35
The effect of Giardia lamblia trophozoites on lipolysis in vitro P. KATELARIS*, F. SEOWWM. NGU Gastroenterology Unit, Concord Hospital, Sydney, Australia 2139 (Received 6 November 1990; revised 7 February 1991; accepted 11 February 1991)
SUMMARY
Infection with Giardia lamblia often causes only minor mucosal changes to the small intestine yet frank fat malabsorption may still occur. Some evidence suggests abnormal pancreatic exocrine function in subjects with giardiasis although the mechanism and significance of this is unclear. Studies were conducted in vitro to determine the effect of G. lamblia trophozoites or culture filtrates from the organism on lipolysis of triglyceride by porcine pancreatic lipase. Live trophozoites significantly inhibited lipolysis. The degree of inhibition increased with longer duration of lipase exposure to trophozoites. Total amounts of enzyme inhibited were proportional to enzyme concentration, while the percentage inhibition was greatest at lowest concentration. At a lipase concentration of 1-7 i.u./ml, enzyme activity was reduced by 89-7 ° 0 compared to controls after incubation for 4 h with trophozoites. The effect was abolished using killed, intact trophozoites. Culture filtrates of G. lamblia did not inhibit lipolysis. Specificity of the effect was suggested by the failure of another flagellate protozoan, Trichomonas vaginalis, to inhibit lipase. In this assay system the inhibition of lipolysis was not dependent on the bile salt concentration present. The impact of this effect in vivo remains to be determined but it may contribute to fat malabsorption in giardiasis. Key words: Giardia lamblia, lipase inhibition, pathogenesis of malabsorption.
INTRODUCTION
Giardia lamblia (G. intestinalis) is a common protozoal pathogen worldwide. It preferentially colonizes the upper small intestine in humans. Although many infected subjects are asymptomatic it may cause acute or chronic diarrhoea often with malabsorption of fat and other nutrients. The pathogenesis of the malabsorption is controversial and likely to be multifactorial. Attention has focused on functional and morphological changes in the small bowel mucosa; however, these are often minor and not invariably present (Hartong, Gourley & Arvanitakis, 1979; Owen, Nemanic & Stevens, 1979; Gillon, 1984). Other mechanisms postulated include physical occlusion of the mucosa (Veghelyi, 1937), secondary bacterial overgrowth (Tomkins et al. 1978), deconjugation (Tandon et al. 1977) or consumption of bile salts (Halliday, Clark & Farthing, 1988) and diminished pancreatic exocrine function. Gupta & Mehta (1973) reported decreased lipase and tryptic activity in duodenal juice from children with giardiasis. Similarly, Chawla et al. (1975) found diminished tryptic activity in duodenal aspirates from symptomatic subjects, which corrected with eradication of the organism. In neither instance was
* Present address and address for reprints: Dr Peter Katelaris, Gastroenterology Department, St Bartholomew's Hospital, London EC1A 7BE. Parasitology (1991), 103, 35-39 Printed in Great Britain
the mechanism for these observations elucidated. The aim of this study was to examine the direct effect of Giardia trophozoites on pancreatic lipase activity in vitro.
MATERIALS AND METHODS
Microorganisms and culture media Giardia lamblia trophozoites (Portland 1) were originally provided by Ms N. Villa, Cel-Labs, Manly Vale, Australia and have been maintained in axenic culture in this laboratory for 2 years. Trichomonas vaginalis were provided by Ms J. Chan, Prince of Wales Hospital, Randwick, Australia from a patient's vaginal swab, and axenized using methods previously described (Philip, Carter-Scott & Rogers, 1987). Giardia and Trichomonas trophozoites were cultured in modified TYI-S-33 medium (Keister, 1983) at 37 °C. Absence of bacterial contamination in axenic cultures was confirmed by periodic sterility testing. For experiments, Giardia were cultured in 50 ml sterile culture flasks with a base surface area of 25 cm 2 (Corning, USA), until a dense adherent monolayer had formed. In preliminary studies, medium was decanted from 72-h-old cultures and the monolayer dislodged by chilling. Trophozoites were fixed with 1 % formalin and counted on an automated cell counter (Coulter Electronics,
P. Katelaris, F. Seow and M. Ngu
Hialeah, USA). per monolayer
The mean used for
5 - 4 X 1 0 ' ± S . E . M . 0-29 ( « =
36
trophozoite count experiments was
11).
Reagents and assay Type 11 porcine pancreatic triacylglycerol lipase was used (Sigma Chemical Company, St Louis, USA). This contains co-lipase in a 1:1 molar ratio and shows extensive homology with human lipase. Fresh solutions were made for each experiment by solubilizing 10 g of enzyme in 10 ml of bicarbonatebuffered saline. This was then centrifuged at 1000 £ for 10 min and the supernatant fraction was filter sterilized by passage through a 0'22 /tm membrane filter. Lipase activity was calibrated against purified porcine lipase of known activity (United States Biochemical Corporation, Cleveland, USA). Solutions were diluted with buffer as required. Pancreatic lipase activity (PLA) was assayed according to the method of Tietz & Fiereck (1966), using olive oil emulsion as substrate (Sigma Diagnostics) and an Orion 960 pH autochemistry system. Total bile acids were assayed using a Sterognost-3a Flu enzymatic kit (Nyegaard Diagnostica, Oslo, Norway) based on the method of Mashige et al. (1978).
Experimental design Media from 72-h-old Giardia cultures were decanted and the adherent monolayer gently washed twice with warmed fresh media. New medium (7-5 ml) containing lipase was then added to monolayers and incubated at 37 °C. As controls, for each experiment an equal amount of lipase was incubated with medium alone. PLA was determined in samples of media drawn as required. Effect of Giardia trophozoites on PLA with time. Monolayers were incubated with media containing 170 i.u./ml lipase. The PLA of the media was assayed after 1 , 4 , 8 and 22 h incubation with Giardia. Effect of varying concentrations of lipase. Lipase solution was diluted to give activities of 1-7, 5-4, 17, 54 and 170 i.u./ml of medium and incubated with Giardia monolayers and control medium. PLA was determined in each after 4 h. Effect of dead trophozoites on PLA. Giardia monolayers were killed by the addition of 8 mmol/1 acetic acid for 1 h. Adequacy of killing was assessed by absence of motility and adherence, inability to exclude trypan blue and failure to propagate on subculturing. After killing, trophozoites were centrifuged at 100 gior 10 min, washed twice in fresh media and resus-
pended in 7-5 ml of medium at 37 °C. On lightmicroscopic examination, trophozoites were intact with normal morphology. Lipase, 170 i.u./ml, was added and incubated for 4 h then assayed for PLA. A live monolayer and a control were paired with each experiment. Effect of Trichomonas vaginalis on PLA. As inhibition of lipase could potentially result from nonspecific factors due to contact of lipase with any living cell, experiments were done using this protozoan. Cultures of Trichomonas trophozoites 48 h old were spun and resuspended in 7-5 ml of fresh medium (mean 4-6 x 1 0 ? + S.E.M. 0-6 organisms, n = 6). Lipase, in concentrations of 17 and 170 i.u./ml, was added. After incubation for 4 h PLA was assayed. Each experiment was paired with a Giardia monolayer and a control. Effect of culture filtrates and bile salts on PLA. New cultures were initiated by the addition of 1-6—2-0 x 106 trophozoites/50 ml of medium. Culture filtrates of Giardia were obtained from cultures after 72 h growth. The mean total number of organisms at this time was 1-5 x 108 + S.E.M. 0-02, n = 6. Lipase, 17 i.u./ml, was incubated for 4 h with 75 ml of culture filtrate and with fresh media as controls and PLA was measured. Bile acid concentrations were measured in the media, before and after 72 h growth of Giardia. TYI-S-33 medium used in all experiments contained 0-6 g/1 bovine bile (Sigma Chemical Company). To determine whether the concentration of bile acids was diminished in media used in experiments, total bile acids were measured following 4, 8 and 22 h incubation with Giardia, and compared with concentrations in fresh media. As bile acids affect lipolysis, the PLA of enzyme added to fresh medium devoid of bile was measured to assess the impact of bile acids on the assay.
Statistical analysis The significance of differences between observations was evaluated with paired i-tests or analysis of variance where appropriate.
RESULTS
PLA was inhibited after incubation with Giardia in a manner suggesting first-order kinetics. The effect of Giardia on PLA with time is shown in Fig. 1. PLA was decreased after incubation with Giardia in a time-dependent manner while the PLA in control medium remained stable. The inhibitory effect of Giardia on PLA was linearly related to the log of lipase concentration.
Effect of Giardia on lipolysis
37
200 180 .u./mu
160 140
ipase aci IVIt
120 100 80 60
-i
40 20 0
4
8
20
24
G/ard/a-lipase incubation time (h)
Fig. 1. The effect of Giardia lamblia trophozoites on lipase activity with time at an initial lipase concentration of 170 i.u./ml of medium. Results are plotted as mean + S.E.M.; n = 9, 9, 6 and 3 for 1, 4, 8 and 22 h respectively. (O) Control; ( # ) Giardia.
100
E 90
Lipase i
ibi
6 80 70 a 60 50 j= 40 c 30 20 10 0
20
40 60 80 100 120 140 Lipase concentration (i.u./ml)
160
180
Fig. 2. The effect of Giardia lamblia on total amount of lipase inhibited at varying concentrations of lipase after incubation for 4 h. Results are plotted as mean ±S.E.M.
(B)
(A)
o t 120-1 17
i.u./ml
170
i.u./ml
12O
1 170
8 ioo-
100-
80-
80-
60-
60-
40-
40-
20-
20-
0
0
i.u./ml
Fig. 3. (A) The effect on PLA of live Giardia lamblia ( • ) compared with Trichomonas vaginalis (E3) at two concentrations of enzyme. Paired t-Xest: P = 0-005 (n = 3) and P = 0-0002 (n = 6) for PLA of 17 and 170 i.u./ml respectively. (B) The effect of live Giardia ( • ) compared to killed Giardia (H) on PLA at 170 i.u./ml. Paired Z-test: P = 00002 (n = 5).
Percentage inhibition of PLA increased with decreasing enzyme concentration. Thus, at enzyme concentrations of 17 and 17i.u/ml, PLA was
reduced by 89-7 and 73-0% respectively, compared to the activity of the control after incubation for 4 h. However, the total amount of enzyme inhibited rose proportionally with higher initial enzyme concentration (Fig. 2). The inhibitory effect of Giardia on PLA was abolished by killing trophozoites with acetic acid. No difference in PLA compared to controls was evident after incubation for 4 h at an enzyme activity of 170 i.u./ml. T. vaginalis did not show any inhibitory effect on PLA after incubation for 4 h with lipase at this or 10-fold lower concentration of enzyme (Fig. 3). Culture filtrates of Giardia had no effect on PLA even at low lipase concentration (17 i.u./ml), after incubation for 4 h with the enzyme. Total bile acid concentration in culture filtrates after 72 h growth was significantly reduced compared to control media (365 + 5/imol/l versus 387 ± 3 /tmol/1, P = 0-02), although the difference (5-7%) was small. The total bile acid concentration in the medium after 4, 8 and 22 h incubation with a Giardia monolayer was not diminished compared to control medium. This excludes the possibility that the inhibition of PLA seen with Giardia monolayers was due to diminished concentrations of bile acids. Moreover, PLA assayed in bile-free control medium was not significantly different from that in either control medium with bile or culture filtrates, indicating that bile acids were not critical to lipolysis in this assay system.
DISCUSSION
These experiments indicate that live G. lamblia trophozoites inhibit lipolysis in vitro. This effect is time and concentration dependent. As neither dead organisms nor live T. vaginalis inhibited PLA, the effect is unlikely to be due to non-specific adsorption of enzyme onto the surface of the organism. Furthermore, although Giardia has been shown to consume bile salts, and bile salts are important for the hydrolysis of lipid by lipase in vivo, this was not the mechanism of inhibition of PLA in these experiments, as Giardia did not reduce bile acid concentrations in the media while incubated with lipase over the time tested. In addition, bile-free media did not significantly affect the assay compared to bile-containing media. According to Borgstrom & Erlanson (1978) a major function of bile acids in lipid absorption is to clear the lipid-water interface of proteins which may otherwise block the binding of lipase. The lack of effect of bile acids per se on the assay was probably due to the low concentration of protein in the assay system, and possibly the relatively high concentrations of lipase and co-lipase present. Smith, Horsburgh & Brown (1981) have shown that sonicates of Giardia devoid of cell membrane
P. Katelaris, F. Seow and M. Ngu
fragments also inhibit lipolysis. Their work suggested that the factor responsible was a high molecular weight heat-labile constituent of the organism. It is therefore possible that lipase is inhibited by a molecule produced intracellularly and expressed on the cell surface. Alternatively, the trophozoite may transport the enzyme across the cell membrane and into the cytosol. Either of these theories would be consistent with the kinetics of the inhibition observed in the experiments reported here. Giardia are known to produce excretorysecretory products in culture (Nash, Gillin & Smith, 1983). The lack of effect of culture filtrates on PLA suggests that these products are not responsible for the inhibition of PLA seen. Diminished pancreatic exocrine function in patients with giardiasis has been reported. A reduction in both tryptic activity (Chawla et al. 1975) and lipase activity was seen in duodenal aspirates of infected subjects. Faecal fat excretion correlated with lipase activity (Gupta et al. 1973). Okada et al. (1983) demonstrated abnormal pancreatic chymotrypsin activity using a BTPABA test in adult subjects with giardiasis. The mechanism of these effects has not been elucidated. Suggested but unproven hypotheses include direct blockage of the pancreatic duct, pancreatic parenchymal damage, disturbed enzyme activation in the duodenum, direct enzyme inhibition or deranged hormonal feedback for pancreatic secretion. Halliday et al. (1988) have suggested that reduction of intraluminal bile salt concentration in vivo due to consumption by Giardia may contribute to fat malabsorption by directly affecting micellular solubilization of lipid or by impeding the action of lipase. A direct inhibition of pancreatic lipase activity by live Giardia trophozoites, independent ofbile acid concentration, has been demonstrated in this work. In clinical infection, the duration of exposure of secreted lipase to the organism would vary depending on small intestinal transit time. However, up to 1012 organisms have been estimated to be present during infection, spread over predominantly the upper small intestine, ensuring a large area of contact between the enzyme and the organism. As Giardia is unable to synthesize lipid de novo and appears to be dependent on the surroundings for pre-formed lipid (Jarroll et al. 1981), the ability of Giardia to inhibit lipolysis could be advantageous for the organism by promoting an environment where lipid is plentiful. Furthermore, products of lipolysis found in human duodenal fluid have been shown to be toxic to trophozoites in vitro (Das et al. 1988). Inhibition of lipase by Giardia may aid the organism counter this host defence. These in vitro observations suggest a mechanism for fat malabsorption in giardiasis which is not dependent on mucosal abnormality. The impact of this effect in vivo remains to be determined but it may contribute to steatorrhoea in giardiasis.
38 Dr Katelaris was supported by a Research Fellowship from the Medical Foundation of the University of Sydney. The authors thank Ms M. Armstrong for performing the bile acid assays and the Microbiology Department at Concord Hospital for technical assistance. REFERENCES BORGSTROM, B. & ERLANSON, c. (1978). Interactions of
serum albumin and other proteins with porcine pancreatic lipase. Gastroenterology 75, 382—6. CHAWLA, L. S., SEHGAL, A. K., BROOR, S. L., VERMA, R. S. &
CHHUTTANI, P. N. (1975). Tryptic activity in the duodenal aspirate following a standard test meal in giardiasis. Scandinavian Journal of Gastroenterology 10, 445-7. DAS, S., REINER, D. S., ZENIAN, J., HOGAN, D. L., KOSS, M. A., WANG, C.-S. & GILLIN, F. D. ( 1 9 8 8 ) . K i l l i n g of
Giardia lamblia trophozoites by human intestinal fluid in vitro. Journal of Infectious Diseases 157, 1257—60. GILLON, j . (1984). Giardiasis: review of epidemiology, pathogenetic mechanisms and host responses. Quarterly Journal of Medicine 209, 29-39. GUPTA, p. K. & MEHTA, s. (1973). Giardiasis in children: a study of pancreatic functions. Indian Journal of Medical Research 61, 743-8. HALLIDAY, C. E. W., CLARK, C. & FARTHING, M. J. G. ( 1 9 8 8 ) .
Giardia—bile salt interactions in vitro and in vivo. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 428-32. HARTONG, W. A., GOURLEY, W. K. & ARVANITAKIS, C.
(1979). Giardiasis: clinical spectrum and functionalstructural abnormalities of the small intestine mucosa. Gastroenterology 77, 61—9. J A R R O L L , E. L . , M U L L E R , P . J . , M E Y E R , E. A. & M O R S E , S. A.
(1981). Lipid and carbohydrate metabolism of Giardia lamblia. Molecular and Biochemical Parasitology 2, 187-96. KEISTER, D. B. (1983). Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Transactions of the Royal Society of Tropical Medicine and Hygiene 77, 487-8. MASHIGE, F., TANAKA, N . , MAKI, A., KAMEI, S. &
YAMANAKA, M. (1978). Direct spectrophotometry of total bile acids in serum. Clinical Chemistry 27, 1352-6. NASH, T. E., GILLIN, F. D. & SMITH, P. D. ( 1 9 8 3 ) .
Excretory-secretory products of Giardia lamblia. Journal of Immunology 131, 2004-10. OKADA, M., FUCHIGAMI, T., RI, S., KOHROGI, N. & OMAE, T.
(1983). The BTPABA pancreatic function test in giardiasis. Postgraduate Medical Journal 59, 79-82. OWEN, R. L., NEMANIC, P. & STEVENS, D. P. ( 1 9 7 9 ) .
Ultrastructural observations on giardiasis in a murine model. Gastroenterology 76, 757-69. PHILIP, A., CARTER-SCOTT, P. & ROGERS, C. ( 1 9 8 7 ) . A n
agar culture technique to quantitate Trichomonas vaginalis from women. Journal of Infectious Diseases 155, 304-8. SMITH, P. D., HORSBURGH, C. R. & BROWN, W. R. ( 1 9 8 1 ) . In
vitro studies on bile acid deconjugation and lipolysis inhibition by Giardia lamblia. Digestive Diseases and Sciences 26, 700-4. TANDON, B. N . , TANDON, R. K., SATPATHY, B. K. &
SHRINIWAS. (1977). Mechanism of malabsorption in
Effect of Giardia on lipolysis giardiasis: a study of bacterial flora and bile salt deconjugation in upper jejunum. Gut 18, 176-81. TIETZ, N. w. & FIERECK, E. A. (1966). A specific method for serum lipase determination. Clinica Chimica Ada 13, 352. TOMKINS, A. M., DRASAR, B. S., BRADLEY, A. K. &
39 WILLIAMSON, W. A. (1978). Bacterial colonisation of jejunal mucosa in giardiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 72, 33-6. VEGHELYI, P. (1937). Celiac disease imitated by giardiasis. American Journal of Diseases of Children 57, 894—9.
The effect of Giardia lamblia trophozoites on lipolysis in vitro P. KATELARIS*, F. SEOWWM. NGU Gastroenterology Unit, Concord Hospital, Sydney, Australia 2139 (Received 6 November 1990; revised 7 February 1991; accepted 11 February 1991)
SUMMARY
Infection with Giardia lamblia often causes only minor mucosal changes to the small intestine yet frank fat malabsorption may still occur. Some evidence suggests abnormal pancreatic exocrine function in subjects with giardiasis although the mechanism and significance of this is unclear. Studies were conducted in vitro to determine the effect of G. lamblia trophozoites or culture filtrates from the organism on lipolysis of triglyceride by porcine pancreatic lipase. Live trophozoites significantly inhibited lipolysis. The degree of inhibition increased with longer duration of lipase exposure to trophozoites. Total amounts of enzyme inhibited were proportional to enzyme concentration, while the percentage inhibition was greatest at lowest concentration. At a lipase concentration of 1-7 i.u./ml, enzyme activity was reduced by 89-7 ° 0 compared to controls after incubation for 4 h with trophozoites. The effect was abolished using killed, intact trophozoites. Culture filtrates of G. lamblia did not inhibit lipolysis. Specificity of the effect was suggested by the failure of another flagellate protozoan, Trichomonas vaginalis, to inhibit lipase. In this assay system the inhibition of lipolysis was not dependent on the bile salt concentration present. The impact of this effect in vivo remains to be determined but it may contribute to fat malabsorption in giardiasis. Key words: Giardia lamblia, lipase inhibition, pathogenesis of malabsorption.
INTRODUCTION
Giardia lamblia (G. intestinalis) is a common protozoal pathogen worldwide. It preferentially colonizes the upper small intestine in humans. Although many infected subjects are asymptomatic it may cause acute or chronic diarrhoea often with malabsorption of fat and other nutrients. The pathogenesis of the malabsorption is controversial and likely to be multifactorial. Attention has focused on functional and morphological changes in the small bowel mucosa; however, these are often minor and not invariably present (Hartong, Gourley & Arvanitakis, 1979; Owen, Nemanic & Stevens, 1979; Gillon, 1984). Other mechanisms postulated include physical occlusion of the mucosa (Veghelyi, 1937), secondary bacterial overgrowth (Tomkins et al. 1978), deconjugation (Tandon et al. 1977) or consumption of bile salts (Halliday, Clark & Farthing, 1988) and diminished pancreatic exocrine function. Gupta & Mehta (1973) reported decreased lipase and tryptic activity in duodenal juice from children with giardiasis. Similarly, Chawla et al. (1975) found diminished tryptic activity in duodenal aspirates from symptomatic subjects, which corrected with eradication of the organism. In neither instance was
* Present address and address for reprints: Dr Peter Katelaris, Gastroenterology Department, St Bartholomew's Hospital, London EC1A 7BE. Parasitology (1991), 103, 35-39 Printed in Great Britain
the mechanism for these observations elucidated. The aim of this study was to examine the direct effect of Giardia trophozoites on pancreatic lipase activity in vitro.
MATERIALS AND METHODS
Microorganisms and culture media Giardia lamblia trophozoites (Portland 1) were originally provided by Ms N. Villa, Cel-Labs, Manly Vale, Australia and have been maintained in axenic culture in this laboratory for 2 years. Trichomonas vaginalis were provided by Ms J. Chan, Prince of Wales Hospital, Randwick, Australia from a patient's vaginal swab, and axenized using methods previously described (Philip, Carter-Scott & Rogers, 1987). Giardia and Trichomonas trophozoites were cultured in modified TYI-S-33 medium (Keister, 1983) at 37 °C. Absence of bacterial contamination in axenic cultures was confirmed by periodic sterility testing. For experiments, Giardia were cultured in 50 ml sterile culture flasks with a base surface area of 25 cm 2 (Corning, USA), until a dense adherent monolayer had formed. In preliminary studies, medium was decanted from 72-h-old cultures and the monolayer dislodged by chilling. Trophozoites were fixed with 1 % formalin and counted on an automated cell counter (Coulter Electronics,
P. Katelaris, F. Seow and M. Ngu
Hialeah, USA). per monolayer
The mean used for
5 - 4 X 1 0 ' ± S . E . M . 0-29 ( « =
36
trophozoite count experiments was
11).
Reagents and assay Type 11 porcine pancreatic triacylglycerol lipase was used (Sigma Chemical Company, St Louis, USA). This contains co-lipase in a 1:1 molar ratio and shows extensive homology with human lipase. Fresh solutions were made for each experiment by solubilizing 10 g of enzyme in 10 ml of bicarbonatebuffered saline. This was then centrifuged at 1000 £ for 10 min and the supernatant fraction was filter sterilized by passage through a 0'22 /tm membrane filter. Lipase activity was calibrated against purified porcine lipase of known activity (United States Biochemical Corporation, Cleveland, USA). Solutions were diluted with buffer as required. Pancreatic lipase activity (PLA) was assayed according to the method of Tietz & Fiereck (1966), using olive oil emulsion as substrate (Sigma Diagnostics) and an Orion 960 pH autochemistry system. Total bile acids were assayed using a Sterognost-3a Flu enzymatic kit (Nyegaard Diagnostica, Oslo, Norway) based on the method of Mashige et al. (1978).
Experimental design Media from 72-h-old Giardia cultures were decanted and the adherent monolayer gently washed twice with warmed fresh media. New medium (7-5 ml) containing lipase was then added to monolayers and incubated at 37 °C. As controls, for each experiment an equal amount of lipase was incubated with medium alone. PLA was determined in samples of media drawn as required. Effect of Giardia trophozoites on PLA with time. Monolayers were incubated with media containing 170 i.u./ml lipase. The PLA of the media was assayed after 1 , 4 , 8 and 22 h incubation with Giardia. Effect of varying concentrations of lipase. Lipase solution was diluted to give activities of 1-7, 5-4, 17, 54 and 170 i.u./ml of medium and incubated with Giardia monolayers and control medium. PLA was determined in each after 4 h. Effect of dead trophozoites on PLA. Giardia monolayers were killed by the addition of 8 mmol/1 acetic acid for 1 h. Adequacy of killing was assessed by absence of motility and adherence, inability to exclude trypan blue and failure to propagate on subculturing. After killing, trophozoites were centrifuged at 100 gior 10 min, washed twice in fresh media and resus-
pended in 7-5 ml of medium at 37 °C. On lightmicroscopic examination, trophozoites were intact with normal morphology. Lipase, 170 i.u./ml, was added and incubated for 4 h then assayed for PLA. A live monolayer and a control were paired with each experiment. Effect of Trichomonas vaginalis on PLA. As inhibition of lipase could potentially result from nonspecific factors due to contact of lipase with any living cell, experiments were done using this protozoan. Cultures of Trichomonas trophozoites 48 h old were spun and resuspended in 7-5 ml of fresh medium (mean 4-6 x 1 0 ? + S.E.M. 0-6 organisms, n = 6). Lipase, in concentrations of 17 and 170 i.u./ml, was added. After incubation for 4 h PLA was assayed. Each experiment was paired with a Giardia monolayer and a control. Effect of culture filtrates and bile salts on PLA. New cultures were initiated by the addition of 1-6—2-0 x 106 trophozoites/50 ml of medium. Culture filtrates of Giardia were obtained from cultures after 72 h growth. The mean total number of organisms at this time was 1-5 x 108 + S.E.M. 0-02, n = 6. Lipase, 17 i.u./ml, was incubated for 4 h with 75 ml of culture filtrate and with fresh media as controls and PLA was measured. Bile acid concentrations were measured in the media, before and after 72 h growth of Giardia. TYI-S-33 medium used in all experiments contained 0-6 g/1 bovine bile (Sigma Chemical Company). To determine whether the concentration of bile acids was diminished in media used in experiments, total bile acids were measured following 4, 8 and 22 h incubation with Giardia, and compared with concentrations in fresh media. As bile acids affect lipolysis, the PLA of enzyme added to fresh medium devoid of bile was measured to assess the impact of bile acids on the assay.
Statistical analysis The significance of differences between observations was evaluated with paired i-tests or analysis of variance where appropriate.
RESULTS
PLA was inhibited after incubation with Giardia in a manner suggesting first-order kinetics. The effect of Giardia on PLA with time is shown in Fig. 1. PLA was decreased after incubation with Giardia in a time-dependent manner while the PLA in control medium remained stable. The inhibitory effect of Giardia on PLA was linearly related to the log of lipase concentration.
Effect of Giardia on lipolysis
37
200 180 .u./mu
160 140
ipase aci IVIt
120 100 80 60
-i
40 20 0
4
8
20
24
G/ard/a-lipase incubation time (h)
Fig. 1. The effect of Giardia lamblia trophozoites on lipase activity with time at an initial lipase concentration of 170 i.u./ml of medium. Results are plotted as mean + S.E.M.; n = 9, 9, 6 and 3 for 1, 4, 8 and 22 h respectively. (O) Control; ( # ) Giardia.
100
E 90
Lipase i
ibi
6 80 70 a 60 50 j= 40 c 30 20 10 0
20
40 60 80 100 120 140 Lipase concentration (i.u./ml)
160
180
Fig. 2. The effect of Giardia lamblia on total amount of lipase inhibited at varying concentrations of lipase after incubation for 4 h. Results are plotted as mean ±S.E.M.
(B)
(A)
o t 120-1 17
i.u./ml
170
i.u./ml
12O
1 170
8 ioo-
100-
80-
80-
60-
60-
40-
40-
20-
20-
0
0
i.u./ml
Fig. 3. (A) The effect on PLA of live Giardia lamblia ( • ) compared with Trichomonas vaginalis (E3) at two concentrations of enzyme. Paired t-Xest: P = 0-005 (n = 3) and P = 0-0002 (n = 6) for PLA of 17 and 170 i.u./ml respectively. (B) The effect of live Giardia ( • ) compared to killed Giardia (H) on PLA at 170 i.u./ml. Paired Z-test: P = 00002 (n = 5).
Percentage inhibition of PLA increased with decreasing enzyme concentration. Thus, at enzyme concentrations of 17 and 17i.u/ml, PLA was
reduced by 89-7 and 73-0% respectively, compared to the activity of the control after incubation for 4 h. However, the total amount of enzyme inhibited rose proportionally with higher initial enzyme concentration (Fig. 2). The inhibitory effect of Giardia on PLA was abolished by killing trophozoites with acetic acid. No difference in PLA compared to controls was evident after incubation for 4 h at an enzyme activity of 170 i.u./ml. T. vaginalis did not show any inhibitory effect on PLA after incubation for 4 h with lipase at this or 10-fold lower concentration of enzyme (Fig. 3). Culture filtrates of Giardia had no effect on PLA even at low lipase concentration (17 i.u./ml), after incubation for 4 h with the enzyme. Total bile acid concentration in culture filtrates after 72 h growth was significantly reduced compared to control media (365 + 5/imol/l versus 387 ± 3 /tmol/1, P = 0-02), although the difference (5-7%) was small. The total bile acid concentration in the medium after 4, 8 and 22 h incubation with a Giardia monolayer was not diminished compared to control medium. This excludes the possibility that the inhibition of PLA seen with Giardia monolayers was due to diminished concentrations of bile acids. Moreover, PLA assayed in bile-free control medium was not significantly different from that in either control medium with bile or culture filtrates, indicating that bile acids were not critical to lipolysis in this assay system.
DISCUSSION
These experiments indicate that live G. lamblia trophozoites inhibit lipolysis in vitro. This effect is time and concentration dependent. As neither dead organisms nor live T. vaginalis inhibited PLA, the effect is unlikely to be due to non-specific adsorption of enzyme onto the surface of the organism. Furthermore, although Giardia has been shown to consume bile salts, and bile salts are important for the hydrolysis of lipid by lipase in vivo, this was not the mechanism of inhibition of PLA in these experiments, as Giardia did not reduce bile acid concentrations in the media while incubated with lipase over the time tested. In addition, bile-free media did not significantly affect the assay compared to bile-containing media. According to Borgstrom & Erlanson (1978) a major function of bile acids in lipid absorption is to clear the lipid-water interface of proteins which may otherwise block the binding of lipase. The lack of effect of bile acids per se on the assay was probably due to the low concentration of protein in the assay system, and possibly the relatively high concentrations of lipase and co-lipase present. Smith, Horsburgh & Brown (1981) have shown that sonicates of Giardia devoid of cell membrane
P. Katelaris, F. Seow and M. Ngu
fragments also inhibit lipolysis. Their work suggested that the factor responsible was a high molecular weight heat-labile constituent of the organism. It is therefore possible that lipase is inhibited by a molecule produced intracellularly and expressed on the cell surface. Alternatively, the trophozoite may transport the enzyme across the cell membrane and into the cytosol. Either of these theories would be consistent with the kinetics of the inhibition observed in the experiments reported here. Giardia are known to produce excretorysecretory products in culture (Nash, Gillin & Smith, 1983). The lack of effect of culture filtrates on PLA suggests that these products are not responsible for the inhibition of PLA seen. Diminished pancreatic exocrine function in patients with giardiasis has been reported. A reduction in both tryptic activity (Chawla et al. 1975) and lipase activity was seen in duodenal aspirates of infected subjects. Faecal fat excretion correlated with lipase activity (Gupta et al. 1973). Okada et al. (1983) demonstrated abnormal pancreatic chymotrypsin activity using a BTPABA test in adult subjects with giardiasis. The mechanism of these effects has not been elucidated. Suggested but unproven hypotheses include direct blockage of the pancreatic duct, pancreatic parenchymal damage, disturbed enzyme activation in the duodenum, direct enzyme inhibition or deranged hormonal feedback for pancreatic secretion. Halliday et al. (1988) have suggested that reduction of intraluminal bile salt concentration in vivo due to consumption by Giardia may contribute to fat malabsorption by directly affecting micellular solubilization of lipid or by impeding the action of lipase. A direct inhibition of pancreatic lipase activity by live Giardia trophozoites, independent ofbile acid concentration, has been demonstrated in this work. In clinical infection, the duration of exposure of secreted lipase to the organism would vary depending on small intestinal transit time. However, up to 1012 organisms have been estimated to be present during infection, spread over predominantly the upper small intestine, ensuring a large area of contact between the enzyme and the organism. As Giardia is unable to synthesize lipid de novo and appears to be dependent on the surroundings for pre-formed lipid (Jarroll et al. 1981), the ability of Giardia to inhibit lipolysis could be advantageous for the organism by promoting an environment where lipid is plentiful. Furthermore, products of lipolysis found in human duodenal fluid have been shown to be toxic to trophozoites in vitro (Das et al. 1988). Inhibition of lipase by Giardia may aid the organism counter this host defence. These in vitro observations suggest a mechanism for fat malabsorption in giardiasis which is not dependent on mucosal abnormality. The impact of this effect in vivo remains to be determined but it may contribute to steatorrhoea in giardiasis.
38 Dr Katelaris was supported by a Research Fellowship from the Medical Foundation of the University of Sydney. The authors thank Ms M. Armstrong for performing the bile acid assays and the Microbiology Department at Concord Hospital for technical assistance. REFERENCES BORGSTROM, B. & ERLANSON, c. (1978). Interactions of
serum albumin and other proteins with porcine pancreatic lipase. Gastroenterology 75, 382—6. CHAWLA, L. S., SEHGAL, A. K., BROOR, S. L., VERMA, R. S. &
CHHUTTANI, P. N. (1975). Tryptic activity in the duodenal aspirate following a standard test meal in giardiasis. Scandinavian Journal of Gastroenterology 10, 445-7. DAS, S., REINER, D. S., ZENIAN, J., HOGAN, D. L., KOSS, M. A., WANG, C.-S. & GILLIN, F. D. ( 1 9 8 8 ) . K i l l i n g of
Giardia lamblia trophozoites by human intestinal fluid in vitro. Journal of Infectious Diseases 157, 1257—60. GILLON, j . (1984). Giardiasis: review of epidemiology, pathogenetic mechanisms and host responses. Quarterly Journal of Medicine 209, 29-39. GUPTA, p. K. & MEHTA, s. (1973). Giardiasis in children: a study of pancreatic functions. Indian Journal of Medical Research 61, 743-8. HALLIDAY, C. E. W., CLARK, C. & FARTHING, M. J. G. ( 1 9 8 8 ) .
Giardia—bile salt interactions in vitro and in vivo. Transactions of the Royal Society of Tropical Medicine and Hygiene 82, 428-32. HARTONG, W. A., GOURLEY, W. K. & ARVANITAKIS, C.
(1979). Giardiasis: clinical spectrum and functionalstructural abnormalities of the small intestine mucosa. Gastroenterology 77, 61—9. J A R R O L L , E. L . , M U L L E R , P . J . , M E Y E R , E. A. & M O R S E , S. A.
(1981). Lipid and carbohydrate metabolism of Giardia lamblia. Molecular and Biochemical Parasitology 2, 187-96. KEISTER, D. B. (1983). Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Transactions of the Royal Society of Tropical Medicine and Hygiene 77, 487-8. MASHIGE, F., TANAKA, N . , MAKI, A., KAMEI, S. &
YAMANAKA, M. (1978). Direct spectrophotometry of total bile acids in serum. Clinical Chemistry 27, 1352-6. NASH, T. E., GILLIN, F. D. & SMITH, P. D. ( 1 9 8 3 ) .
Excretory-secretory products of Giardia lamblia. Journal of Immunology 131, 2004-10. OKADA, M., FUCHIGAMI, T., RI, S., KOHROGI, N. & OMAE, T.
(1983). The BTPABA pancreatic function test in giardiasis. Postgraduate Medical Journal 59, 79-82. OWEN, R. L., NEMANIC, P. & STEVENS, D. P. ( 1 9 7 9 ) .
Ultrastructural observations on giardiasis in a murine model. Gastroenterology 76, 757-69. PHILIP, A., CARTER-SCOTT, P. & ROGERS, C. ( 1 9 8 7 ) . A n
agar culture technique to quantitate Trichomonas vaginalis from women. Journal of Infectious Diseases 155, 304-8. SMITH, P. D., HORSBURGH, C. R. & BROWN, W. R. ( 1 9 8 1 ) . In
vitro studies on bile acid deconjugation and lipolysis inhibition by Giardia lamblia. Digestive Diseases and Sciences 26, 700-4. TANDON, B. N . , TANDON, R. K., SATPATHY, B. K. &
SHRINIWAS. (1977). Mechanism of malabsorption in
Effect of Giardia on lipolysis giardiasis: a study of bacterial flora and bile salt deconjugation in upper jejunum. Gut 18, 176-81. TIETZ, N. w. & FIERECK, E. A. (1966). A specific method for serum lipase determination. Clinica Chimica Ada 13, 352. TOMKINS, A. M., DRASAR, B. S., BRADLEY, A. K. &
39 WILLIAMSON, W. A. (1978). Bacterial colonisation of jejunal mucosa in giardiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 72, 33-6. VEGHELYI, P. (1937). Celiac disease imitated by giardiasis. American Journal of Diseases of Children 57, 894—9.
