Salivary antimicrobial proteins in patients with Crohn's disease

Vol. 76 No. 3

L MEDI

November 1993

Editor: If. Dean Millard

Salivary antimicrobial proteins in patients with Crohn’s disease Bo Sundh, DDS,” Ingegerd Johansson, DDS, Qdont Dr,b Claes-Goran Emilson, DDS, Odont Dr,a Svante Nordgren, MD, Med Dr,C and Dowen Birkhed, DDS, Odont Dr,a Goteborg and Umei, Sweden UNIVERSITY

OF GGTEBORG

AND UNIVERSITY

OF UMEA

The aim was to study the level of salivary proteins with antimicrobial properties in persons with Crohn’s disease. Twenty-five patients were recruited, ‘I3 with ongoing symptoms (acute group) and 12 free of clinical signs of the disease at the time of the investigation (nonacute group). A control group matched to the nonacute group was also included in the study. Unstimulated and stimulated whole saliva samples were collected, and the secretion rates estimated. Unstimulated saliva was analyzed for concentrations of total protein, peroxidase, thiocyanate, slgA, lactoferrin, lysozyme, and for specific bacteria aggregation ability. Numbers of mutans streptococci and lactobacilli in saliva were determined, and dental caries status was examined. No differences were found among the groups regarding salivary flow rate, total protein, or any of the antimicrobial proteins. However, three patients with Crohn’s disease had no detectable slgA in saliva compared with none in the control group. The lactobacillus count and the number of decayed tooth surfaces were higher in the nonacute group than in the control group. (ORAL SURC ORAL MED ORAL PATHOL 1993;76:564-9)

Crohn’s disease is characterized by chronic segmental granulomatous inflammatory involvement of the intestine. The disease, first described in 1932 by Crohn, Ginzburg, and Oppenheimer’ and referred to as “regional ileitis,” might affect the whole gastrointestinal tract from the mouth to the anus. Fully developed there is a transmural inflammation of the gut with hypertrophic nodular mucosa and deep longitudinal ulcers. The disease may be complicated by abscesses, fistulas, and sinus formations, which may lead to stenosis, adhesions, and frequent episodes of bowel obstruction. Manifestations in the oral mucosa may also appear.2 In Sweden, the incidence of Crohn’s disease is 6 to 7 cases per 100,000 inhabitants, and the most prevalent age of diagnosis is 15 to 30 years.3 Supported by the Swedish Dental Association and the Faculty of Odontology, University of Giiteborg. aDepartment of Cariology, Faculty of Odontology, University of Goteborg. bDepartment of Cariology, Faculty of Odontology, University of Umea. CDepartment of Surgery, Sahlgren’s Hospital, University of Giiteberg. Copyright @ 1993 by Mosby-Year Book, Inc. 0030-4220/93/$1.00 + .lO 7/13/50086

564

Common symptoms associated with Crohn’s disease are abdominal pain, diarrhea, weight loss, and anorexia. These symptoms are often paralleled by malabsorption, malnutrition, and anemia. Symptomatic relief may be achieved by treatment with corticosteroids, sulfasalazine, and, in some cases, by parenteral nutrition or a low-fat diet. 4, 5 Surgery, with resection of the diseased intestinal segment, will be required for the majority of patients. The disease has a strong tendency to recur, and repeated resections are often needed. In patients with inflammatory involvement of the large bowel, proctocolectomy and ileostomy have to be performed. Patients with Crohn’s disease may, in acute periods, complain of dry mouth and later also of an increased caries activity. However, few studies have been published in this field even if there are indications from the literature that persons with Crohn’s disease have a higher prevalence of dental caries than healthy controls.6, 7 This is particularly the case in patients in whom an extensive small bowel resection has been performed.* High levels of cariogenic bacteria in saliva, such as mutans streptococci and lactobacilli, have also been reported. 7,9 However, the salivary flow rates and the concentrations of calcium, phosphate,

Sundh et al.

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume

565

76, Number 5

Table I. Inter alia and medical history of the 25 recruited patients with Crohn’s disease Acute group (n = 13) Mean Men/Women Age (years) Number of years since diagnosis Number of bowel resections Small bowel length (percentage of total small bowel)

Nonacute group (n = 12) Range

Mean

23-72 o-41 O-6 50-100

46 21 5 30

716 38 11 2 92*

Range

814 26-69 11-31 l-10 25-50

*8 of the 13 patients had not had small bowel resections.

and total protein, as well as the activity of amylase in chewing-stimulated whole saliva, were similar in a group of patients with chronic Crohn’s disease compared with population reference values.7 Salivary proteins with antibacterial properties have been claimed tlo influence the caries process. To date, there is limited knowledge on the role of salivary proteins in vivo, but a large number of in vitro studies show that they may reduce bacterial metabolism as well as adhesion of oral microorganisms.” The aim of the present investigation was to study the level of salivary proteins with antimicrobial properties in patients with Crohn’s disease. MATERIAL Subjects

AND MIETHODS

Twenty-five consecutive patients with Crohn’s disease, attending the Department of Surgery, Sahlgren’s Hospital, GGteborg, Sweden, were recruited for the study. They formed two separate cohorts. The first, the “acute group” (n = 13), included persons with acute disease and with surgical treatment planned within a short period of time. These patients had no or only minor (< 50 cm) earlier loss of their small intestine. At the time of the planned surgery, segments (30 to 100 cm) of the small intestine were extirpated. The second cohort, the “nonacute group” (n = 12), had no clinical signs of active inflammatory disease at the time of thle investigation. These patients had a long history of Crohn’s disease and had, by surgical treatment, lost 50% to 80% of the small bowel and, occasionally, al.so the colon. An age- and sex-matched control group to the nonacute group (n = 12) was selected from he,althy persons, referred to the Department of Oral S’urgery, Faculty of Odontology, Giiteborg, mainly for third molar surgery. Data on age, gender, and medical history of the acute and nonacute group are presented in Table I. In the control group, 8 men and 4 women with a mean age of 46 years, (range, 26 to 69 years), participated.

Saliva collection

Duplicate samples of unstimulated and stimulated whole saliva from each participant were collected within 3 days of each other. In the acute group, follow-up samples were taken in 10 of the 13 patients 3 to 9 months after the surgical treatment. The subjects were told not to drink, eat, or smoke 1 hour before saliva collection. The sampling took place between 9 AM and 3 PM . For each person, the second samples were always collected at approximately the same time of the day as the first. Unstimulated whole saliva was drooled for 10 minutes, and stimulated whole saliva was collected by expectoration of saliva while chewing on an 1-gm piece of paraffin wax for 5 minutes. The saliva samples were collected into graded, ice-chilled test tubes. The flow rates of both unstimulated and stimulated whole saliva were then calculated and expressed as milliliter/ minute. Volumes of unstimulated saliva, needed for the various protein analyses were immediately pipetted into separate test tubes and stored at -70 “. The samples, frozen on carbon dioxide ice, were transported by air to the University of Umd where the analyses were carried out within 3 months after arrival. Analyses of salivary whole saliva

proteins

in unstimulated

The concentrations of total protein’ 1were assessed, with bovine serum albumin (Sigma Chemical Co, St Louis, MO.) used as a standard. The activities of salivary peroxidase were measured, with pyrogallol used as a chromogen,12 and the concentration of thiocyanate,13 using iron (III) chloride (Merck, Darmstadt, Germany) as a standard. Total IgA and lactoferrin concentrations were measured with Enzyme-Linked ImmunoSorbent Assay technique. For determination of sIgA, a-chain specific goat-anti-human-serum-IgA (No. 109-005-011, Jackson Immuno Research, West Grove, Pa.) was used as a primary antibody, and the same antibody with conjugated alkaline phosphatase

566 Sundh et al.

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY November 1993

was used as a secondary antibody. Human serum protein calibrator (No. X908, Dakopatts, Glostrup, Denmark) was used as a standard, For lactoferrin analyses, rabbit-anti-human-lactoferrin (No. A186, Dakopatts) was used as a primary antibody and antihuman lactoferrin PEROX (No. P062, Bindingsite, Birmingham, England) as a secondary antibody. Human colostral lactoferrin (No. L488 1, Sigma Chemical Co) was used as a standard. The lytic activity of lysozyme was determined on micrococcus diffusion plates made of agarose gel, pH 6.3, with Micrococcus lysodeikticus.14 Chicken egg white lysozyme (No. L-7001, Grade III, Sigma Chemical Co) was used as a standard. The activity of chicken egg white lysozyme was compared with human urinary lysozyme, which is used in the lysozyme kits manufactured by Kallestad (Chaska, Mich.). Chicken egg white lysozyme was found to give four times higher activities than urinary lysozyme. The activities of lysozyme were therefore corrected by a factor of four. The aggregation ability of BAGP, a glycoprotein with the ability to aggregate Streptococcus mutans, strain TH 16, serotype c, was followed spectrophotometrically for 60 minutes.15 Microbiologic

analyses

One milliliter of the stimulated whole saliva was transferred to a bottle with transport medium (VMG II, Janke and Kunkel, Staufen, Germany) for cultivation. The samples were dispersed on a Whirlimixer, serially diluted in 0.05 mol/L phosphate buffer, and 0.25 ~1 were spotted, in duplicate, on MSB agar16 and Rogosa selective Lactobacillus agar17 for cultivation of mutans streptococci and lactobacilli, respectively. All samples were processed within 24 hours after collection. Registration

of dental

caries

Recordings of decayed, missing, and filled tooth surfaces (DMFS) and teeth (DMFT) were made by one of us (B.S.) under optimal clinical conditions. Mirror, explorer, and four posterior bite-wing radiographs were used. Incipient and macroscopic caries defects were diagnosed separately on each tooth surface (DS) on the basis of clinical as well as radiographic findings. In addition to the caries examination, a history of drug consumption including the intake of antibiotics during the previous 2 months was taken. Statistical

analyses

Because no differences were found between the duplicate saliva samples, their mean values were used for

the statistical analyses. Bacteriologic data were logarithmically transformed to improve normality. The differences in the three groups were analyzed with one-factor analysis of variance, followed by a twotailed t test. The t test was applied when the analysis of variance test could reject the null-hypothesis that all groups were samples from the same population. Bacteriologic data were tested nonparametrically with the Mann-Whitney U-test. P < 0.05 was considered statistically significant. RESULTS Medication

Eight of the patients with Crohn’s disease, four in each group, reported to be free of medication at the time of the investigation. Thus 17 of the 25 patients used some kind of drug. The most common were those with an antidiarrheal effect, such as codeine-phosphate, cholestyramine, and loperamide, which were used by 11 patients. None in the nonacute and three in the acute group had an ongoing medication with sulfasalazine, corticosteroids, or other anti-inflammatory agents. Four of the 12 patients in the nonacute and none in the acute group reported regular intake of vitamins or minerals, such as vitamin Brz and magnesium. The hospital records showed, however, that all patients received Brz injections on a regular basis. None of the persons in the control group had any medication or disease at the time of saliva collection. Salivary flow

rate

The mean flow rate for unstimulated and stimulated whole saliva did not differ among the three groups (Table II). Two persons in the acute group, two in the nonacute group, and one in the control group had low values for unstimulated saliva (< 0.1 ml/minute) at one of the two sampling occasions. One patient in the nonacute group had low flow rate for stimulated saliva (< 0.7 ml/minute). Salivary

antimicrobial

proteins

Table II shows the concentration of total protein and the concentrations or activities of the various antimicrobial proteins in unstimulated saliva in the three studied groups. No significant differences were found among the groups for any of the variables. However, two patients in the acute group and one in the nonacute group had no detectable amounts of sIgA in their saliva at either of the two sampling occasions. None in the control group showed any deficiency in this respect. When the values from the follow-up saliva samples, collected 6 to 9 months after surgery in the acute

Sundh et al.

ORAL SURGERYOR,4L MEDICINE ORAL PATHOLOGY Volume 76, Number 5

Table II. Flow rates of stimulated and unstimulated

saliva and content of antimicrobial

Acute group (n = 13) Stimulated saliva Flow rate (ml/min) Unstimulated saliva Flow rate (ml/min) Total Iprotein (mg/L)

Nonacute group

1.9 1.5-2.3

1.9 1.3-2.4

2.1 1.5-2.1

0.33 0.20-0.46

0.26 0.15-0.37

0.35 0.23-0.47 838 637-1,040 0.07 0.03-0.12 237 165-310 22 16.9-26.7 42 28-56 2.6

1,194 879-1,509 0.11

426-1,480

BAGP (m-value?) s&A h/L) Lactoferrin

(mg/L)

Lysozyme (mg/L) SCN- l(mmol/L) SPO (AA400//ml/min)

proteins Control group* (n = 12)

(n = 12)

953 0.09 0.03-0.14 181 56-305 21 O-41.2 62 23-100 2.1 1.1-3.0 2.1

0.07-0.14 289 199-379 28

11.4-45.0 56 37-15 2.9

0.9-3.2

567

1.9-3.8

1.9-3.2

3.3 2.5-4.1

2.6 2.2-3.1

Data are expressedas means and 95% confidence intervals *Matched to the noaacute ;group. PAccording to Ericson et al.”

Table Ill. Colony-forming whole saliva

units (CFU X 103/ml saliva) of mutans streptococci and lactobaccilli Acute group (n = 13)

Mutans streptococci Lactobacilli

2,973 363-5,582

155 o-395

!

Nonacute group (n = 12)

I

in stimulated

Control group* (n = 12)

9,130t

1,297

O-20,237 448$

O-2,858 28 O-640

O-928

Data are expressedas means and 95% confidence intervals calculated on logarithmically transformed data. *Matched to the nonacute group. ip = 0.057 $p < 0.001 compared with the control group using Mann-Whitney U-test.

group (n = lo), were compared with the baseline values, no signific,ant differences were found for any of the salivary components studied. Cariogenic microorganisms The numbem of Icolony-forming units of mutans streptococci and lactobacilli per milliliter of saliva are given in Table III. The nonacute group had slightly higher values for mutans streptococci (p = 0.057) and significantly higher numbers of lactobacilli (JJ < 0.001) compared with the control group. Dental caries The prevalence of dental caries, given as mean values of DMFS AND DMFT, is presented in Table IV.

The mean values of DS, including both incipient and macroscopic caries defects, are also presented. The nonacute group had significantly higher DS value than the control group (p < 0.01). Otherwise, no differences were found among the three groups. DISCUSSION The main objective of this investigation was to study the salivary proteins with antimicrobial properties in patients with Crohn’s disease. Thus the concentrations of total protein, secretory IgA, and lactoferrin and the activities of lysozyme, salivary peroxidase, and an agglutinin (BAGP) in unstimulated whole saliva were compared between diseased and healthy groups. The number of participants was small

568

Sundh et al.

ORAL SURGERY ORAL MEDICINE

ORAL PATHOLOGY

November 1993 Table IV.

Caries prevalence (DS, DMFS, and DMFT) Acute group (n = 13)

Nonacute group

Control group*

(n = 12)

(II = 12)

3.2 1-5 48 26-70 17 13-22

7.41 3-12 80 60-99 22 20-25

l.6 l-2 69 48-89 22 19-26

Decayed tooth surfaces (DS) Decayed, missing, and filled surfaces (DMFS) Decayed, missing, and filled teeth (DMFT) Data expressedas means and 95% confidence intervals. *Matched to the nonacute group. tp < 0.01 compared with the control group.

because of the strict selection criteria. It was noted, however, that the two groups with Crohn’s diseasehad larger variations for several of the analyzed variables (sIgA, SPQ, and lactoferrin) than the control group, mainly becauseof lower minimum values (Table II). One may also notice that the average value for sIgA was slightly lower for the acute group than for the nonacute group and the control group. Two patients in the acute group and one in the nonacute group had no detectable amounts of sIgA in their saliva. It has been reported that the prevalence of IgA immunodeficiency is 1 per 600 healthy personsin Sweden and 1 per 200 hospitalized patients.‘* IgA-immunodeficient patients have no IgA secreted in their saliva. We found that 3 of the 25 patients with Crohn’s disease lacked salivary IgA. This would correspond to 24 per 200 (or 72 per 600 persons). The information on the relationship between Crohn’s disease and IgA levels in exocrine secretions is conflicting. An increased number of IgA-producing cells has been found in labial salivary glands of Crohn’s patients compared with healthy controls. l9 It has beenreported, however, that a high disease activity leads to decreased concentration of IgA in parotid saliva,2 whereas others have found elevated IgA levels in stimulated whole saliva and no difference in parotid saliva.20It is interesting to notice that a seriesof studies on patients with Crohn’s disease report impaired secretion of the dimeric secretory IgA from intestinal mucosal cells.21-24These findings support the hypothesis that the secretion of IgA is decreasedin Crohn’s disease and thereby also the mucosal defense.When an association between Crohn’s disease and sIgA in whole saliva is to be evaluated, it is of importance to consider contamination by monomeric serum-derived IgA. We have in the present study analyzed total IgA in uncentrifuged resting whole saliva, where the monomer fraction is reported to be 13% to 17% depending on the state of the gingiva.25 This would correspond to a contribution of approximately 20 to 50 mg/L sa-

liva and thus slightly obscure an even more impaired secretion of sIgA. The antimicrobial mechanisms of the proteins analyzed in the present study may all be related to metabolic processesin oral microorganisms or to their adhesion to surfaces in the mouth. Even if not statistically significant, there was a tendency (p = 0.057) that the nonacute group had higher numbers of mutans streptococci in their saliva than the matched control group. This is in accordance with an earlier study that showed that patients with Crohn’s disease had higher numbers of these cariogenic microorganisms than adults living in the samearea.7The present study also showed that there was a significant difference between the salivary levels of lactobacilli, with the higher counts for the nonacute group compared with the controls. These observations, which are in agreement with other studies9 indicate that the patient with a long history of Crohn’s disease and major loss of the small intestine as a result of disease progression may harbor high numbers of cariogenic microorganisms in the mouth. Patients with Crohn’s diseasehave beenreported to have higher caries prevalence than healthy controls.7, s This was supported by the findings in the present study that the DS score was significantly higher in the group with chronic Crohn’s diseasethan in the age- and sex-matched control group. It must be remembered, however, that dental caries is a multifactorial diseasein which the effects of the two main initiating factors (dental plaque and fermentable carbohydrates) are modified by various endogenic and exogenic factors, such asfluoride exposure,saliva, and oral muscular activity. These factors must all be kept in mind if variation in caries development between personsis to be fully understood. The reason for the increased caries prevalence in patients with Crohn’s diseaseis unclear, but it is probably enhanced by the dietary regimen induced by the disease.4T 5 The change of food habits may result in frequent eating to

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 76, Number 5

maintain body weight and in a specially fat-reduced diet to prevent diarrhea. The energy normally provided by fat is often replaced by low-molecularweight carbohydrates, especially sucrose. This increased exposure to fermentable carbohydrates is likely to contribute to the increased caries risk. It has earlier been reported that patients with Crohn’s diseaseseemto have normal stimulated salivary flow rate and buffer capacity.7 This was confirmed in the present study that found that the patients in both the acute and nonacute groups had similar secretion rates of unstimulated and stimulated whole saliva asithe healthy controls. These values are also in accord with the variations found in a healthy population. 26None of the patients with Crohn’s diseaseincluded in this study had active diarrhea, which may have an effect on the salivary secretion rates. CONCLUSION

The conclusion made from this study is that patients with chronic Crohn’s disease have an increased risk to develop dental caries. However, this increased risk (cannotbe associatedwith impairments in the protection provided by various salivary antimicrobial proteins. Other risk factors, such as frequent exposure to fermentable carbohydrates and increased numbers of calriogenic microorganisms, are therefore more likely to explain the enhanced caries development. REFERENCES 1. Crohn BB, Ginzburg L, Oppenheimer GD. Regional ileitis: a pathological and clinical entity. JAMA 1932;99:1323-9. 2. Basu MK, Asquith P, Thompson RA, Cooke WT. Oral manifestations of Crohn’s disease. Gut 1975;16:249-54. 3. Lindberg E, JBrnerot G. The incidence of Crohn’s disease is not decreasing in Sweden. Stand J Gastroenterol 1991;26:495500. 4. Martini G, Brandes J. Increased consumption of refined carbohydrates in patients with Crohn’s disease. Klin Wochenschr 197&54:367-‘il. 5. Andersson H, Bosaeus I, Hellberg R, Hulten L. Effect of a low-fat diet a,nd antidiarrhoeal agents on bowel habits after excisional bowel surgery for classical Crohn’s disease. Acta Chir Stand 1982;148:285-90. 6. Rooney TP. Dental caries prevalence in patients with Crohn’s disease. ORALSURC ORALMEDORALPATHOL~~~~;~~:~~~-~. 7. Sundh B, Emilsson CG. Salivary and microbial conditions and dental health in patients with Crohn’s disease: a 3-year study. ORALSURGORALMEDORALPATHOL 1989;67:286-90. 8. Sundh B, H&&n L. Oral status in patients with Crohn’s disease. Acta Chir Stand 1982;148:531-4.

Sundh et al.

569

9. Bevenius J. Caries risk in patients with Crohn’s disease: a pilot study. ORAL SURC ORAL MED ORAL PATHOL 1988;65:304-7. defense factors in human sa10. Tenovuo J. Nonimmunoglobulin liva. In: Tenovuo J, ed. Human saliva: clinical chemistry and microbiology, Vol II. Boca Raton, Florida: CRC Press Inc, 1989:55-91. 11. Spector T. Refinement of the Coomassie blue method of protein quantitation. Analyt Biochem 1978;86:142-6. 12. Gothefors L, Marklund S. Lactoperoxidase activity in human milk and in saliva of newborn infants. Infect Immun 1975; 11:1210-5. 13. Pruitt KM, Adamson M, Arnold R. Lactoperoxidase binding to streptococci. Infect Immun 1979;25:304-9. 14. Osserman EF, Lawlor DP. Serum and urinary lysozyme (muramidase) in monocytic and monomyelocytic leukemia. J Exp Med 1966;124:921-51. 15. Ericson T, Pruitt K, Wedel H. The reaction of salivary substances with bacteria. J Oral Path01 1975;4:307-23. 16. Gold OG, Jordan HV, van Houte J. A selective medium for Streptococcus mutans. Arch Oral Biol 1973;18:1351-64. 17. Rogosa M, Mitchell JA, Wiseman RF. A selective medium for the isolation and enumeration of oral lactobacilli. J Dent Res 1951;30:682-9. 18. Hansson LA, Wigzell H, eds. Immunology. Stockholm: Almqvist & Wiksell 1988: part II, 64. 19. Crama-Bohbaoth G, Bosman FT, Vermeer BJ, et al. Immunohistological findings in lip biopsy specimens from patients with Crohn’s disease and healthy subjects. Gut 1983;24:202-5. 20. Crama-Bohbaoth G, Lems-van Kan P, Weterman IT, Biemond I, Pena AS. Immunological findings in whole and parotid saliva of patients with Crohn’s disease and healthy controls. Dig Dis Sci 1984;29:1089-92. 21. Wu KC, Mahida UR, Priddle JD, Jewel DP. Immunoglobulin production by isolated intestinal mononuclear cells from patients with ulcerative colitis and Crohn’s disease. Clin Exp Immunol 1989;78:37-41. 22. MacDermott RP, Nash GS, Nahm MH. Antibody secretion by human intestinal mononuclear cells from normal controls and inflammatory bowel disease patients. Immunol Invest 1989;18:449-57. 23. O’Mahony S, Barton JR, Crichton S, Ferguson A. Appraisal of gut lavage in the study of intestinal humoral immunity. Gut 1990;31:1341-4. 24. Marteau P, Colombel JF, Nemeth J, Vaerman JP, Dive JC, Rambaud JC. Immunological study of histologically noninvolved jejunum during Crohn’s disease: evidence for reduced in vivo secretion of secretory IgA. Clin Exp Immunol 1990; 80:196-201. 25. Brandtzaeg P, Fjellanger I, Gjeruldsen S T. Human secretory immunoglobulins, I: salivary secretions from individuals with normal or low levels of serum immunoglobulins. Stand J Haematol 1990;12:970-3. 26. Heintze U, Birkhed D, Bjijrn H. Secretion rate and buffer effect of resting and stimulated whole saliva as a function of age and sex. Swed Dent J 1983;7:227-38. Reprint requests: Bo Sundh, DDS Faculty of Odontology Medicinaregatan 12 S-4 13 90 Gateborg Sweden