Replacement of fetal calf serum by human serum as supplementation for human fibroblast culture

Introduction: Fetal calf serum (FCS) is commonly used as a supplement in the culture medium for fibroblast cells. This supplementation is far from ideal as sample quality varies from batch to batch and the composition of FCS is not completely known. In addition, FCS may be contaminated with viruses and/or prions and may also cause adverse immunologic responses in humans. Due to these facts, a worldwide effort is being made to find alternatives for xenobiotic elements in cell cultures. Human serum could be a safer alternative, especially for clinical application. Methods: We investigated human serum as a substitute for FCS in human fibroblast culture. Fresh human serum was obtained from 10 healthy volunteers. Fibroblasts were cultivated in multiwell plates containing either Dulbecco’s modified Eagle’s medium (DMEM) plus 10% FCS (D10) or DMEM plus 10% human serum (D10H). Cell counts were obtained between 24 and 264 hours of cultivation; results were expressed as the mean number of cells + standard error of the mean to create cell proliferation curves. Results: There was no statistical difference in fibroblast proliferation between the two groups. Human serum supported human fibroblast growth and proliferation, suggesting that it may be a potential substitute for FCS in human cell culture. Cells cultivated with human serum presented a different morphology, appearing smaller and more rounded as compared to cells cultivated in D10. Conclusions: These results demonstrate that human serum can be substituted for FCS in human fibroblasts culture and that fibroblasts cultivated in the presence of human serum have a morphology that is similar to in vivo fibroblasts.


INTRODUCTION
Since the beginning of the twentieth century, scientists have searched for methods to isolate and grow skin and/or skin cells for research and clinical application 1 .Currently, liquid or gel cell culture is a fundamental tool for research and for the growth of cells for therapeutic use 2 .For cells to survive and grow in vitro, special combinations of nutrients are required to be present in culture media to provide the optimal conditions for different cell types 3 .The medium used for fibroblast cultivation is universally supplemented with fetal calf serum (FCS), a complex mixture containing growth factors that are necessary to maintain cellular function and permit proliferation 4 .
FCS is easily obtained and contains a high concentration of growth factors and a low concentration of gamma globulins, compared to other sera originating from animals.This fact contributed to the adoption of FCS as the standard supplement for cell culture media 5 .
Normally, FCS is used to supplement the culture medium at a concentration of 5% to 20%.A thousand different components are present in its composition, of which only approximately 200 have been defined.These components include hormones, vitamins, nucleosides, amino acids, lipids, transport proteins (albumin, globin, transferrin), binding factors (fibronectin and laminin), stabilizing factors, detoxifying agents, proliferation factors, and growth factors 6 .
This form of supplementation is not ideal because the component concentrations can change from batch to batch and the full composition of FCS is not known.Further, FCS may contain pathogenic viruses or prions.Additionally, immunological complications may occur since FCS contains proteins that induce immune reactions 7,8 .
FCS leads to increased DNA synthesis in spleen cells of healthy rats 9 and increases the formation of B-lymphocyte plaques in rats, leading to increased DNA and IgM synthesis in these cells 10 .FCS also induces a blastogenic response in peripheral blood lymphocytes 11 , and its presence in human tumor cell culture media may be responsible for antigenic alterations of these cells' membranes 12 .FCS has also been demonstrated to induce nonspecific cytotoxic responses in human lymphocytes, an occurrence which does not happen when the cells are grown in media supplemented with human serum.Both of these responses in cultured human cells were shown to be dose-dependent and triggered by a single exposure 13 .Some studies have shown that patients undergoing cardiomyoplasty, using myofibroblasts grown in the presence of FCS, developed ventricular arrhythmias and suffered sudden death 14 .Burn patients who received transplants of in vitro cultivated keratinocytes have developed immune responses, which led to rejection of the graft 15 .
It is important to emphasize that FCS is classified as a low risk product for the transmission of bovine spongiform encephalopathy (BSE) since the prion that causes this disease cannot cross the placental barrier.The BSE prion has a tropism for the central nervous system of animals and is rarely found in the blood.However, many other diseases can be transmitted through serum, restricting the use of sera derived from animals in areas where such diseases are endemic.Thus, according to international trade rules, FCS must be obtained from areas which are free of cattle diseases; however, quality control tools are relatively primitive and expensive 16 .
Currently, there is a worldwide effort to find alternatives for the use of xenobiotic elements in cell culture media 5,17,18 .The utility of human serum in promoting cell growth was described by Chang 19 following a lengthy study of human epithelial cells cultured in the presence of human serum.These studies demonstrated that sera from different individuals differentially impacted the growth rate of epithelial cells.Other studies with human cells, including conjunctival epithelial cells 20 , chondrocytes 21 , and bone marrow cells 22 , have demonstrated that human serum is capable of supporting the growth and normal phenotype of these cells.Indeed, these studies demonstrated that the cells had a higher rate of multiplication in media supplemented with human serum as compared to that supplemented with FCS 23 .Tallheden et al. 24 demonstrated that human serum has higher levels of growth factors derived from the epidermis and platelets as compared to FCS.
Based on this knowledge, the present research was designed to examine replacing FCS with human serum as a culture enrichment factor.The cells chosen for this initial study were dermal fibroblasts -cells which have a key role in the process of wound contraction.Thus, this paper presents the results of an investigation into the substitution of FCS by human serum in culture media and its impact on promoting the in vitro proliferation of human fibroblasts.

Acquisition of Human Serum
Human serum, 10-20 mL (mean = 15.7 ± 3.9 mL), was recovered from blood samples (400 mL) obtained from 10 healthy adult volunteers (both sexes), submitted for the following serological assessments: HIV, hepatitis B, hepatitis C, syphilis, Chagas disease, HTLV I and II, cytomegalovirus, and toxoplasmosis.The volunteers were informed about the purpose and consequences of the study and then signed an informed consent form.The blood samples were collected by venipuncture, performed at the Blood Bank of the Central Institute of HCFMUSP.The blood samples were sent to the laboratory where the serum was separated by centrifugation (10,000 rpm for 20 minutes).

Study Groups
Cell culture medium containing 10% human serum (D 10 H) was produced by adding an appropriate volume of pooled human serum to Dulbecco's modified Eagle's medium (DMEM; GIBCO 11965.092,Life Technologies, Baltimore, MD, USA) containing 2 mM glutamine (GIBCO 25.030.081,Life Technologies), 100 U/mL penicillin, 100 μg/mL streptomycin, and 25 μg/mL amphotericin B (Antibiotic Antimycotic; GIBCO 15249.062,Life Technologies).This medium was filtered to remove any contaminants and pathogens acquired during the preparation, minimizing the potential risk of loss of growth factors.
Medium supplemented with 10% FCS (D 10 ) was identically prepared except that the human serum was replaced with FCS (GIBCO 16000.044,Life Technologies).

Fibroblast Cultures
Using the explant method, primary cultures of fibroblasts were initiated from full-thickness human skin (n = 4) derived from cosmetic mastoplasty procedures.The explants were cultivated in culture flasks with 25 cm² surface areas (90025 TPP) containing D 10 H or D 10 at 37°C in a humidified, 5% carbon dioxide (CO 2 ) atmosphere.The culture medium was changed every 72 hours.The macroscopic behavior of the skin fragments was observed over time.These primary cultures of fibroblasts reached subconfluence after about four weeks of growth and, subsequently, were trypsinized and subcultured in the same medium.Thereafter, the cells were subcultured at weekly intervals.

Fibroblast Proliferation
Fibroblast growth assays were conducted using nonconfluent, asynchronous cultures.This was accomplished by dispensing 1-mL volumes of culture medium, containing1.5 x 10³ freshly trypsinized fibroblasts, into triplicate wells of 9.62 cm 2 (6 multiwell plates; Falcon 35 3047).Cultures were incubated for 11 days.After 24, 48, 72, 96, 120, and 264 hours, the fibroblasts were trypsinized and the number of released cells were counted using a Neubauer hemocytometer; cell viability was assessed by dye exclusion using 0.4% trypan blue (Sigma T-8154).The tests were performed on triplicate cultures with the number of cells from each culture being counted twice; values were expressed as the number of cells per well + standard error of the mean (SEM).

Statistics
Cohort or follow-up studies for independent samples were performed; P < 0.05 was considered significant.

Macroscopic Aspects of the Cultures
Skin fragments demonstrated greater difficulty adhering to the flask surface in the presence of D 10 H as compared to D 10 (Figure 1).

Morphology of Dermal Fibroblasts
Observed with an inverted light microscope, skin explants in D 10 H-containing media took one week longer to release fibroblasts when compared to explants cultured in media with D 10 .Cells in D 10 H-containing media were more elongated and presented more intracytoplasmic granules than those maintained in D 10 -containing media (Figure 2).After several subcultures, the fibroblasts changed their phenotype, becoming more rounded in the presence of D 10 H than those cultured with D 10 (Figure 3).

Fibroblast Proliferation Assay
Fibroblast proliferation was observed throughout the experiments.After 48 hours, there were 6,565 + 590.83 fibroblasts in the medium with human serum and 6,041.5 + 216.81 in the medium with FCS.After 11 days, the number of cells was 99,791.5 + 24,793.93 in D 10 H media and 68,645.75+ 22,452.39 in D 10 media (Figure 4).

DISCUSSION
FCS is the most widely used animal serum supplement for cell culture media because it is easily available, is easily stored, and contains high concentrations of growth factors.However, this supplement is far from ideal because its nutrient concentration varies from batch to batch, its composition is not fully defined, and it may be contaminated with viruses or prions 7,8 .For clinical applications, it would be safer to avoid using xenobiotic elements in human cell culture.
The use of human blood elements in cell culture has shown promising results, which has encouraged researchers to replace FCS in in vitro cultures of human fibroblasts by a similar nutrient derived from humans.Kurita et al. 25 compared fibroblast growth in media supplemented with three human blood derivatives: serum, platelet-rich plasma, and platelet-poor plasma.They observed that serum and platelet-rich plasma stimulated intensive fibroblast growth, with results similar to those observed in FCS-supplemented media; the proliferation rate was significantly lower using platelet-poor plasma.These findings correlate with the presence of growth factors in platelets, such as platelet-derived growth factor that is known to be mitogenic for fibroblasts 26 .They also observed that the ability to stimulate cell proliferation varied somewhat between plasma or serum samples obtained from different individuals, but the variation was not linked to donor gender or age.
Remarkably, in this study, we observed that adhesion of skin fragments to the surface of the culture flask was delayed in the presence of human serum (Figure 1).Since this result was not yet reported, further studies are necessary to confirm and clarify this finding.Phenotypic differences between in vivo and in vitro grown fibroblasts have previously been reported.One such observation was that monolayers of fibroblasts cultured in the presence of FCS presented as wider cells as compared to the more rounded cells found, in vivo, in extracellular matrix 27 .Similarly, we also observed morphological differences between fibroblasts grown in the presence of FCS and those grown in human serum-supplemented media.Cultured fibroblasts showed a more rounded morphology when exposed to D 10 H medium than when grown in D 10 , suggesting that the presence of human serum is responsible for the in vivo phenotype (Figures 2 and 3).
However, this delayed adhesion was not observed to have an impact on the proliferation rate of these cells; the proliferation rate of the fibroblasts was the same regardless of whether the cells were grown in D 10 H or D 10 (Figure 4), as previously reported by Cooper & Goldstein 28 .However, these authors also reported that the number of rounds of replication that occurred over the lifespan of the cultured cells was significantly reduced in the presence of human serum.   analyzed the lifespan and protein synthesis of cultured fibroblasts in 10% human serumsupplemented media compared to that of cells cultured in 10% FCS-supplemented medium.The authors observed that cellular proliferation was faster in the human serumcontaining media, leading to shorter duration of cellular duplication and maintenance of the physiological phenotype of the fibroblasts.However, this increased cellular proliferation rate was not observed in the present study, possibly due to the difference in the composition of the medias used in the two studies.Mazlyzam et al. 29 used Ham's F12 and DMEM supplemented with 10% FCS, while in our study we did not include Ham's F12 in the media.
According to the Mazlyzam et al. 29 report, the level of type I collagen gene expression was maintained in cultured fibroblasts with an increase of its expression in early passages and up-regulation of genes related to the expression of type III collagen.This is important for the down-regulation of gene expression of alpha-smooth muscle actin and of extracellular proteins involved in wound contraction and cicatrix formation 29 .
Kondo et al. 30 obtained human fetal skin fibroblasts and compared in vitro cell migration in culture media containing either human serum or FCS.In 48% of the samples, human serum inhibited cell migration and maintained higher levels of proliferation.These authors also observed that the inhibition was dependent on the human serum concentration and on the age of the serum donor, with serum from older donors yielding a more pronounced inhibition of migration 31 .This inhibition may be attributed to an interference with collagen synthesis, but the authors also raised the hypothesis that inhibition would result from the levels of alpha 2-macroglobulin and low-density lipoprotein.
We are performing other studies to examine the possibility of substituting other human blood components, such as a platelet lysate, for FCS.We intend to compare the results obtained from the new studies with those obtained by other authors.The ultimate aim of this work is to produce humanized skin that can be used clinically in the treatment of chronic ulcers of the lower limb.This next phase of work will be conducted at the Cellular Therapy Laboratory of the Central Institute of the Clinics Hospital.

CONCLUSION
The results obtained in this study indicate that human serum can substitute for FCS in human fibroblast culture medium.This substitution is especially important for clinical application because it minimizes the possibility of immunological complications related to the use of xenobiotic components.

Number of cells
Committee for Analysis of Research Projects (CAPPesq), Clinics Hospital of the Medical College of the University of São Paulo (FMUSP), number 2010 (01/23/2008), and was conducted in the Research Cell Culture and Wound Healing Laboratory within the Department of Plastic Surgery of FMUSP.

Figure 1 -
Figure 1 -Primary culture in 25-cm² surface flasks one week after explanting.On the right, three bottles containing D 10 H and on the left, three bottles containing D 10 .A lower adherence of skin fragments in the presence of human serum as compared to the medium with fetal calf serum was observed.

Figure 2 -
Figure 2 -Cultured fibroblasts at two weeks.In the presence of human serum (A), the cells were more elongated and presented larger number of intracytoplasmic granules, as compared to fibroblasts grown in fetal calf serum (B).In both cases, keratinocytes are present among the fibroblasts (OM -10x).

Figure 3 -
Figure 3 -Details of cell culture after three weeks.Breakdown in the organization and small cell volume were observed among the fibroblasts grown in the presence of human serum (A and C) as compared to fibroblasts grown in fetal bovine serum (B and D).Fibroblasts cultured in the presence of human serum demonstrated a more rounded morphology (E) as compared to the cells grown with fetal calf serum (F) (OM -10x).Number of cells + standard error of the mean (SEM); P < 0.001.

Figure 4 -
Figure 4 -Comparison of the proliferation rate of human fibroblasts (n = 4) cultured in the presence of fetal calf serum or human serum over time (hours).SEM = standard error of the mean; FCS = fetal calf serum; HS = human serum.