2005;37:270–278
Discussion
275
A previous work by Gillette and Allansmith [23] using immunofluorescence detection by specific antibodies failed to detect any LF in human ocular tissues other than the lachrymal gland. However, since immunofluorescence is a much less sensitive technique than PCR or protein detection by immunoprecipitation and biotin labeling, it is likely that LF expression levels in conjunctival and corneal epithelial tissues were below the detection limit of that technique. Moreover, it cannot be ruled out that human ocular surface tissues may also express different (lower) amounts of LF than bovine tissues. For instance, it has been reported that LF mRNA is highly expressed in the uterus and spleen of both mice and humans but not in porcine counterparts [24], and that multiple steroid response elements are present in human and mouse, but not in bovine or porcine promoters [25]. Western immunoprecipitation analyses shown in figure 3 hint at the nature and possible roles of LF produced by BCEC. The presence of the protein in ocular tissues is demonstrated in figure 3a. The apparent molecular weight of lachrymal gland LF was slightly lower as compared to standard or conjunctival LF, and it might be due to differences in glycosylation and/or iron saturation. LF appears to be constantly secreted in the medium, where it accumulates, since more protein is detected in 48-hour than in 24-hour conditioned medium (fig. 3b). This increase is consistent with an endogenous production of LF, ruling out that what we observed is residual protein from FCS added at the time of cell plating. Secreted LF may carry out several different activities, as it has been shown that, beside binding free iron, a mechanism by which it can exert both antimicrobial and antioxidant activities [6, 26], it has immunomodulatory and anti-inflammatory properties [27], and multiple enzymatic activities [28], including a prominent DNase activity [29]. Its interactions with the lipid layer of the tear film might contribute to its stabilization [30]. Moreover, LF receptors can be expressed on the cell surface [31, 32], where they may mediate some form of signal transduction [33]. However, detergent and urea cell extracts reveal that part of the protein comes from inside the cell, with more protein extracted by urea (fig. 3c). This indicates that some of the protein is not present in the cytoplasmic membranous system (extractable by detergents), but associated with the detergent-insoluble portion (e.g. nucleus and/or cytoskeleton) of the cell, extractable with urea. The presence of LF in the nucleus is consistent with its DNA-binding ability and a proposed role as a regulator of transcription [34–38]. The two lower-molecular-weight
276
Ophthalmic Res 2005;37:270–278
bands visible in the blot could represent proteins that are associated with LF in the insoluble fraction, and coimmunoprecipitate with it. Understanding the exact molecular mechanisms governing and regulating LF gene expression in conjunctival epithelium might be of importance in all those pathological conditions, such as Sjögren’s and dry eye syndromes, in which a reduced level of lachrymal LF protein has been detected [9]. In these cases, an increased synthesis and secretion of LF by the epithelium might contribute to the restoration of the impaired stability of the lachrymal film, and to a better defense of the diseased ocular surface from microbial infections. In order to address tissue-specific LF promoter regulation, we established an in vitro primary cell culture system of BCEC. In fact, it is intuitive that regulation of gene expression is usually much closer to the in vivo situation in primary cell cultures than in established cell lines, although primary cells do not always represent an easy choice, given the difficulties encountered in their culture and transfection. However, differently from what is often observed, BCEC cultivated in monolayer turned out to be a valid model for LF promoter study, given that they could be easily subcultured for several passages, and transfected, using a commercial transfection reagent, with an efficiency comparable to that observed for established cell lines (fig. 4a). The observed kinetics of promoter activity, peaking at the second passage (P2) and decreasing to basal levels at P4, is certainly intriguing. In our earlier work on conjunctival epithelial cell cultures [14, 15], we showed that, under air-lift culture conditions and without subculturing, their phenotype remains quite stable for many days [15]. Conjunctival cells grown on transwells filters under submerged conditions tend to degenerate faster [14]. It might then be possible that some degree of cell instability due to the culturing conditions has led to this peculiar pattern of promoter activity, which, however, is reproducible. It is important to stress that LF promoter activity is hardly an in vitro artifact following displacement of conjunctival cells from their original environment, as there is a good agreement between data obtained from living tissues and from cell cultures. Since ours is the first report of LF expression in conjunctival epithelial cells, no transcription factors have been described yet that regulate LF gene expression in that environment. We may speculate that, based on the sequence analysis of the bovine LF promoter [39], showing the presence of Sp1 binding sites, and on the ubiquitous distribution of Sp1 [40], this transcription factor
Santagati/La Terra Mulè/Amico/Pistone/ Rusciano/Enea
could be easily involved in the regulation of LF expression also in conjunctival epithelial cells. Moreover, since both human and mouse LF gene promoters contain estrogen response elements [41, 42], LF expression is decreased in dry eye syndrome [9], and hormone imbalance is a likely cause of dry eye in postmenopausal women [43]; it appears likely that estrogen could be involved in LF regulation. Studies are currently under way to address these possibilities. The different trends in LF expression in tissue culture revealed by real-time PCR (fig. 2b) and promoter activity (fig. 4b) may be explained by the fact that we have used the human LF gene promoter sequence to drive reporter gene expression in bovine conjunctival cells. This human promoter was derived from human blood, and, as we determined by sequence analysis, shares 99.7% nucleotide homology with the two published human LF sequences (referred to as M_73700 and S_52659 at the National Center for Biotechnology Information, Bethesda, Md., USA; data not shown). This means that, according to the study published by Teng et al. [44], the cloned LF promoter bears multiple response elements that could be recognized by different transcription factors. The mix of these factors may slightly vary between human and bovine cells, and also during cell passages, thus leading to different levels of LF expression [45, 46]. Real-time PCR, on the other hand, reflects the amount of species-specific endogenous mRNA present at that moment within the cell, and thus represents a more accurate picture of the
cell transcriptional activity balanced by mRNA stability. In fact, different stabilities of the two mRNAs (LF or CAT) might also have an influence on the amounts finally detected by the two methodologies. Further studies are in progress to confirm these findings in human tissues and cells, and to determine which transcription factors could be regulating human LF gene expression in the human homologous conjunctival cell system. In conclusion, we have presented here novel findings indicating that bovine ocular surface epithelial cells produce LF, and illustrated a suitable model system based on primary cell cultures to address the regulation of LF gene expression in ocular surface tissues. These results may lead to the discovery of regulatory agents that might be topically employed to stimulate endogenous LF production in patients affected by dry eye or Sjögren’s syndromes, and thus ameliorate their pathological condition. There is, in fact, preliminary evidence that LF supplementation in severe dry eye patients is associated with an improvement of dry eye symptoms, tear stability and vital staining scores [47].
Acknowledgment The authors wish to thank Dr. Asero from SIFI RSI for critically reading the manuscript and for helpful discussions.
References 1 Lonnerdal B, Iyer S: Lactoferrin: molecular structure and biological function. Annu Rev Nutr 1995;15:93–110. 2 Nuijens JH, van Berkel PH, Schanbacher FL: Structure and biological actions of lactoferrin. J Mammary Gland Biol Neoplasia 1996; 1: 285–295. 3 Janssen PT, van Bijsterveld OP: Origin and biosynthesis of human tear fluid proteins. Invest Ophthalmol Vis Sci 1983;24:623–630. 4 Kijlstra A, Jeurissen SH, Koning KM: Lactoferrin levels in normal human tears. Br J Ophthalmol 1983;67:199–202. 5 Miano F, Mazzone M, Giannetto A, et al: Interface properties of simplified tear-like fluids in relation to lipid and aqueous layers composition. Adv Exp Med Biol 2002;506:405–417. 6 Kuizenga A, van Haeringen NJ, Kijlstra A: Inhibition of hydroxyl radical formation by human tears. Invest Ophthalmol Vis Sci 1987;28: 305–313.
Lactoferrin Expression in Ocular Tissues
7 Shimmura S, Suematsu M, Shimoyama M, Tsubota K, Oguchi Y, Ishimura Y: Subthreshold UV radiation-induced peroxide formation in cultured corneal epithelial cells: the protective effects of lactoferrin. Exp Eye Res 1996; 63:519–526. 8 Johnson ME, Murphy PJ: Changes in the tear film and ocular surface from dry eye syndrome. Prog Retin Eye Res 2004;23:449–474. 9 Ohashi Y, Ishida R, Kojima T, et al: Abnormal protein profiles in tears with dry eye syndrome. Am J Ophthalmol 2003;136:291–299. 10 Danjo Y, Lee M, Horimoto K, Hamano T: Ocular surface damage and tear lactoferrin in dry eye syndrome. Acta Ophthalmol (Copenh) 1994;72:433–437. 11 Fujihara T, Nagano T, Nakamura M, Shirasawa E: Lactoferrin suppresses loss of corneal epithelial integrity in a rabbit short-term dry eye model. J Ocul Pharmacol Ther 1998; 14: 99–107.
12 Omstead DR, Baird LG, Christenson L, et al: Voluntary guidance for the development of tissue-engineered products. Tissue Eng 1998; 4: 239–266. 13 Germain L, Carrier P, Auger FA, Salesse C, Guerin SL: Can we produce a human corneal equivalent by tissue engineering? Prog Retin Eye Res 2000;19:497–527. 14 Civiale C, Paladino G, Marino C, Trombetta F, Pulvirenti T, Enea V: Multilayer primary epithelial cell culture from bovine conjunctiva as a model for in vitro toxicity tests. Ophthalmic Res 2003;35:126–136. 15 Paladino G, Marino C, La Terra Mule S, Civiale C, Rusciano D, Enea V: Cytokeratin expression in primary epithelial cell culture from bovine conjunctiva. Tissue Cell 2004;36:323– 332. 16 Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156–159.
Ophthalmic Res 2005;37:270–278
277
17 Ault GS, Ryschkewitsch CF, Stoner GL: Typespecific amplification of viral DNA using touchdown and hot start PCR. J Virol Methods 1994;46:145–156. 18 User Bulletin No. 2 ABI PRISM 7700 Sequence Detection System, 1997, 11. 19 Geng K, Li Y, Bezault J, Furmanski P: Induction of lactoferrin expression in murine ES cells by retinoic acid and estrogen. Exp Cell Res 1998;245:214–220. 20 Hori Y, Spurr-Michaud S, Russo CL, Argueso P, Gipson IK: Differential regulation of membrane-associated mucins in the human ocular surface epithelium. Invest Ophthalmol Vis Sci 2004;45:114–122. 21 Haynes RJ, Tighe PJ, Dua HS: Antimicrobial defensin peptides of the human ocular surface. Br J Ophthalmol 1999;83:737–741. 22 Lehmann OJ, Hussain IR, Watt PJ: Investigation of beta defensin gene expression in the ocular anterior segment by semiquantitative RT-PCR. Br J Ophthalmol 2000;84:523–526. 23 Gillette TE, Allansmith MR: Lactoferrin in human ocular tissues. Am J Ophthalmol 1980;90: 30–37. 24 Teng CT: Regulation of lactoferrin gene expression by estrogen and epidermal growth factor: molecular mechanism. Cell Biochem Biophys 1999;31:49–64. 25 Teng CT: Lactoferrin gene expression and regulation: an overview. Biochem Cell Biol 2002; 80:7–16. 26 Ward PP, Conneely OM: Lactoferrin: role in iron homeostasis and host defense against microbial infection. Biometals 2004; 17: 203– 208. 27 Legrand D, Elass E, Pierce A, Mazurier J: Lactoferrin and host defence: an overview of its immuno-modulating and anti-inflammatory properties. Biometals 2004;17:225–229.
278
28 Kanyshkova TG, Babina SE, Semenov DV, et al: Multiple enzymic activities of human milk lactoferrin. Eur J Biochem 2003; 270: 3353– 3361. 29 Babina SE, Kanyshkova TG, Buneva VN, Nevinsky GA: Lactoferrin is the major deoxyribonuclease of human milk. Biochemistry (Mosc) 2004;69:1006–1015. 30 Millar TJ, Tragoulias ST, Anderton PJ: The contribution of individual tear components to the surface pressure of tears. Ocul Surf 2005; 3:92. 31 Suzuki YA, Shin K, Lonnerdal B: Molecular cloning and functional expression of a human intestinal lactoferrin receptor. Biochemistry 2001;40:15771–15779. 32 Suzuki YA, Lonnerdal B: Characterization of mammalian receptors for lactoferrin. Biochem Cell Biol 2002;80:75–80. 33 Takayama Y, Takahashi H, Mizumachi K, Takezawa T: Low density lipoprotein receptorrelated protein (LRP) is required for lactoferrin-enhanced collagen gel contractile activity of human fibroblasts. J Biol Chem 2003; 278: 22112–22118. 34 Garre C, Bianchi-Scarra G, Sirito M, Musso M, Ravazzolo R: Lactoferrin binding sites and nuclear localization in K562(S) cells. J Cell Physiol 1992;153:477–482. 35 He J, Furmanski P: Sequence specificity and transcriptional activation in the binding of lactoferrin to DNA. Nature 1995;373:721–724. 36 Kanyshkova TG, Semenov DV, Buneva VN, Nevinsky GA: Human milk lactoferrin binds two DNA molecules with different affinities. FEBS Lett 1999;451:235–237. 37 Penco S, Scarfi S, Giovine M, et al: Identification of an import signal for, and the nuclear localization of, human lactoferrin. Biotechnol Appl Biochem 2001;34:151–159. 38 Son KN, Park J, Chung CK, et al: Human lactoferrin activates transcription of IL-1beta gene in mammalian cells. Biochem Biophys Res Commun 2002;290:236–241.
Ophthalmic Res 2005;37:270–278
39 Seyfert HM, Tuckoricz A, Interthal H, Koczan D, Hobom G: Structure of the bovine lactoferrin-encoding gene and its promoter. Gene 1994;143:265–269. 40 Kaczynski J, Cook T, Urrutia R: Sp1- and Kruppel-like transcription factors. Genome Biol 2003;4:206. 41 Teng CT, Gladwell W, Beard C, Walmer D, Teng CS, Brenner R: Lactoferrin gene expression is estrogen responsive in human and rhesus monkey endometrium. Mol Hum Reprod 2002;8:58–67. 42 Stokes K, Alston-Mills B, Teng C: Estrogen response element and the promoter context of the human and mouse lactoferrin genes influence estrogen receptor alpha-mediated transactivation activity in mammary gland cells. J Mol Endocrinol 2004;33:315–334. 43 Smith JA, Vitale S, Reed GF, et al: Dry eye signs and symptoms in women with premature ovarian failure. Arch Ophthalmol 2004; 122: 151–156. 44 Teng CT, Liu Y, Yang N, Walmer D, Panella T: Differential molecular mechanism of the estrogen action that regulates lactoferrin gene in human and mouse. Mol Endocrinol 1992; 6: 1969–1981. 45 Adhikary G, Crish J, Lass J, Eckert RL: Regulation of involucrin expression in normal human corneal epithelial cells: a role for activator protein one. Invest Ophthalmol Vis Sci 2004; 45:1080–1087. 46 Chen TT, Wu RL, Castro-Munozledo F, Sun TT: Regulation of K3 keratin gene transcription by Sp1 and AP-2 in differentiating rabbit corneal epithelial cells. Mol Cell Biol 1997;17: 3056–3064. 47 Dogru M, Matsumoto Y, Saeki M, Goto E, Tsubota K: Tear function and ocular surface alterations with lactoferrin treatment in severe dry eyes. Ocul Surf 2005;3(suppl):S57.
Santagati/La Terra Mulè/Amico/Pistone/ Rusciano/Enea
