|
 |
| ORIGINAL ARTICLE |
|
| Year : 2004 | Volume
: 52
| Issue : 1 | Page : 51-56 |
|
Elevated free tear lactoferrin levels in leprosy are associated with type 2 reactions
E Daniel, M Duriasamy, Gigi J Ebenezer, Shobhana, Charles K Job
Department of Ophthalmology, Schieffelin Leprosy Research and Training Center, Tamil Nadu, India
Correspondence Address:
E Daniel
Department of Ophthalmology, Schieffelin Leprosy Research and Training Center, Tamil Nadu
India
 Source of Support: None, Conflict of Interest: None  | Check |
PMID: 15132380 
Purpose: To determine the association of demographics, leprosy and ocular characteristics with altered levels of lactoferrin in the tears of normal subjects and leprosy patients, and to detect the presence of antibodies to lactoferrin in these tear samples.
Method: We collected light-stimulated tears from 298 leprosy patients and an equal number of normal subjects using the glass capillary method. Free lactoferrin levels were estimated using ELISA and the presence of antibodies to lactoferrin was detected using the immuno-blotting method. Significant associations were looked for between tear lactoferrin levels and demographic characteristics, leprosy characteristics such as type of disease, duration of disease, reactions, deformity and bacterial load, and ocular complications, using chi-square and regression analysis.
Results: Tear lactoferrin levels with a mean (SD) of 2.55 (2.83)mg/ml in the control group were significantly different (P<0.000) from leprosy patients with a smean (SD) of 5.66 (7.21)mg/ml. Age showed an inverse correlation with tear lactoferrin levels in controls. Increased bacterial load, grade 2 leg deformity and Type 2 reactions were significantly associated (P<0.05) with increased tear lactoferrin levels. Type 2 reactions remained significantly associated ( P=0.01) on multiple regression analysis. Tear lactoferrin levels were not associated with gender, serum lactoferrin levels, Type 1 reactions, face patches, treatment status, orbicularis oculi weakness, lagophthalmos, ectropion, entropion, corneal opacity, cataract and iridocyclitis.
Conclusion: Age is inversely related to tear lactoferrin levels in normal subjects. Free lactoferrin levels in tears are significantly higher in leprosy patients compared with normal controls. Type 2 reactions in leprosy are significantly associated with elevated tear lactoferrin levels.
Keywords: Leprosy, tear lactoferrin levels, antibodies to tear lactoferrin, Type 2 reaction.
How to cite this article:
Daniel E, Duriasamy M, Ebenezer GJ, Shobhana, Job CK. Elevated free tear lactoferrin levels in leprosy are associated with type 2 reactions. Indian J Ophthalmol 2004;52:51-6 |
Lactoferrin is a glycoprotein present in exocrine secretions like milk, tears, nasal exudate, saliva, bronchial mucus, gastrointestinal fluids, cervicovaginal mucus and seminal fluid. It is also produced in polymorphonuclear leukocytes.[1] Several studies have shown that lactoferrin has various bactericidal effects. It has exhibited significant activity against Escherichia coli, Proteus mirabilis, Staphyloccocus aureus, Candida albicans and several other bacterial pathogens.[2],[3] Lactoferrin also exhibits immuno-modulatory activities. It is thought to promote the growth and differentiation of T lymphocytes and known to bind uniquely in the region of major histocompatability (MHC) proteins and the CD4 and CD8 determinants on T4 (helper) and T8 (suppressor) lymphocytes. It also appears to play a role in the regulation of cytokines and lymphokines.[4],[5] Lactoferrin levels have been investigated in the tears of diabetics,[6] HIV patients,[7] chronic hepatitis patients,[8] chronic meibonitis associated with acne rosacea[9] and in the tears of contact lens wearers developing giant papillary conjunctivitis.[10]
Lactoferrin levels are raised in lepromatous leprosy patients with Type 2 reactions.[11] This study was done to determine whether tear lactoferrin levels were associated with demography, leprosy and ocular characteristics. Although leprosy is a chronic infectious disease, it is also, to a great extent, considered an immunological disease. Mycobacterium leprae induces a complex humoral and cellular immune response in the host. Documented evidence exists that there is cross-reactivity between human lactoferrin and the 65 kDa heat shock protein of the leprosy bacilli.[12] Since the organism can form immune complexes with the antibody to lactoferrin, we also detected antibodies to lactoferrin in tear samples of patients and controls to determine any association between the presence or absence of antibodies raised against lactoferrin and variables that exist in leprosy and eye complications related to a host-immune response.
| Materials and Methods | |  |
Patients and Controls
Consecutive leprosy patients admitted as in-patients to the hospital and attending the out-patient department of the Schieffelin Leprosy Research and Training Center, conforming to certain stratifying criteria were invited to participate in the study. The stratification included (1) smear negative leprosy patients who were newly diagnosed, under treatment and released from treatment (RFT), (2) smear positive patients who were newly diagnosed, under treatment and RFT, (3) patients with past and present Type 1 and Type 2 reaction, and (4) patients with eye complications related to leprosy like lagophthalmos and iridocyclitis. An equal number of persons who did not have leprosy but belonged to the same socio-economic status (SES) served as controls. These were selected as much as possible to match the age and gender of patients. Individuals with any ocular abnormalities or any known systemic disease were not included in the control group.
After informed consent was obtained, each of the recruited patients and controls were asked to complete a demography questionnaire. The patients were then assessed for leprosy characteristics that included type of leprosy (categorised according to the Ridley and Jopling classification[13] and the World Health Organization (W.H.O.) treatment classification),[14] the approximate duration of the disease, skin smear positivity for acid fast bacilli, deformity grading (according to the W.H.O. classification),[15] presence or absence of Type 1 or Type 2 reaction, treatment status, relapse, presence or absence of face patch, extra clofazimine therapy and corticosteroid therapy. This was followed by an ocular examination which included specifically checking for madarosis, orbicularis oculi weakness, lagophthalmos, ectropion, entropion, corneal opacity, features of past or present iridocyclitis, naso-lacrimal duct patency and cataract.
Estimation of Free lactoferrin
Tears were collected from the right lower conjunctival sac of patients and controls by stimulating the eyes with a bright light source and using sterile glass capillaries. Tears were taken from the left eye of patients only if a local ocular condition specified as a stratifying criterion warranted such a collection. 5ml of blood was collected by venous puncture with sterile disposable syringes and the serum separated by centrifugation. The collected tears and serum were stored at -20°C till use. Tear and serum lactoferrin were quantified as per the protocol used by Kijlstra et al,16 with following modification. The reaction volume was scaled down to suit the microtitre plates and orthophenylene diamine with 30% hydrogen peroxide was used as substrate instead of ABTS (2-2' - Azino di - (3-ethyl-benzthiazoline - 6 - sulphonate). The competitive binding of anti-human lactoferrin to free lactoferrin in samples against bound lactoferrin was measured and the quantity calculated with the help of a standard graph. Briefly, all the wells except the blank were coated with AQ 100 µl of 0.05µg /ml human milk lactoferrin (Sigma) diluted in coating buffer for 1 hour at 37 °C. Washing of excess lactoferrin was done with 0.1% PBST (0.1% Tween-20 in phosphate buffered saline) thrice. Standards were included in each plate with serially diluted lactoferrin ranging from 0.1µg/ml to 0.0015625µg/ml. Lactoferrin standards and samples were assayed in triplicates. In test wells, 100µl samples of tear at four different dilutions (1:104 to 1:107) and serum 1: 10 and neat were added. Dilutions were made with incubation buffer (3% bovine serum albumin (BSA) in 0.1% PBST). To this 100ul of 1:2000 of rabbit anti human lactoferrin antibody (DAKO A/S Copenhagen, Denmark) prepared in 0.1 % PBST was added and incubated at room temperature on shaker for one hour to facilitate competitive binding. Plates were then washed thrice with 0.1% PBST. 100ul of goat anti rabbit - horseradish peroxidase conjugate (DAKO) diluted 1:2000 with 0.1 % PBST was added and incubated at room temperature for one hour. Excess conjugate was removed by washing thrice with 0.1%PBST. 100 l of freshly prepared Ortho-phenylene diamine (Sigma Chemicals Co, St. Louis, MO, USA ) with 30% H2O2 was added to each well and the plate left in darkness at room temperature for 30 minutes. The reaction was stopped with 100µl of 0.5M sulfuric acid and the Optical density read at 490nm.
Detection of antibody to lactoferrin
Cross reactivity of anti-65kDa antibody was confirmed against heat killed M leprae and purified human lactoferrin. Pooled serum from 10 patients with high bacterial index (BI) and Type 2 reactions was used as positive control. Dot immunoblotting protocol was designed and standardised using the positive controls and was based on a protocol followed in an earlier study.17 Briefly, 5µg of purified human milk lactoferrin mildly denatured with 1 N Sodium hydroxide was blotted to nitrocellulose membrane (Schleicher & Schuell, Inc, Keene, NH, USA) and allowed to air dry. To confirm cross reactivity with M leprae antigens, heat-killed bacilli at a concentration of 1 x 107 per ml was blotted and the membrane blocked with 100µl of 3% BSA in 0.3% Tween 20 in Tris buffered saline (TBST) to prevent nonspecific binding. Tears were diluted to 1:10,000 and incubated with the membrane-bound antigens for an hour at room temperature. Excess tear samples were washed thrice with 0.1% TBS. Antibodies bound to lactoferrin and M.leprae in the membrane were detected using anti-human IgG, IgA and IgM (DAKO) conjugated with alkaline phosphatase. Excess conjugate were washed and the colour developed with 5-Bromo-4-Chloro-3-Indolyl Phosphate/Nitro Blue Tetrazolium substrate. A bluish-violet color indicates the presence of antibodies to lactoferrin. A consensus of four independent observers for the development of colour for both lactoferrin and heat killed M.leprae was considered positive.
Results were entered into an Excel database and simple linear regression and the multiple linear regression analysis using different models were done using the STATA 7.0 package.
| Results | | |
Two hundred and ninety eight leprosy patients and an equal number of controls agreed to participate in the study. Their distribution by age and gender is given in [Table - 1]. The age of the patients and controls ranged from 15 to 79 years. There were 230 and 226 males in the patient and control group respectively. There was no significant difference in the age or gender of both groups. Regression of tear lactoferrin and age in the control group showed an inverse correlation with a coefficient (SE) of -0.03232 (0.01047) and a 95%C.I. of
-0.0529295 to -0.0117149 (P<0.002). There was no significant association of tear lactoferrin with gender, either in the controls or cases. Tear lactoferrin levels ranged from 0.08 to 27mg/ml with a mean (SD) of 2.55 (2.83) mg/ml in the control group. In leprosy patients, tear lactoferrin ranged from 0.2 to 51mg/ml with a mean (SD) of 5.66 (7.21) mg/ml [Figure - 1]. The concentration of tear lactoferrin in these two groups was statistically different (P<0.000). There were extreme outlying values (46mg/ml and 51mg/ml) in the tear lactoferrin levels of two leprosy patients. The estimation of lactoferrin in the tears obtained from these two patients was repeated twice by two independent observers with no change observed from the original values. We did not discard these two high values in leprosy patients during analysis.
Among leprosy patients, 212 belonged to the Multibacillary (MB) group and 86 to the Paucibacillary (PB) group. The approximate duration of disease ranged from 1 year to 51 years with a mean (SD) of 11.27 (13.3) years. Five (15%) were relapsed patients. Thirty-four patients had had only dapsone monotherapy, 166 had only multi-drug therapy (MDT), 44 had both dapsone and MDT and 54 newly diagnosed patients were yet to start anti-leprosy therapy. Twelve patients had classical face patches over the malar area, 65 patients had patches anywhere in the face and 221 patients were free of any face patch. A classical face patch was either an erythematous or a hypo-pigmented patch that occurred over the eyelids and/or over the malar area. The malar area was defined as an area comprising of 2cm on either side of a line drawn from the tragus to the lateral canthal point. 52 patients had a history of Type 1 reaction and 49 patients had signs of Type 1 reaction at the time of examination. 46 patients had a history of Type 2 reaction and 28 had Type 2 reaction at the time of tear collection. 54 patients had had extra clofazamine and 180 oral corticosteroids during the course of their disease. The deformity grading in all four limbs of the patients studied is given in [Table - 2]. 142 patients had Grade 2 deformity in one or more limbs, 46 had Grade 2 deformity in all the limbs and 84 had no deformity in any of the limbs. The average bacterial index (B.I.) for acid-fast bacilli in the routine skin smears ranged from 0 to 4.5 with a mean (SD) of 1.38 (1.65) at the time of diagnosis of leprosy. 143 patients were smear negative when they were diagnosed with leprosy. The highest B.I. at any one site ranged from 0 to 6 with a mean (SD) of 1.32 (1.80) in patients at diagnosis. The average BI ranged from 0 to 4.5 with a mean (SD) of 0.98 (1.44) at the time of tear collection. 178 patients were smear negative at the time of tear collection. The highest B.I. at any one site ranged from 0 to 6 with a mean (SD) of 1.32 (1.79) in patients at tear collection.
Thirty-nine patients had madarosis, 69 had orbicularis oculi weakness, 66 had lagophthalmos, 18 had ectropion, 3 had entropion, 27 had signs of past iridocyclitis, 9 had active iridocyclitis, 57 had corneal opacity, 96 had cataract and one patient had blocked naso-lacrimal ducts. Simple linear regression analysis results of tear lactoferrin with demographic, leprosy and ocular variables is given in [Table - 3]. Multiple linear regression between tear demography and all the variables of leprosy patients that showed significant associations in the simple linear regression analysis is given in [Table - 4]. Chi-square analysis was also done after categorising tear lactoferrin into two groups, one >20 mg/ml and the other with values of 20mg/ml or less. There was significant association (P<0.04) only with increasing bacterial index.
Antibody to lactoferrin was present in the tears of only 7 patients (2.35%). It was also present in 6 control eyes (2.01%). Antibody to lactoferrin was associated with increasing bacterial index, both at the time of diagnosis of the disease as well as nearer to the time of tear collection ( P<0.001). Antibody to lactoferrin was not associated with any other demographic, leprosy or ocular variables.
| Discussion | | |
Tear lactoferrin levels have been estimated in health and diseases in healthy subjects. In two different studies, tear lactoferrin was estimated at 2.2 ± 1.12mg/ml (38 subjects)[13] and 1.73 ± 0.46mg/ml (12 subjects).[18] In our study the free tear lactoferrin level was estimated at 2.55 ± 2.83mg/ml. The large standard deviation accompanying the mean free lactoferrin value demonstrates the wide variation in individual values and the extreme outlying values obtained in this sample of patients. The method of tear collection, whether by Schirmer's strips or by glass capillaries after tear gas stimulation, does not significantly alter the lactoferrin concentration in tears.[16] Although other protein concentrations are shown to change with progressively increasing stimuli, tear lactoferrin levels usually remain relatively constant.[19] While one study found age and gender did not alter levels of lactoferrin significantly,[16] another found an inverse correlation with age. We found a similar significant association (P<0.002), with a decrease of 0.32 ml/mg of tear lactoferrin for every 10 years' increase in age of the controls. Leprosy patients showed similar significant association (P<0.038).This association lost its significance when the two patients with the extreme outlying values were removed from the analysis (P>0.05). The two extreme values were present in young males but otherwise showed no relation to any of the variables.
Tear lactoferrin levels in systemic and ocular diseases have shown varying results. Tear lactoferrin level is reduced in chronic hepatitis C infection and in HIV positive asymptomatic patients.[7],[8] No correlation between lactoferrin content of tears could be demonstrated between normal subjects and diabetics.[6] In our study skin smear results for acid-fast bacilli, both at the time of diagnosis of the disease and the latest results closer to when the tears were collected, were available for all leprosy patients. In patients who showed the presence of acid-fast bacilli, particularly with the BI of 2+ and above, it is likely that the conjunctival and lid tissue had M.leprae in considerable quantities. Increased levels of mycobacteria were significantly associated with increasing levels of lactoferrin.
One other important, consistent, significant, association for increased tear lactoferrin was the Type 2 reaction, also known as Erythema Nodosun Leprosum (ENL). Type 2 reactions occur in lepromatous leprosy, especially in patients with a high load of M leprae . They are characterised by painful and tender nodules on the skin. Apart from the skin lesions they can also cause neuritis, iridocyclitis and orchitis. Immune complexes are formed in these reactions. Histopathologically there is dense infiltration of neutrophils. An earlier study has also shown that serum lactoferrin was raised considerably in lepromatous patients with Type 2 reactions.[11] In our study, patients with Type 2 reactions were 50 times more likely to have significantly raised lactoferrin levels than patients without active Type 2 reaction. The raised levels of lactoferrin in the tears of patients with Type 2 reactions may be a reflection of the neutrophilia present in these patients. It is known that lactoferrin is released from neutrophil granules during inflammation. Lactoferrin in tears is secreted by the lacrimal gland and considered to be a regulated lacrimal gland protein. It is possible that patients with Type 2 reactions also have some amount of dacryoadenitis and the inflammation of the lacrimal glands during the reaction may be responsible for the raised tear lactoferrin levels. Increased levels of tear lactoferrin were not significantly associated in patients with past history of Type 2 reactions suggesting that raised tear lactoferrin levels return to lower normal levels after the acute episode. It is possible that tear lactoferrin has the potential to be a predictive inflammatory marker of Type 2 reactions. Longitudinal follow-up of multi-bacillary leprosy patients and estimation of tear lactoferrin at regular intervals in these patients, including those developing an acute phase of Type 2 reaction, could substantiate this hypothesis.
Iridocyclitis is considered a localised form of Type 2 reaction. It could occur in isolation in lepromatous leprosy patients or as part of the other manifestations of Type 2 reactions. Elevated levels of lactoferrin were not associated with past or present iridocyclitis. Raised lactoferrin levels have been documented in arthritic joint exudates[20] and have been suggested to play a role in inflammations. However, iridocyclitis, although an inflammatory reaction, does not alter the levels of lactoferrin in tears. This is probably because the inflammation is intraocular and not related to the lacrimal glands or the tissues that are directly in contact with the tear film. Local inflammations like chronic meibominitis associated with acne rosacea,[9] contact lens wearers developing giant papillary conjunctivitis,[10] cataract surgery,[21] and keratoconjunctivitis sicca[22] reduce tear lactoferrin levels. Lactoferrin levels were raised in patients who had grade 2 leg deformities. Elevated lactoferrin is demonstrated in neurodegenerative diseases like Alzheimer's disease, Down's syndrome, amyotrophic lateral sclerosis, Parkinsonism-dementia complex of Guam and sporadic amyotrophic lateral sclerosis.[23] An excessive accumulation of lactoferrin, as well as transported iron and aluminum, may lead to a cytotoxic effect resulting in the formation of intracellular lesions and neuronal death. This could account for the association of grade 2 deformities with raised free tear lactoferrin levels.
We did not find a significant difference in the amounts of antibodies to lactoferrin in the tears of leprosy patients as compared to controls. The only variable that was associated significantly with the antibodies to lactoferrin among leprosy patients was increased presence of the leprosy bacilli in the body of these patients. Neither leprosy reactions nor local ocular inflammations showed any relationship to the presence of antibodies to lactoferrin in tears.
| Acknowledgement | | |
:
Dr. Brennen's laboratory monoclonal antibodies for 65kDa M.leprae antigens. Mr Yowan, Dr Sheena Koshy and Dr Kala Gopalakrishnan recruited and examined the leprosy patients.
| References | | |
| 1. | Weinberg ED.Human lactoferrin: A novel therapeutic with broad spectrum potential. J Pharm Pharmacol 2001;53:1303-10.  |
| 2. | Vorland LH, Ulvatne H, Andersen J, Haukland HH, Rekdal O, Svendsen JS and Gutteberg TJ. Antibacterial effects of lactoferricin B. Scand J Infect Dis. 1999; 31:179-84. |
| 3. | Arnold RR, Brewer M and Gauthier JJ. Bactericidal activity of human lactoferrin : Sensitivity of a variety of micro-organisms Infect Immun 1980;28:893-98. |
| 4. | Britigan BE, Serody JS, Cohen MS. The role of lactoferrin as an anti-inflammatory molecule. Adv Exp Med Biol. 1994; 357:143-56. [ PUBMED] |
| 5. | Lonnerdal B, Iyer S. Lactoferrin. Molecular structure and biological function. Annu Rev Nutr. 1995;15:93-110. [ PUBMED] [ FULLTEXT] |
| 6. | Jensen OL, Gluud BS, Birgens HS. The concentration of lactoferrin in tears of normals and of diabetics. Eur J Ophthalmol 2000;10:11-14. |
| 7. | Comerie-Smith SE, Nunez J, Hosmer M, Farris RL. Tear lactoferrin levels and ocular bacterial flora in HIV positive patients. Adv Exp Med Biol 1994;350:339-44. [ PUBMED] |
| 8. | Abe T, Nakajima A, Matsunaga M, Sakuragi S, Komatsu M. Decreased tear lactoferrin concentration in patients with chronic hepatitis C. Br J Ophthalmol 1999;83:684-87. [ PUBMED] [ FULLTEXT] |
| 9. | Kiratli H, Irkec M, Orhan M. Tear lactoferrin levels in chronic meibomitis associated with acne rosacea. Eur J Ophthalmol 2000;10:11-14. [ PUBMED] |
| 10. | Velasco Cabrera MJ, Garcia Sanchez J, Bermudez Rodriguez FJ. Lactoferrin in tears in contact lens wearers. CLAO J 1997;23:127-29. [ PUBMED] |
| 11. | Parkash O, Girdhar BK and Sengupta U. Serum lactoferrin in lepromatous leprosy patients Lepr Rev 1993;64:295-301. |
| 12. | Esaguy N, Aguas AP, Embden JDA, Silva MT. Mycobacteria and human auto-immune disease: Direct evidence of cross-reactivity between human lactoferrin and the 65-kilodalton protein of tubercle and leprosy bacilli. Infect Immun 1991;59:1117-25. |
| 13. | Ridley DS and Jopling WH. Classification of leprosy according to immunity: A five-group system. Int J Lepr 1966;34;255-73. |
| 14. | WHO Expert Committee on Leprosy. Geneva: World Health Organisation Tech Rep Ser 1988;768;1-51. |
| 15. | WHO Study Group. Chemotherapy of leprosy for control programs. Geneva: World Health Organization Tech Rep Ser 1982;675:21-26. |
| 16. | Kijlstra A, Jeurissen SHM, Koning KM. Lactoferrin levels in normal human tears. Br J Ophthalmol 1983;67:199-202. |
| 17. | Locht H, Skogh T, Kihlstrom E. Antilactoferrin antibodies and other types of anti-neutrophil cytoplasmic antibodies (ANCA) in reactive arthritis and ankylosing spondilytis. Clin Exp Immunol 1999;117:568-73. [ PUBMED] [ FULLTEXT] |
| 18. | Ballow M, Donshik PC, Rapacz P, Samartino L. Tear lactoferrin levels in patients with external inflammatory ocular disease. Invest Ophthalmol Vis Sci 1987 ;28:543-45. [ PUBMED] |
| 19. | Fullard RJ, Tucker DL. Changes in human tear protein levels with progressively increasing stimulus. Invest Ophthalmol Vis Sci 1991;32:2290-301. [ PUBMED] |
| 20. | Malmquist J, Thovell JI, Wollheim FA. Lactoferrin and lysozyme in arthritic exudates Acta Med Scand 1977;202:313-18. |
| 21. | McCollum CJ, Foulks GN, Bodner B, Shepard J, Daniels K, Gross V, Kelly L, Cavanagh HD. Rapid assay of lactoferrin in keratoconjunctivitis sicca. Cornea 1994 ;13:505-508. [ PUBMED] |
| 22. | Kumar R, Parmar IP, Chhillar N, Lal H. Tear lactoferrin concentration during postoperative ocular inflammation in cataract surgery. Acta Ophthalmol Scand 1997;75:142-44. [ PUBMED] |
| 23. | Leveugle B, Spik G, Perl DP, Bouras C, Fillit HM, Hof PR. The iron-binding protein lactotransferrin is present in pathologic lesions in a variety of neurodegenerative disorders: A comparative immunohistochemical analysis. Brain Res 1994;650:20-31. |
[Figure - 1]
[Table - 1], [Table - 2], [Table - 3], [Table - 4]
| This article has been cited by | | 1 |
Role of lactoferrin in the tear film |
|
| Flanagan, J.L., Willcox, M.D.P. | | Biochimie. 2009; 91(1): 35-43 | | [Pubmed] | | | 2 |
Proteomic analysis of dog tears for potential cancer markers |
|
| de Freitas Campos, C., Cole, N., Dyk, D.V., Walsh, B.J., Diakos, P., Almeida, D., Torrecilhas, A., (...), Willcox, M.D.P. | | Research in Veterinary Science. 2008; 85(2): 349-352 | | [Pubmed] | |
|
 |
|
|
|
Search Pubmed for