Materials and Methods

Sample population—Blood samples collected from 15 Holstein bull calves were used in the study. Birth of each of these calves was observed, and each calf was provided colostrum by use of an orogastric feeder shortly after birth. Calves were raised in the Food Animal Clinic of the College of Veterinary Medicine at the University of Missouri. Calves were housed separately and fed 2 L of a commercial milk replacer twice daily until they were 8 weeks old. Calves were allowed ad libitum access to choice alfalfa hay, a commercial calf starter ration, and water. Blood samples were collected from calves for use in lymphocyte proliferation analysis (calves 1 to 8; 1 to 4 months of age), PCR assay of COX-2 expression (calves 9 to 12 and 15; 4 to 8 months of age), PCR assay of MMP-9 expression (calves 9 to 12; 4 to 8 months of age), and measurement of PGE₂ concentrations (calves 9 to 14; 4 to 11 months of age). The experiments were evaluated and approved by the University of Missouri Institutional Animal Care and Use Committee.

Isolation of PBMCs and initial culture—Blood samples (50 mL) were collected by jugular venipuncture into tubes containing acid citrate dextrose (acid citrate dextrose-to-blood ratio, 1:9). Blood was centrifuged at 1,000 × g for 15 minutes at 23°C. After centrifugation, PBMCs were isolated by initial harvest of buffy coats.¹⁰ Buffy coats were suspended in HBSS^a and layered over an equal volume of nonionic, tri-iodated, water-soluble, low-density gradient medium^b (specific gravity, 1.083). Samples were centrifuged at 342 × g for 30 minutes at 23°C, and cells suspended between the HBSS and nonionic, tri-iodated, water-soluble, lowdensity gradient medium were aspirated with a pipette. Following harvest, cells were washed twice, resuspended in HBSS, and placed on ice. Viability of PBMCs was determined by exclusion of 0.1% trypan blue dye.¹¹ Differential cytologic examination was performed by counting 200 cells prepared by use of a cytocentrifuge system^c and stained with Wright stain.

PBMC treatment—Mixed populations of PBMCs (4 × 10⁶ cells) isolated from each of the calves were placed into 12-well culture plates in 2 mL of RPMI 1640^d supplemented with 10% heat-inactivated FBS,^e 0.1mM nonessential amino acids,^f 2mM L-glutamine,^g 55μM 2-mercaptoethanol,^h and penicillin-streptomycinⁱ (50 U/mL and 50 μg/mL, respectively). After isolation, cells were cultured for 24 hours at 37°C in an atmosphere of 5% carbon dioxide. After culture for 24 hours, medium in wells was removed and replaced with 1 mL of fresh RPMI 1640 with FBS and supplements. Nonadherent cells in the removed media were pelleted by centrifugation (117 × g for 5 minutes at 23°C), and the pellet was resuspended in 1 mL of RPMI 1640 with FBS and supplements. Viability of nonadherent cells was evaluated by trypan blue dye exclusion, and the remaining suspension was added to the appropriate wells (final volume, 2 mL). Viability of adherent cells was assessed subjectively by visual examination of these cells in the tissue culture plates.

Cell treatments included a control treatment (no treatment), lactoferrin^j (200 ng/mL), LPS stimulation (Escherichia coli strain O55:B5^k; 1 μg/mL), and treatment with lactoferrin (200 ng/mL) followed by LPS stimulation (1 μg/mL). For the latter, lactoferrin was added to wells 1 hour before LPS was added; lactoferrin remained in the wells for the duration of the experiment.

Incorporation of ³H-thymidine by proliferating PBMCs—The PBMCs from calves 1 to 8 were used for proliferation studies, as described¹² but with modifications. After addition of each treatment, cells remained in culture for 48 hours, after which ³H-thymidine^l (1 μCi/well) was added to each well. Twenty-four hours later, contents of each well were thoroughly mixed by pipetting, and 100 μL of medium and cells were transferred to 96-well tissue culture plates (in triplicate). Cells were harvested onto filter mats^m by use of an automated cell harvester.ⁿ Wells were not scraped prior to transfer of suspended cells to ensure evaluation of nonadherent cellular proliferation. Radioactivity retained in the filter mat, expressed as mean CPM, was determined by use of a workstation.^o Incorporation of ³H-thymidine by PBMCs (in CPM) was converted to the incorporation index by standardizing each treatment to results of control cell incubations conducted concurrently by use of the following equation:

View Expanded

Gene expression—The PBMCs from calves 9 to 12 and 15 were used to evaluate COX-2 gene expression, whereas PBMCs isolated from calves 9 to 12 were used to evaluate MMP-9 gene expression. Cells were cultured by use of the same conditions as described for the proliferation experiments, except that isolated PBMCs were cultured for 24 hours in medium after addition of treatments.^13,14 Twenty-four hours after addition of treatments (control treatment, lactoferrin, LPS stimulation, and lactoferrin followed by LPS stimulation), culture medium was aspirated and centrifuged (117 × g) to enable collection of nonadherent cells; supernatant was aspirated to leave only the pellets of the nonadherent cells. Adherent cells were immediately lysed with a monophasic solution of phenol and guanidine isothiocyanate^p and incubated for 5 minutes on ice. The solution for the lysed adherent cells was aspirated and added to each tube containing the nonadherent cell pellet. The resulting total RNA preparation was purified by use of phenol-chloroform extraction and ethanol precipitation, as described elsewhere.^13,15 Concentration and purity of RNA was determined by use of a spectrophotometer, and 2 μg of RNA was used to generate cDNA in reverse transcriptase reactions.^13,14 Expression of COX-2, MMP-9, and GAPDH was determined by use of PCR assays, as described elsewhere.^13,14 Briefly, cDNA was prepared by addition of 0.5 μg of oligo-(dT)_12–18, PCR buffer (20mM Tris HCl [pH, 8.4] and 50mM KCl), 2.5mM MgCl₂, 10mM dithiothreitol, 0.2mM of each dNTP, and 50 units of MuLV–RT (final volume, 50 μL). A negative control sample was assayed simultaneously by substituting diethylpyrocarbonate-treated water for the MuLV–RT to control for genomic DNA. Cycling conditions for the cDNA synthesis were 2 minutes at 42°C, 50 minutes at 42°C, 15 minutes at 70°C, and 5 minutes at 5°C.¹³ Cycling conditions for PCR assays were as described elsewhere,^13,14 with minor modifications (Appendix). The PCR assays for each calf and each gene were conducted simultaneously. The PCR amplification consisted of 5 μL of reverse transcription template or water, 1.5mM MgCl₂, 1 × PCR buffer, 0.2mM of each dNTP, 10μM of each primer, and 1.5 units of Taq polymerase (final volume, 100 μL), with the aforementioned thermocycler conditions.^13,14 Products of PCR amplification (COX-2, 449 bp; MMP-9, 658 bp; and GAPDH, 468 bp) were analyzed on ethidium bromide–stained 2% agarose gels. Scanned gels were analyzed by inverting each gel image such that product bands appeared black on a white background. Stored images were analyzed by use of an image-analysis program.^q Gels were evaluated such that the gene product of each inflammatory mediator and corresponding GAPDH were scanned concurrently. Band density was determined as peak area for each gene and standardized on the basis of the peak area for GAPDH. The ratios of COX-2 to GAPDH or MMP-9 to GAPDH were adjusted by dividing the GAPDH-standardized gene expression for COX-2 and MMP-9 to provide fold increases (or fold decreases) in gene expression by use of the following equation¹³:

View Expanded

Production of PGE₂—For samples obtained from calves 9 to 14, supernatants of cell cultures of PBMC incubations used for gene expression experiments were evaluated for PGE₂ production by use of a competitive enzymeimmunoassay.^13,r The assay was based on the addition of sample or standards containing PGE₂, which compete with PGE₂-acetylcholinesterase conjugate (tracer) for binding to a monoclonal anti-PGE₂ antibody. Plates coated with goat polyclonal anti-mouse antibody were used, and capture monoclonal antibody bound to PGE₂ or PGE₂-acetylcholinesterase conjugate were added to each well. Absorbance of each well was proportional to the concentration of bound PGE₂-acetylcholinesterase conjugate, which was inversely proportional to the PGE₂ concentration in the sample or standard.^r

Each sample was diluted 1:100 with assay sample buffer, and aliquots (50 μL) were analyzed in triplicate in accordance with the manufacturer's recommendations. An 8-point standard curve was established with known concentrations of PGE₂ by serial dilution of a stock solution (10 ng/mL; provided with the kit); serial dilutions were achieved by use of assay sample buffer. To each well, 50 μL of PGE₂-acetylcholinesterase conjugate and 50 μL of monoclonal anti-PGE₂ were added. Values for standards and samples were determined simultaneously in an identical manner. After addition of culture supernatants to wells, plates were covered with plastic film and incubated for 18 hours at 4°C in the dark. Wells were emptied and rinsed 5 times with wash buffer, after which binding of PGE₂-acetylcholinesterase conjugate was determined by use of 200 μL of an acetylcholinesterase substrate (5,5′-dithio-bis-[2-nitrobenzoic acid]).

Blank wells (empty wells) and wells for total activity (5 μL of PGE₂-acetylcholinesterase conjugate), NSB (50 μL of PGE₂-acetylcholinesterase conjugate), and maximal binding (ie, B₀; 50 μL of PGE₂-acetylcholinesterase conjugate and monoclonal anti-PGE₂ antibody) were included for calculation of PGE₂ concentrations. Values for B₀ were corrected by determining the mean absorbance values for duplicate wells and subtracting the mean of the NSB wells. For each standard concentration and each sample, the ratio of the percentage of PGE₂ bound versus the B_o value was determined by subtracting absorbance for NSB from the absorbance for the standard or sample and dividing the difference by the corrected value for B₀ (ie, B₀ – NSB). These ratios were then multiplied by 100 to determine the percentage of bound PGE₂. The data were entered into a spreadsheet and subjected to 4-point logistic regression analysis in accordance with the manufacturer's recommendations. Data were standardized on the basis of values for PGE₂ production in control cells to account for animal-to-animal and day-to-day variation by use of the following equation: PGE₂ production = (PGE₂ for treated PBMCs – PGE₂ for control cells)/PGE₂ for control cells. The stimulation indices were logarithmically transformed for data analysis.

Statistical analysis—Normality of proliferation data was evaluated by use of the Kolmogorov-Smirnov test, and equal variance was evaluated by use of the Levene median test. Proliferation indices were logarithmically transformed and compared by use of a Kruskal-Wallis 1-way ANOVA on ranks. Differences between treatments were considered significant at values of P < 0.05. The LPS-induced gene expression and effects of lactoferrin on basal and LPS-induced gene expression were adjusted on the basis of results for GAPDH. The ratio of control or treated cellular gene expression was logarithmically transformed and compared by use of a 1-way ANOVA, with post hoc testing with the Holm-Sidak test. Concentrations of PGE₂ were compared by use of a 1-way ANOVA.

Results

Cell culture, cell viability, and differential cell counts—Mean ± SD cell viability after isolation was approximately 89.4 ± 7.3% (range, 74% to 98%). Cellular differential counts for lymphocytes and monocytes revealed approximately 85 ± 9% (range, 70% to 97%) and 15 ± 9% (range, 3.5% to 22%), respectively. Cellular viability of nonadherent cells after 24 hours of culture was 94.4 ± 4% (range, 89% to 100%).

Effect of lactoferrin on ³H-thymidine incorporation—Stimulation indices (standardized on the basis of results for control cells analyzed simultaneously) revealed that LPS induced bovine PBMCs to incorporate ³H-thymidine, compared with results for lactoferrin-treated cells (Figure 1). In PBMCs treated with lactoferrin followed by LPS, significantly lower amounts of ³H-thymidine were incorporated, compared with results for LPS-treated cells. The proliferation index for lactoferrin-treated PBMCs did not differ significantly from that for the control wells.

Median ± median deviation for the ³H-thymidine incorporation index in PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 8 calves. The ³H-thymidine incorporation index was determined by subtracting the CPM for the control cells from the CPM for the treated cells and was standardized on the basis of the CPM for the control cells. *Value differs significantly (P < 0.05) from values for other treatments.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Median ± median deviation for the ³H-thymidine incorporation index in PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 8 calves. The ³H-thymidine incorporation index was determined by subtracting the CPM for the control cells from the CPM for the treated cells and was standardized on the basis of the CPM for the control cells. *Value differs significantly (P < 0.05) from values for other treatments.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Median ± median deviation for the ³H-thymidine incorporation index in PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 8 calves. The ³H-thymidine incorporation index was determined by subtracting the CPM for the control cells from the CPM for the treated cells and was standardized on the basis of the CPM for the control cells. *Value differs significantly (P < 0.05) from values for other treatments.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Effect of lactoferrin on LPS-induced gene expression—Resting PBMCs cultured in vitro revealed variations in COX-2 gene expression, which was markedly less than that for LPS-treated PBMCs (Figure 2). Cells treated with lactoferrin did not differ significantly from control PBMCs with regard to COX-2 gene expression. Stimulation of PBMCs by LPS was associated with significant increases in COX-2 gene expression, compared with gene expression for control cells (mean ± SD increase of 5 ± 3-fold; P = 0.006) and lactoferrin-treated PBMCs (mean increase of 4.5 ± 1.8-fold; P = 0.006). Treatment with LPS induced a significant (P = 0.005) increase of 9 ± 6-fold in COX-2 gene expression, compared with gene expression in PBMCs treated with lactoferrin followed by LPS.

Mean ± SD fold increase in gene expression for COX-2 (A) and MMP-9 (B) in PBMCs incubated with no treatment (control group [gray bar]), lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 5 and 4 calves for panels A and B, respectively. Results represent the fold increase in gene expression and were standardized on the basis of gene expression for a housekeeping gene (GAPDH). *,†Value differs significantly (*P = 0.006; †P = 0.005), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean ± SD fold increase in gene expression for COX-2 (A) and MMP-9 (B) in PBMCs incubated with no treatment (control group [gray bar]), lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 5 and 4 calves for panels A and B, respectively. Results represent the fold increase in gene expression and were standardized on the basis of gene expression for a housekeeping gene (GAPDH). *,†Value differs significantly (*P = 0.006; †P = 0.005), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Mean ± SD fold increase in gene expression for COX-2 (A) and MMP-9 (B) in PBMCs incubated with no treatment (control group [gray bar]), lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 5 and 4 calves for panels A and B, respectively. Results represent the fold increase in gene expression and were standardized on the basis of gene expression for a housekeeping gene (GAPDH). *,†Value differs significantly (*P = 0.006; †P = 0.005), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Furthermore, LPS treatment resulted in a significant increase (> 2-fold) in MMP-9 gene expression, compared with gene expression for control cells (Figure 2). The MMP-9 gene expression for PBMCs treated with lactoferrin followed by LPS was significantly (P = 0.005) reduced, compared with gene expression of PBMCs treated with LPS alone. Peripheral blood mononuclear cells treated with nothing (control) and lactoferrin each expressed significantly (P = 0.006) less MMP-9 gene than that expressed by cells treated with LPS. The MMP-9 gene expression for PBMCs treated with lactoferrin followed by LPS did not differ significantly from gene expression for PBMCs incubated with lactoferrin alone or from gene expression for control cells.

Effect of lactoferrin on production of PGE₂—Concentrations of PGE₂ produced by control cells ranged from 13 to 49 ng/mL. Addition of lactoferrin to PBMCs did not significantly alter basal PGE₂ production. The LPS-treated PBMCs produced significantly more PGE₂ than did control cells and lactoferrin-treated PBMCs (Figure 3). Treatment with lactoferrin followed by LPS stimulation significantly attenuated production of PGE₂, compared with production by PBMCs treated with LPS alone.

Median ± median deviation PGE₂ production by PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 6 calves. Results are reported as the fold increase for the PGE₂ stimulation index. Data were standardized on the basis of values for PGE₂ production in control cells by use of the following equation: PGE₂ production = (PGE₂ for treated PBMCs – PGE₂ for control cells)/PGE₂ for control cells. *Value differs significantly (P < 0.05), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Median ± median deviation PGE₂ production by PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 6 calves. Results are reported as the fold increase for the PGE₂ stimulation index. Data were standardized on the basis of values for PGE₂ production in control cells by use of the following equation: PGE₂ production = (PGE₂ for treated PBMCs – PGE₂ for control cells)/PGE₂ for control cells. *Value differs significantly (P < 0.05), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Median ± median deviation PGE₂ production by PBMCs incubated with lactoferrin (200 ng/mL [black bar]), LPS (1 μg/mL [cross-hatched bar]), or lactoferrin (200 ng/mL) followed 1 hour later by LPS (1 μg/mL [diagonal-striped bar]). The PBMCs were obtained from 6 calves. Results are reported as the fold increase for the PGE₂ stimulation index. Data were standardized on the basis of values for PGE₂ production in control cells by use of the following equation: PGE₂ production = (PGE₂ for treated PBMCs – PGE₂ for control cells)/PGE₂ for control cells. *Value differs significantly (P < 0.05), compared with the value for PBMCs treated with LPS alone.

Citation: American Journal of Veterinary Research 69, 9; 10.2460/ajvr.69.9.1164

• Download Figure
• Download figure as PowerPoint slide

Discussion

Gram-negative bacterial infections are common in livestock and are often severe in neonates. The processes leading to morbidity result from exaggerated responses of the host innate immune system to gramnegative bacteria or their cellular products.¹⁶ Lipopolysaccharide is a component of the outer membrane of gram-negative bacterial cell walls, is mitogenic, and has profound effects on the humoral and cellular immune systems.¹⁷ During endotoxemia, systemic generation of proinflammatory molecules subsequent to activation of mononuclear cells and the culmination of a myriad of intracellular processes, including gene and protein expression, protein secretion, and activation of other cellular processes, lead to septic shock.^18–20 Septic shock has been associated with cell-generated cytokines, chemokines, and eicosanoid mediators.^18,19 These soluble molecules also play roles in cellular proliferation.²¹

The objectives of the study reported here were to determine the ability of lactoferrin to alter LPS-induced ₃ H-thymidine incorporation by PBMCs as a measure of cell proliferation. We also evaluated whether lactoferrin would modify LPS-induced production of proinflammatory cytokines and lipid mediators in vitro. The experimental design was based on the cellular effects of bacterial LPS in the presence of lactoferrin. Several factors were assumed during the performance of these experiments. First, ³H-thymidine incorporation into PBMCs subsequent to LPS stimulation requires antigen-presenting cells (monocytes) and lymphocytes. Incorporation of ³H-thymidine was used as an end point indicative of DNA replication. Second, PBMC proliferation requires iron provided through interactions with transferrin in vitro. Third, interactions of LPS with the acute-phase reactant, LPS-binding protein, enhance LPS-induced activation of monocytes, which are the most efficient activator of differentiated effector lymphocytes and one of the major antigen-presenting cells of hosts.^17,18 Fetal bovine serum is a natural source of both transferrin and LPS-binding protein.^22,23 The in vitro assay included all factors required for cellular proliferation to determine whether lactoferrin modulates LPS-induced proliferation of bovine PBMCs.^22,23 Fourth, the mixed population of PBMCs was appropriate for evaluation of ³H-thymidine incorporation of nonadherent cells, which are primarily lymphocytes. Gene expression and prostanoid evaluations represented contributions of all PBMCs in each sample. We assumed that lymphocytes would proliferate, and this proliferation was the result of gene expression and prostanoid production of all PBMCs.

Results of our PBMC proliferation experiments differed from those of other investigations in terms of maximal proliferation indices after addition of LPS.¹⁷ We attributed these differences to several factors. Maximal LPS stimulation is believed to require serum LPS-binding protein. Because our study used heat-inactivated FBS, reduction of the serum activity of LPS-binding protein could be expected.²³ To improve the interaction of LPS with cellular receptors, we dissolved LPS in FBS to solubilize aggregate LPS prior to incubation with cells, as described elsewhere.²⁴ Furthermore, the concentration of LPS (1 μg/mL) used in our incubations was relatively high, which should have improved exposure of cells to this stimulus. Despite our efforts to limit the impact of heat-treated FBS, these manipulations may have resulted in lower ³H-thymidine incorporation indices. However, the general response of LPS-treated PBMCs was to induce ³H-thymidine incorporation, gene expression, and prostanoid production. The general response to treatment of PBMCs with lactoferrin followed by LPS stimulation was to significantly reduce LPS-induced effects.

In our study, lactoferrin (200 ng/mL) significantly reduced ³H-thymidine incorporation induced by LPS. The reduction in LPS-induced ³H-thymidine incorporation associated with lactoferrin treatment of PBMCs in vitro suggested that lactoferrin may be capable of inhibiting LPS-stimulated proliferation of PBMCs in vivo. We attributed this to the reported binding affinity of lactoferrin for LPS, which prevents interaction between LPS and cells bearing appropriate receptors.²⁵ These results are strengthened by results of another study²⁶ in which supplemental lactoferrin provided immediately after birth was absorbed systemically. Increases in serum concentrations of lactoferrin may be useful in reducing the systemic effects of LPS on neonates.²⁶

The assay was not able to discriminate between the effects of lactoferrin on LPS binding and lactoferrin-mediated iron sequestration on reduced cellular incorporation of ³H-thymidine. To evaluate whether lactoferrin sequestration of iron in the media results in reduced ³H-thymidine incorporation, it would have been necessary to quantitate the amount of iron and transferrin in the media and determine the saturation of transferrin and lactoferrin prior to addition of LPS.^27,28 Furthermore, it would have required us to conduct experiments to evaluate unscheduled DNA synthesis in the absence of cell division. This was beyond the scope of the study reported here.

The ability of lactoferrin to interfere with LPS-induced ³H-thymidine incorporation by PBMCs was also associated with diminished expression of COX-2, which is the inducible form of COX responsible for PG production. These results were supported by the reduction in the LPS-induced production of PGE₂ by lactoferrin-treated PBMCs. The LPS-induced expression of COX-2 and production of PGE₂ by bovine alveolar macrophages in vitro have been reported.¹³ Induction of COXs is associated with availability of substrates resulting from activation of membrane phospholipases.¹³ Activation of phospholipase is greatly enhanced by bacterial LPS.^13,29,30 Results of our in vitro study support the possibility that lactoferrin interferes with LPS-induced production of prostanoids through a mechanism involving the expression of an LPS-inducible gene (ie, COX-2). In other studies, investigators have reported that prostanoids play a role in physiologic and morphologic changes associated with gram-negative bacterial infections.^31,32 Inhibition of LPS binding may reduce the systemic effects of endotoxin on a host, including release of arachidonic acid from membranes and production of bioactive prostanoids. Studies evaluating the amount (if any) of inducible COX protein would provide information regarding this mechanism.

In another study¹⁴ conducted by our laboratory group, we determined that the expression of the MMP-9 gene and MMP-9 protein by bovine alveolar macrophages in vitro is induced by exposure to bacterial LPS. The production and release of this protease in cattle have been associated with pneumonia caused by gramnegative bacteria.³³ In the study reported here, we determined that treatment of PBMCs with lactoferrin followed by LPS stimulation inhibited LPS-induced MMP-9 gene expression. Unfortunately, we could not easily study the protein expression of MMP-9 by PBMCs for reasons related to cellular growth requirements. The experimental design hinged on a source of LPS-binding protein, a natural constituent of FBS. Serum concentrations of MMP-9 are usually low but can be measured zymographically and immunologically. However, we were not convinced that detection of differences in MMP-9 protein produced by PBMCs, compared with MMP-9 protein contained in FBS, would have sufficient sensitivity to warrant further evaluation of this protein.

In another study²¹ conducted to evaluate the effects of human lactoferrin on lymphocyte proliferation, investigators reported that lactoferrin mitigates the proliferative effects of high concentrations of iron, which suggests that lactoferrin has a protective effect mediated through binding of iron. Because we were more interested in determining the ability of lactoferrin to modulate the effect of LPS, similar investigations were beyond the scope of the study reported here. The role of lactoferrin in mitigating LPS-induced cellular production of cytokine has been evaluated,²¹ and analysis of results indicates that lactoferrin has the ability to prevent LPS interaction with cells.

Data for the study reported here supported a role of lactoferrin in antagonism of LPS-induced ³H-thymidine incorporation, gene expression, and mediator production in PBMCs. We presumed this to be via the ability of lactoferrin to bind LPS. However, other mechanisms in which expression of inflammatory mediators may be inhibited should not be disregarded without further investigation.³⁴ Lactoferrin inhibition of LPS-induced cellular effects is not fully understood; however, binding and sequestration of lipid A may be an important and natural way to reduce the systemic effects of endotoxin. Lactoferrin inhibition of bacterial growth by sequestration of iron may also limit the effects of sepsis on neonates.³⁵ Analysis of our results suggests that lactoferrin modulates LPS-induced cellular effects and corroborates results of other studies^4,34,35 in which investigators indicated that lactoferrin has multiple functions in vitro. Results of the study reported here provide in vitro data that suggest lactoferrin may be effective in limiting some aspects of gram-negative bacterial inflammation in cattle.