Amino sugar phosphate levels in Giardia change during cyst wall formation

Keriman Sener1, Zuojun Shen2,{dagger}, David S. Newburg2 and Edward L. Jarroll1

1 Department of Biology, Northeastern University, Boston, MA 02115, USA
2 Program in Glycobiology, Shriver Center at University of Massachusetts Medical School, 200 Trapelo Road, Waltham, MA 02452, USA

Correspondence
David S. Newburg
david.newburg{at}umassmed.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The parasite Giardia intestinalis exists as a trophozoite (vegetative) that infects the human small intestine, and a cyst (infective) that is shed in host faeces. Cyst viability in the environment depends upon a protective cyst wall, which consists of proteins and a unique {beta}(1-3) GalNAc homopolymer. UDP-GalNAc, the precursor for this polysaccharide, is synthesized from glucose by an enzyme pathway that involves amino sugar phosphate intermediates. Using a novel method of microanalysis by capillary electrophoresis, the levels of amino sugar phosphate intermediates in trophozoites before encystment, during a period of active encystment and after the peak of encystment were measured. These levels were used to deduce metabolic control of amino sugar phosphates associated with encystment. Levels of amino sugar phosphate intermediates increased during encystment, and then decreased to nearly non-encysting levels. The most pronounced increase was in glucosamine 6-phosphate, which is the first substrate unique in this pathway, and which is the positive effector for the pathway's putative rate-controlling enzyme, UDP-GlcNAc pyrophosphorylase. Moreover, more UDP-GalNAc than UDP-GlcNAc, its direct precursor, was detected at 24 h. It is postulated that the enhanced UDP-GalNAc is a result of enhanced synthesis of UDP-GlcNAc by the pyrophosphorylase, and its preferential conversion to UDP-GalNAc. These results suggest that kinetics of amino sugar phosphate synthesis in encysting Giardia favours the direction that supports cyst wall synthesis. The enzymes involved in synthesis of UDP-GalNAc and its conversion to cyst wall might be potential targets for therapeutic inhibitors of Giardia infection.


Abbreviations: CWS, cyst wall synthase; GalN 1-P, galactosamine 1-phosphate; GlcN 1-P, glucosamine 1-phosphate; GlcN 6-P, glucosamine 6-phosphate; GlcNAc 1-P, N-acetylglucosamine 1-phosphate; GlcNAc 6-P, N-acetylglucosamine 6-phosphate

{dagger}Present address: Anhui Provincial Center for Clinical Laboratories, Anhui Medical University, Hefei, P. R. China.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Giardia intestinalis (syn. duodenalis, lamblia), a flagellate protozoan parasite, is a major cause of enteric disease with approximately 108 cases annually (Lane & Lloyd, 2002). Giardiasis may result in acute or chronic diarrhoea, dehydration, abdominal discomfort and weight loss (Farthing, 1994).

Giardia's life cycle includes trophozoites and cysts. The vegetative trophozoite infects the proximal small intestine and may cause disease. In response to bile, the trophozoite can encyst (elaborate a cyst wall), and after being shed via the faeces into the environment, cysts remain viable for several months in cool moist conditions and resist chlorination (deRegnier et al., 1989; Farthing, 1996). It is as cysts that Giardia survives outside of the host and can thus be transmitted from one host to the next.

Giardia's cyst wall is composed of an inner membranous and an outer filamentous portion (Feely et al., 1984). The outer filamentous wall is made up of proteins (37 %) and carbohydrate filaments (63 %) composed mainly of a [D-galnac-{beta}(1-3)-D-GalNAc]n homopolymer (Gerwig et al., 2002). The precursor for the GalNAc homopolymer is UDP-GalNAc, which is synthesized from glucose by a pathway of inducible enzymes and amino sugar phosphate intermediates shown in Fig. 1. The enzymes involved in the synthesis of UDP-GalNAc include glucosamine-6-phosphate isomerase, glucosamine-6-phosphate N-acetylase, phosphoacetylglucosamine mutase, UDP-GlcNAc pyrophosphorylase and UDP-GlcNAc 4'-epimerase. All of these enzymes have been localized to the cytosol of encysting Giardia (Macechko et al., 1992). UDP-GalNAc is then converted into a unique {beta}(1-3)-D-GalNAc homopolysaccharide by the action of a particle-associated {beta}(1-3)GalNAc transferase, tentatively termed cyst wall synthase (CWS) (Jarroll et al., 2001). When trophozoites are induced to encyst, the first five of these enzymes are transcriptionally activated (Lopez et al., 2003) and the specific activities of all these enzymes increase (Macechko et al., 1992). Although CWS, which catalyses the ultimate step of the pathway, has not yet been cloned, we hypothesize that its transcription could also be induced, as it is part of the same pathway and its activity also increases during encystment. However, changes in levels of enzyme expression provide only an indirect measure of potential metabolic changes and control in an organism; changes in enzyme activity are often more relevant towards understanding the proximate basis of fine metabolic control. In this context, Bulik et al. (2000) showed that the GlcN 6-P concentration increased threefold during encystment and that GlcN 6-P can allosterically activate UDP-GlcNAc pyrophosphorylase in the direction of UDP-GlcNAc synthesis. However, changes in enzyme activity, measured in isolated enzymes in vitro, may not reflect their true activity in intact cells. Thus a more direct means of understanding the control of a metabolic pathway requires simultaneous measurement of levels of metabolic intermediates for each of the steps of the pathways under different physiological states. A long-standing difficulty with such simultaneous measurements of sugar phosphate intermediates in any metabolic pathway is their low concentrations in cells (Bessman, 1974).



View larger version (33K):
[in this window]
[in a new window]
 
Fig. 1. Amino sugar phosphate pathway supporting cyst wall {beta}(1-3)GalNAc homopolymer synthesis in Giardia.

 
Recent advances in capillary electrophoresis held the promise of the resolution and detection of these anionic sugar compounds in picogram quantities. Therefore, the methodology described herein was developed to allow the measurement of hexosamine phosphates in the Giardia trophozoite under basal conditions, while the cell is undergoing active encystment, and when encystment is nearing completion. The purposes of this study were to: (1) investigate the relationship between the known elevation of enzymes during encystment and changes in their respective metabolic products; (2) measure the relationship between changes in metabolic intermediates and the production of the ultimate product, UDP-GalNAc; and (3) compare the relative amounts of metabolic intermediates before and during encystment to determine the relative roles for each of the enzymes towards regulating this metabolic pathway and identify the rate-limiting steps in the synthesis of the cyst wall.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Giardia cells.
Giardia (MR4) trophozoites were grown axenically for 4–5 days in modified TYI-S-33 medium (Keister, 1983) without bile at 37 °C to obtain 0 h (non-encysting, 3·0x109 cells) trophozoites. After 4–5 days, trophozoites were harvested by centrifugation, washed three times in 0·25 M sucrose and stored at –80 °C. To obtain 24 or 48 h induced cells (2·3x109 and 2·0x109 cells, respectively), trophozoites were grown identically to 0 h cells and transferred into encystment medium (pH 7·6 with 10 % bovine bile) (Schupp et al., 1988) and harvested after an additional 24 or 48 h of encystment at 37 °C. Harvested cells were washed three times with 0·25 M sucrose, and stored at –80 °C until analysis, typically from 2 to 6 weeks. Encystment was asynchronous; at 24 h the proportion of cells that were completely encysted varied from 10 to 20 % and at 48 h from 25 to 40 %.

Amino sugar phosphate extraction.
Cells from each culture were resuspended in 30 ml ice-cold ultrapure water and divided into three equal aliquots. Each aliquot was treated as follows before homogenization. Two aliquots were made 1 % with phosphatase inhibitor (Phosphatase Inhibitor Cocktail 2; Sigma), and into one of the two, a mixture of authentic sugar phosphate standards containing GlcNAc 6-P, GlcN 1-P, GalN 1-P, GlcNAc 1-P, GlcN 6-P, UDP-GalNAc, UDP-GlcNAc and UDP-Glc (Sigma) was added. The first of these aliquots was used to quantify each peak that corresponded to an amino sugar phosphate of interest. The second with authentic standards was used to confirm the identity of each peak by co-elution with, and to calculate the recovery of, each of the added standards. The third aliquot, without inhibitor, was used to measure the organic phosphate peaks in the absence of phosphatase inhibitor, both with and without treatment with exogenous phosphatase. The three aliquots were homogenized on ice for 4 min using a Polytron homogenizer (model PT 10/35; Brinkman Instruments) at setting no. 4, with a 9 mm microprobe. The homogenates were centrifuged on a swinging bucket rotor (SW28) in a Beckman ultracentrifuge at 100 000 g for 90 min. Supernatants were transferred to 10 000 MWCO centrifugal ultrafilters (Filtron; Gelman) and filtered by centrifugation at 4900 g for 39 h. After measuring filtrate volumes, the filtrates were lyophilized and redissolved in 200 µl ultrapure water. From the third sample an 80 µl aliquot was removed for treatment with alkaline phosphatase as follows: into an 80 µl aliquot, 10 units (10 µl) shrimp alkaline phosphatase (New England BioLabs) and 10 µl 10x alkaline phosphatase buffer (New England BioLabs) were added and the solution was digested at 25 °C for 1 h. Disappearance of the peaks in this sample confirmed that the peaks co-eluting with the amino sugar phosphate standards were also themselves phosphorylated. As positive controls for the phosphatase digestion, a mixture of pure authentic standards at concentrations similar to those found in our samples was incubated with alkaline phosphatase. All samples were analysed by capillary electrophoresis.

Amino sugar phosphate measurement.
Capillary zone electrophoresis conditions were developed for the resolution of the mixture of amino sugar phosphate standards listed above: 30 kV normal polarity (sample loaded in the anode and detected at the cathode) in 30 mM sodium borate running buffer, pH 9·0 at 19 °C. The capillary had an effective length of 56 cm and an interior diameter of 50 µm with extended light path geometry at the detector. The hexosamine phosphates and nucleotide sugars were detected by their absorbance at 200 nm. This is the first technique capable of measuring all of the known intermediates of amino sugar phosphate metabolism as found naturally in Giardia and other tissues (D. S. Newburg and others, unpublished).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Capillary electrophoresis proved to be effective for the analysis of the amino sugar phosphates in Giardia. The electropherograms of a standard mixture of authentic amino sugar phosphates (STD) in Fig. 2 demonstrate that each standard was baseline resolved from the others in a reasonably short run time, with sharp, well-shaped peaks and negligible tailing. In Giardia samples (Fig. 2, Giardia) peaks were detected whose elution times corresponded to those of the standards. A range of concentrations was analysed for each standard, and the area under the peaks was integrated. The resulting curves were linear from the low pg nl–1 range through the low ng nl–1 range. The Giardia samples, as prepared above, contained amino sugar phosphate concentrations that were at least an order of magnitude greater than the lower limit of the linear ranges. The addition of an exogenous mixture of standards to the samples in the presence of phosphatase inhibitor was used for two purposes. An increase in sample peak height in the presence of standards confirmed their identity. The magnitude of the difference between peak areas in the presence or absence of standards was used to confirm that the phosphatase inhibitor was effective at inhibiting the endogenous phosphatase of Giardia during the extended extraction procedure, resulting in negligible loss of sugar phosphates. For Giardia extract lacking phosphatase inhibitor, the disappearance of these same peaks following incubation with phosphatase confirms that the peaks from Giardia that co-eluted with standards were indeed phosphorylated. In the mixture of pure standards, the disappearance of the peaks after incubation with phosphatase confirmed that these amino sugar phosphates are sensitive to digestion with this phosphatase, further reinforcing the conclusion that the co-eluting phosphatase-sensitive peaks in Giardia are indeed the amino sugar phosphates of interest.



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 2. Representative electropherograms of standards and samples of the amino sugar phosphates measured in Giardia. STD is the mixture of authentic amino sugar phosphate standards used in this study. Giardia samples are from the 48 h (post-encystment) cells. Using this method, we were able to resolve, detect and quantify pathway intermediates at concentrations at the low ng nl–1 range, even in the presence of the complex Giardia matrix.

 
Standard curves were determined for each compound of interest over the linear range of the method; these analyses were repeated multiple times over several days. Standard curves were used to assure that the sugar phosphates in Giardia were being measured in the linear range of this technique. These data were also used to calculate the coefficient of variation for the analysis of these compounds, which was approximately 5 % (range 4–7 %), indicating that the difference in concentrations between the Giardia samples before, during and following encystment greatly exceeded the error intrinsic to this method. Furthermore, when other laboratory personnel replicated the entire experiment, from Giardia culture through amino sugar analysis, the results produced were essentially identical to those reported herein.

This technique allowed the simultaneous measurement of amino sugar phosphates in Giardia as its trophozoite at 0 h, at 24 h after exposure to bile (peak of encystment) and at 48 h after exposure (when encystment is approaching completion). The data in Table 1 and Fig. 3 demonstrate that during encystment, the levels of all of the amino sugar phosphates in the UDP-GalNAc synthetic pathway rise above those levels seen in the trophozoite at 0 h in a co-ordinated fashion, and that after encystment, they return towards non-encysting levels. Such an increase is not seen in amino sugar phosphates that are not part of this UDP-GalNAc synthetic pathway, such as the GalN 1-P and GlcN 1-P shown at the bottom of Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Amino sugar phosphates and sugar nucleotides (amol per cell) measured in encysting Giardia

 


View larger version (22K):
[in this window]
[in a new window]
 
Fig. 3. Visual representation of the concentrations of Giardia amino sugar phosphate intermediates. Note the large absolute increase in the amount of GlcN 6-P during encystment, which is a positive allosteric effector for the synthesis of UDP-GlcNAc. The large relative increase in UDP-GlcNAc is consistent with this point of regulatory control during encystment. These amino sugar phosphate intermediates of this pathway displayed a co-ordinated rise during the peak of encystment at 24 h, and went back to almost non-encysting level at 48 h when encystment is completed. The results shown are representative of three independent experiments that displayed essentially similar changes during encystment.

 
It is noteworthy that the largest absolute increase is in levels of GlcN 6-P, the first substrate unique to this pathway, which increases fivefold during encystment. GlcN 6-P is believed to be the positive effector for UDP-GlcNAc pyrophosphorylase, the pathway's putative rate-controlling enzyme that catalyses the conversion of GlcNAc 1-P into UDP-GlcNAc. The greatest relative increase (approx. ninefold during encystment) in concentrations of the amino sugar phosphate pathway intermediates is in UDP-GlcNAc, the product of UDP-GlcNAc pyrophosphorylase. This is consistent with the postulated co-ordinated metabolic control of the pathway by the concentration of its first intermediate activating the pyrophosphorylase.

In all of our measurements of amino sugar phosphate intermediates in Giardia, there are higher levels of UDP-GalNAc than of its precursor, UDP-GlcNAc (Fig. 3). If the synthesis of UDP-GalNAc from the apparent rate-controlling substrate UDP-GlcNAc were controlled by the withdrawal of the UDP-GalNAc for synthesis of the cell wall, the amount of the latter would have been less than that of the former. However, because the amount is greater, it implies that this ultimate conversion step of the pathway is through an active synthetic process that favours the production of UDP-GalNAc from its immediate precursor, UDP-GlcNAc.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Giardia is a major human enteric pathogen, and understanding the details of its life cycle is a priority for learning to control giardiasis. A crucial element in the Giardia life cycle for its survival outside of its host is the cyst wall formation (encystment) that occurs prior to the pathogen exiting the host. The cell wall formed during encystment contains polymers of GalNAc (Gerwig et al., 2002). The enzymes involved in synthesis of amino sugars leading to GalNAc have higher activity in vitro when isolated from encysting Giardia relative to quiescent Giardia (Macechko et al., 1992) and the transcripts for the first five of these enzymes are upregulated during encystment (Lopez et al., 2003). Furthermore, GlcN 6-P upregulates the activity of the putative rate-controlling enzyme UDP-GlcNAc pyrophosphorylase in vitro, and GlcN 6-P concentrations increase in encysting Giardia in vivo (Bulik et al., 2000). However, these previous findings could not address several questions that remained unsolved regarding regulation of the amino sugar synthetic pathway during encystment in vivo. The only method with the potential to allow a practical simultaneous analysis of hexosamine phosphates in Giardia in vivo was the capillary electrophoresis technique that we applied previously to measure GlcN 6-P in encysting Giardia (Bulik et al., 2000). Therefore, we adapted this technique for the measurement of the known sugar phosphate intermediates for UDP-GalNAc synthesis. Authentic standards for these amino sugar phosphate intermediates were used to validate the technique with regard to the retention time for each compound of interest, its limit of detection, sensitivity and range of linearity. However, the small amounts of some of these intermediates in Giardia relative to other irrelevant peaks of compounds in Giardia presented several challenges that needed to be addressed. For example, the large amount of other charged materials in the sample relative to the compounds of interest can cause slight shifts in the elution time for capillary electrophoresis; therefore, we ran all samples with and without added standards, both individually and as a mixture, so that the identity of each peak was confirmed by its co-elution with authentic standard. Also, each sample was treated with phosphatase, and the losses of our peaks of interest confirmed that they were phosphorylated, providing further independent confirmation of the identity for the sugar phosphates, but not for the three amino sugar phosphates carrying UDP, which did not hydrolyse. A phosphatase inhibitor was added to an aliquot of each of our samples as they were thawed, before homogenization, so that the isolated amounts of sugar phosphate intermediates would represent the true amount in the Giardia cells before their processing released endogenous phosphatase. The efficacy of the inhibitor in Giardia in each of the stages of encystment was tested in independent samples to which known amounts of each of the authentic standards had been added. Each standard was quantitatively recovered in the samples to which inhibitor was added.

The co-ordinated increase and decrease through Giardia's encystment process was specific to the metabolic intermediates from GlcN 6-P through UDP-GalNAc, and not observed in similar compounds that are not part of this pathway, such as GlcN 1-P and GalN 1-P. This provides additional confidence that the method correctly measures the compounds of interest, and that the pathway shown in Fig. 1 correctly describes metabolic changes in Giardia that are specific to encystment.

These changes in the level of the sugar phosphate intermediates suggest co-ordinated metabolic control of UDP-GalNAc synthesis associated with encystment. At 24 h all of the pathway intermediates increased in concert when encystment was at its peak, and most of them returned to non-encysting levels by 48 h, when encystment approached completion. These in vivo findings are consistent with the findings of Macechko et al. (1992) that the activity of pathway enzymes in vitro increased by 24 h and sharply decreased by 48 h.

The increase in pathway intermediates was greatest in absolute amount for GlcN 6-P, which is a positive allosteric effector in the synthetic direction for the pathway's putative rate-controlling enzyme UDP-GlcNAc pyrophosphorylase (Bulik et al., 2000). This tends to push the pathway in favour of cyst wall synthesis. In this pathway, the UDP-GlcNAc pyrophosphorylase converts GlcNAc 1-P to UDP-GlcNAc. The proposed regulatory role of this enzyme in this pathway, both with regard to being rate-limiting and of being allosterically activated during encystment, is supported by our finding a ninefold increase (highest of any of the sugar phosphate intermediates) in UDP-GlcNAc during encystment.

In vitro, the equilibrium of the epimerase reaction UDP-GlcNAc to UDP-GalNAc results in higher levels of UDP-GlcNAc than UDP-GalNAc, which suggested that in Giardia the reaction might be ‘pulled’ in the direction of synthesis of UDP-GalNAc by depletion of UDP-GalNAc as it is used in cyst wall synthesis. However, we detected more UDP-GalNAc at 24 h than UDP-GlcNAc, its direct precursor. This suggests that there is a shift in the preferred direction of synthesis by the UDP-GlcNAc 4'-epimerase, the enzyme immediately following the pyrophosphorylase in this pathway. These results suggest that in the cytoplasm of encysting Giardia there are kinetic pressures (such as increased amounts of UDP-GalNAc) forcing the reaction in the direction that supports (‘pushes’) cell wall synthesis. Because this major distinction between in vitro and our in vivo observations is based on differences in measurements of very small quantities, it warrants confirmation by an independent form of analysis. These control points of metabolic pathways essential for encystment in Giardia may be candidate targets for specific inhibitors that may be of therapeutic value against Giardia.


   ACKNOWLEDGEMENTS
 
This study was supported by AI4775 (E. L. J.) and HD13021 (D. S. N.).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bessman, S. P. (1974). An automated apparatus for quantitative aliquot measurement of phosphorylated intermediates. Anal Biochem 59, 524–532.[Medline]

Bulik, A. D., van Ophem, P., Manning, J. M., Shen, Z., Newburg, D. S. & Jarroll, E. L. (2000). UDP-N-acetylglucosamine pyrophosphorylase, a key enzyme in encysting Giardia, is allosterically regulated. J Biol Chem 275, 14722–14728.[Abstract/Free Full Text]

deRegnier, D. P., Cole, L. & Schupp, D. G. (1989). Viability of Giardia cysts suspended in lake, river, and tap water. Appl Environ Microbiol 55, 1223–1229.[Medline]

Farthing, M. J. G. (1994). Giardiasis as a disease. In Giardia: From Molecules to Diseases, pp. 15–37. Edited by R. C. A. Thompson, J. A. Reynoldson & A. J. Lymbery. Wallingford, UK: CAB International.

Farthing, M. J. G. (1996). Giardiasis. Gastroenterol Clin North Am 25, 493–515.[Medline]

Feely, D. E., Erlandsen, S. L. & Chase, D. G. (1984). Structure of the trophozoite and cyst. In Giardia and Giardiasis, pp. 3–31. Edited by S. Erlandsen & E. Meyer. New York: Plenum.

Gerwig, G. J., van Kuik, J. A., Leeflang, B. R., Kamerling, J. P., Viegenthart, J. F. G., Karr, C. D. & Jarroll, E. L. (2002). The Giardia intestinalis filamentous cyst wall contains a novel {beta}(1-3)-N-acetyl-D-galactosamine polymer: a structural and conformational study. Glycobiology 12, 499–505.[Abstract/Free Full Text]

Jarroll, E. L., Macechko, P. T., Steimle, P. & 8 other authors (2001). Regulation of carbohydrate metabolism during Giardia encystment. J Eukaryot Microbiol 48, 22–26.[Medline]

Keister, D. B. (1983). Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Trans R Soc Trop Med Hyg 77, 487–488.[Medline]

Lane, S. & Lloyd, D. (2002). Current trends in research into the waterborne parasite Giardia. Crit Rev Microbiol 28, 123–147.[Medline]

Lopez, A. B., Sener, K., Jarroll, E. L. & van Keulen, H. (2003). Transcription regulation is demonstrated for five key enzymes in Giardia intestinalis cyst wall polysaccharide biosynthesis. Mol Biochem Parasitol 128, 51–57.[CrossRef][Medline]

Macechko, P. T., Steimle, P. A., Lindmark, D. G., Erlandsen, S. L. & Jarroll, E. L. (1992). Galactosamine-synthesizing enzymes are induced when Giardia encyst. Mol Biochem Parasitol 56, 301–310.[CrossRef][Medline]

Schupp, D., Januszka, M., Sherlock, L., Stibbs, H., Meyer, E., Bemrick, W. & Erlandsen, S. (1988). Production of viable Giardia cysts in vitro: determination by fluorogenic dye staining, excystation and animal infectivity in the mouse and Mongolian gerbil. Gastroenterology 95, 1–10.[Medline]

Received 10 November 2003; revised 30 January 2004; accepted 30 January 2004.