Optimal assay conditions for quantifying fibronectin in saliva

LLENA-PUY MC, MONTAÑANA-LLORENS C, FORNER-NAVARRO L .OPTIMAL ASSAY CONDITIONS FOR QUANTIFYING FIBRONECTIN IN SALIVA. MED ORAL 2004;9:191-6.

ABSTRACT

Introduction: Fibronectin (Fn) is a glycoprotein that is present in many body fluids and tissues in both physiological and pathological conditions. It can also be detected in the saliva, although only in very small quantities and frequently in broken chains. It induces bacterial aggregation and its levels fall when those of cariogenic or periodontal pathogenic bacteria rise. The infective capacity of the saliva of patients infected by human immunodeficiency virus (HIV) has been linked to the levels of this protein. In some chronic conditions of the oral mucosa, such as oral lichen planus, the concentration of salivary fibro-nectin is lower than usual. Fibronectin quantity also varies in the presence of some tumours, such as oral squamous cell carcinoma, although it cannot be considered a specific factor.
Aims: Due to the low Fn concentration in saliva and its lability in the soluble form, sample collection and conservation conditions are extremely important. The aim of this study is therefore to standardise these conditions so that the Fn can be quantified in an optimum manner.
Materials and methods: The Fn concentration in human saliva was determined in 20 healthy subjects aged between 28 and 54 by means of the ELISA technique and the concentration of the protein in fresh samples kept at 4ºC for 24 hours was compared with that of frozen samples kept at -40ºC for different periods of time.
Results and Conclusions: After comparing different ways of conserving the saliva samples, we found that the optimum conditions were to collect the samples in glass tubes and to quantify them immediately after collection or conserve them at 4ºC and quantify them within a maximum of 24 hours. Freezing and later thawing for quantification induced losses of up to 60% of the protein.

Key words: Fibronectin, saliva, quantification, storage

INTRODUCTION

Fibronectin (Fn) is a multifactor glycoprotein that is found in soluble form in the blood, saliva and other body fluids (1, 2), while the insoluble form interacts with several extracellular matrix components. Fn levels are positively correlated to acute infectious pathology as well as to chronic processes such as cirrhosis of the liver or liver carcinoma (3, 4). Fn is composed of two 230 KDa sub-units linked by two disulphide bonds with a complex structure, subdivided into a series of small domains, each of which ONE is apparently responsible for certain of its functions. Two of these, the N-terminal and the central domain of the sequence respectively, are in charge of binding to the collagen and fibrin and to the eukaryotic cells (5, 6).

Fn is a non-specific defence factor in saliva, like lysozyme, lactoferrin, beta 2 microglobulin, histidin, mucin, or the peroxidase system (7). It plays an essential role in cell-binding mechanisms, is capable of binding to bacteria (8-12) and to hydroxyapatite and contributes to acquired pellicle and bacterial plaque formation (13). It can also be found in soluble form in the total saliva or in the parotid, submaxillary or sublingual saliva, as well as in the crevicular fluid (14).

The bacterial aggregation capacity of saliva is lessened when the Fn level or other cell binding factors decreases and administering them to a previously depleted saliva restores the interaction between the saliva and the bacteria present in it, such as Streptococcus spp (15). Consequently, Fn is considered to be a factor that is responsible for bacterial aggregation (16). Fn levels in the saliva of patients with periodontal disease are lower than in healthy patients. We know that this protein is one of the factors responsible for Porphyromona gingivalis fimbriae binding (17). An inverse correlation between Streptococcus mutans levels and the concentration of soluble Fn in the saliva has also been shown to exist (18). A recent study has shown that the Qrg-Gly-Asp (RGD) amino acid sequence is the protein's binding domain for Streptococcus mitis (19).

Fn also interacts with the glycoproteins of the human immunodeficiency virus (HIV) capsule and can have a positive effect in reducing the transmission of this virus (20, 21). Hydrogen peroxide induced cell death in fibroblast cultures is slowed when Fn is added to these cultures (22). In patients with oral lichen planus, the levels of salivary Fn are lower than in healthy patients (23). An isoform of this protein obtained by glycosylation of normal Fn and some extra segments of the protein is strongly correlated with a large number of tumours, including oral squamous cell carcinoma, but its levels in saliva do not exhibit any specificity in relation to the tumour (24).

Fn levels in saliva and other human fluids are very low (25) and sample handling is critical. Consequently, the aim of this study is to review the optimum conditions for sample handling and storage in order to permit an appropriate determination of the levels of this protein.

MATERIALS AND METHODS

Between 7 and 10 ml of paraffin stimulated whole saliva were collected from 20 subjects aged between 28 and 54 years. The samples were divided into 7 aliquots. Four were collected in glass tubes, frozen immediately with liquid nitrogen and stored at -40ºC, one was collected in a glass tube and processed immediately to quantify the protein, another was stored in a glass tube at 4ºC for 24 hours before being used and the last was stored in the same way but in a plastic tube. The frozen samples were analysed at 1, 2, 4 and 8 weeks respectively. After thawing at 37ºC, they were kept at 4ºC while being processed. Human plasma Fn (Boheringer Mannheim), diluted in distilled water and frozen at -40ºC, was used as the standard protein.

A modified ELISA technique was used to quantify the Fn. The double antibody method described by Lamberts et al. (26) was used to identify Fn recognition by the antibodies, employing polyclonal anti-human plasma Fn antibodies, AP-conjugated anti-rabbit IgG (Promega Corp.), 9.7% diethanolamine, pH 9.8 (Sigma Aldrich), Mg SO₄ (1mM) (Merck Igoda S.A) and p-nitrophenol (1mM). The method was modified as follows: the human plasma Fn was diluted in PBS in order to use duplicates in a 10 to 250 ng/ml range to obtain concentrations from 1/2 (w/v) to 1/4, 1/10 and 1/24 (v/v). The samples were placed on plates (Nunc MaxiSorp) and incubated for 12 hours at 4ºC with the respective antibodies. They were then washed with fibronectin-free BSA (25). The absorbency was measured at 405 nm after 40 minutes in an ELISA reader (Novapath BIORAD Laboratories S.A).

The percentage protein loss was determined for the different conditions, taking the immediate determination as 100%. The average values and 95% levels of confidence were calculated for each storage group. All the data were compared by analysis of variance and Duncan's test was used for pairwise comparison.

RESULTS

A standard curve for the human plasma Fn was calculated from the average of the three assays per concentration, using the ELISA protocol described above. As can be seen in figure 1, the concentrations from 10 to 100 ng/ml show a linear response with a Spearman regression coefficient of 0.88.

Fig. 1. Standard curve for the different concentrations assayed. 
Standard curve for human olasma Fn concertrations between 10 and 250 ng/ml 
Each point represent the average value of the measurements from three 
different assays. (n=9) (r= 0,88)

 

The protein values were quantified for each of the 20 samples and each of the conservation methods (N=120). A 10% loss was observed in the samples kept in glass tubes for 24 hours at 4ºC and the loss was greater if the samples were kept in plastic tubes. For the frozen samples, despite being stored in glass tubes, the protein loss figures lay in a range of 40% to 60%, as may be observed in figure 2.

Fig. 2. Percentage protein loss according to conservation conditions.
Percentage Fn loss for each of the storage conditions, taking immediate 
quantification as 100%.
B: Samples kept for 24 hours at 4ºC in plastic tubes
C: Samples kept for 24 hours at 4ºC in glass tubes
D, E, F, G: Samples kept at -40ºC in glass tubes for 1, 2, 4 and 
8 weeks respectively

 

Table 1 shows the average values and 95% confidence intervals for each set of storage conditions. No significant differences were found between the protein levels measured immediately after collection and those obtained after being stored in a glass tube for 24 hours at 4ºC. Statistically significant differences were found between these two and all the other groups.

Table 1. Average values and confidential intervals for each of the conservation methods.
+, * Groups where statistically significant differences were found

 

DISCUSSION AND CONCLUSIONS

The results obtained showed high losses of immune-reactive immunoreactive Fn in saliva when stored by the usual conservation freezing and storage methods. Although this has been demonstrated for samples of different body fluids by other authors (26), the high percentage of losses in saliva is a fact that should be considered with a view to careful handling of the samples.

Simply keeping the samples in plastic or handling the saliva with plastic instruments favours protein loss, as the protein coats on the tube walls. Conservation for 24 hours in a glass tube at 4ºC causes minor losses but any of the other storage methods investigated causes very high losses.

Storage at -40ºC and subsequent thawing causes very high losses. These are probably related more to the thawing process than to the freezing of the samples in itself. It is also possible that the presence of protein degradation factors in the total saliva (27) and the well-known lability of the protein in the presence of water (28) result in the antibodies used for the assay being unable to recognise it. The protein's considerable capacity for binding to the cells and the extracellular matrix may be a factor in high losses during the thawing process.

One important fact is that the rabbit antibody allows binding to six fractions of the protein (230, 200, 110, 85, 75 and 65 KDa), as well as to the complete protein, whereas mouse monoclonal anti-serum only recognises the 230 KDa molecule, which is not present in the total saliva (29). This aspect is extremely important in relation to the protein quantification methodology, as it would not be possible to detect the protein with mouse monoclonal anti-serum.

In conclusion, saliva sample collection to quantify Fn should be performed with glass tubes, the samples should be processed no later than 24 hours after collection and during this time they should be stored at 4ºC. Freezing and thawing the samples causes losses of up to 60% of the initial values of the protein.

REFERENCES

1. Ruoslahti E. Fibronectin and its receptor. Ann Rev Biochem 1988;57:375-431.         [ Links ]

2. Tynelius-Bratthall G, Ericson D, Araujo HM. Fibronectin in saliva and gingival crevices. J Periodontal Res 1986;21:563-8.         [ Links ]

3. Llena MC, Cornudella R, Barrau F, Gutiérrez M, Martínez MR, Tres A. Fibronectina y hepatopatías: un nuevo parámetro en la valoración de la función hepatocelular. Rev Esp Enf Ap Digest 1986;70:169-71.         [ Links ]

4. Atance JC, Tres A, Martínez MA, Ortells JL, Llena-Puy MC. Fibronectina: estudio en gastroenteritis agudas. Re Esp En Ap Digest 1987;72:512-4.         [ Links ]

5. Mosher DF. Physiology of fibronectin. Ann Rev Med 1984;35:561-75.         [ Links ]

6. Ruoslahti E, Vaheri A. Interaction of soluble fibroblast surface (fs) antigen with fibrinogen and fibrin. Identity with cold insoluble globulin of human plasma. J Exp Med 1982;141:497-501.         [ Links ]

7. Vanden-Abbeele A, Courtois P, Pourtois M. The antiseptic role of saliva. Rev Belg Med Dent 1992;47:52-8.         [ Links ]

8. Babu JP, Simpson WA, Courtney HS, Beachey E H. Interaction of human plasma fibronectin with cariogenic and non-cariogenic oral streptococci. Infect Immun 1983;41:162-8.         [ Links ]

9. Ericson D, Tynelius-Bratthall G. Absorption of fibronectin from human saliva by strains of oral streptococci. Scand J Dent Res 1986;95:377-9.         [ Links ]

10. Sojar HT, Lee JY, Genco RJ. Fibronectin binding domain of P. Gingivalis fimbriae. Biochem Biophyis Res Commun 1995;216:785-92.         [ Links ]

11. Naito Y, Tohda H, Okuda K, Takazoe I. Adherence and hydrophobicity of invasive and noninvasive strains of Porphyromona Gingivalis. Oral Microbiology Immunol 1993;8:195-202.         [ Links ]

12. Loo CY, Wilcox MD, Knox KW. Surface-associated properties of Actinomyces strains and their potential relation to pathogenesis. Oral Microbiol Immunol 1994;9:12-8.         [ Links ]

13. Tajima K, Yu S, Tmaki N, Ueda H, Hori Y, Kanehisa J, et al. The mechanisms dental plaque and dental calculus formation by fibronectin. IV. Demonstration of fibronectin in dental calculus. J Jap Ass Periodontol 1984;26:701-9.         [ Links ]

14. Tynelius-Bratthall G. Crevicular and salivary fibronectin before and after gingivitis treatment. J Clin Periodontol 1988;15:283-7.         [ Links ]

15. Babu JP, Dabbous MK. Interaction of salivary fibronectin with oral streptococci. J Dent Res 1986;65:1094-9.         [ Links ]

16. Babu JP, Dean JW, Pabst MJ. Attachement of Fusobacterium nucleatum to fibronectin immobilized on gingival epithelial cells or glass coverlips. J Periodontol 1995;66:285-90.         [ Links ]

17. Murkami Y, Hanazawa S, Tanaka S, Iwahashi H, Kitano S, Fujisawa S. Fibronectin in saliva inhibits Porphyromonas Gingivalis fimbria-induced expression of inflammatory cytokine gene in mouse macrophages. FEMS Immunol Med Microbiol 1998;22:257-62.         [ Links ]

18. Llena MC, Motañana C, Forner L. Fibronectin levels in stimulated whole-saliva and their relationhip with cariogenic oral bacteria. Int Dent J 2000;50: 57-9.         [ Links ]

19. Sugano N, Tanaka H, Ito K, Murai S. Arg-Gly-Asp (RGD) peptides inhivit Streptococcus mitis to adherence to fibronectin. J Nihon Univ Sch Dent 1997; 39:154-5.         [ Links ]

20. Su H, Boackle RJ. Interaction of the envelope glycoprotein of human immunodeficiency virus with C1q and fibronectin under conditions present in human saliva. Mol Immunol 1991;28:811-7.         [ Links ]

21. Boackle RJ, Connor MH, Vasely J. High molecular weight non-immunoglo-bulin salivary agglutinins (NIA) bind C1Q globular heads and have the potential to activate the first complement component. Mol Immunol 1993;30:309-19.         [ Links ]

22. Tipton DA, Braxton SD, Dabbous MK. Role of salivary components as modulators of bleaching agent toxicity to human gingival fibroblastsin vitro. J Periodontol 1995;66:766-74.         [ Links ]

23. Zhan F, Wang H, Wang Z. Fibronectin in tissues and saliva of patients with oral lichen planus. Zhonghuan Kou Qiang Yi Xue Za Zhi 1996;31:235-7.         [ Links ]

24. Lyons AJ, Cui N. Salivary oncofetal fibronectin and oral squamous cell carcinoma. J Oral Pathol Med. 2000; 29: 267-70.         [ Links ]

25. Gómez-Lechón MJ, Castell JV. Measurement of fibronectin in human body fluids. J Clin Chem Clin Biochem 1986; 24:333-9.         [ Links ]

26. Lamberts BL, Pederson ED, Bial JJ, Tombasco PK. Fibronectin levels of unestimulated saliva from naval recruits with and without chronic inflammatory periodontal disease. J Clin Periodontol 1989;16:342-6.         [ Links ]

27. Gibbons RJ, Etherden I. Fibronectin-degrading enzymes in saliva and their relation to oral cleanliness. J Periodontal Res 1986;21:386-95.         [ Links ]

28. Gómez-Lechón MJ, Castell JV. Enzyme-linked immunosorbent assay to quantify fibronectin. Anal Biochem 1985;145:1-8.         [ Links ]

29. Kanehisa J, Doi S, Yamanaka T, Takeuchi H. Salivary fibronectin in man: an immunoblotting, radioimmunoassay and immunohistochemical study. Arch Oral Biol 1991;36:265-72.         [ Links ]