Chlorate: a reversible inhibitor of proteoglycan sulphation in Chlamydia trachomatis-infected cells

The identity of host-cell receptors mediating chlamydial attachment has remained undetermined and controversial. In previous studies, it has been suggested that heparan sulphate (HS) glycosaminoglycan (GAG) serves as an initial receptor for the adherence and subsequent infectivity of chlamydia (Davis & Wyrick, 1997). There is still considerable debate regarding whether HS is present on chlamydiae (Zhang & Stephens, 1992) or the host cell (Su et al., 1990). Moreover, the sulphate constituents of proteoglycans may play an important role in the binding and infectivity of chlamydia. This raises the possibility that variation in sulphation might have considerable biological ramifications. The most direct way to examine this would be to establish in vitro conditions that promote undersulphation. This consists of specifically inhibiting 3'-phosphoadenyl 5'-phosphosulphate (PAPS; the active form of sulphate) using different chemicals. Chlorate appeared to be the most effective of the substances used in reducing the sulphation of a variety of macromolecules (Safaiyan et al., 1999). It competes with sulphate ions in binding to ATPsulphurylase, the enzyme which initiates the formation of PAPS (Klaassen & Boles, 1997).

To better understand the importance of sulphated glycosaminoglycans in chlamydial infectivity, and whether they are present on chlamydia or the host cell, we investigated the effect of various concentrations of chlorate on the infectivity of Chlamydia trachomatis serovars LGV1, LGV2 and E grown in two cell lines and whether any effect was reversible.

C. trachomatis serovars.
Serovars LGV1, LGV2 and E64 were grown in semi-confluent McCoy cells for 48 h in Minimum Essential Medium Eagle (EMEM) medium supplemented with 10 % fetal calf serum (FCS) and cycloheximide (2 µg ml^-1). To infect E64, the tissue culture flasks were centrifuged for 1 h at 2000 g. All serovars were incubated at 37 °C in 5 % CO₂ for 4872 h. Elementary bodies (EBs) were harvested and purified according to Caldwell et al. (1981).

Cell lines.
McCoy cells (mouse fibroblast cell line), HeLa 229 cells (cervical carcinoma) and Hec1B cells (endometrial carcinoma) were all obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA). They were passaged regularly in EMEM containing 10 % FCS and maintained at 37 °C in 5 % CO₂.

Inhibition of sulphation.
HeLa cell monolayers were grown to confluence and then maintained for 48 h in EMEM medium supplemented with 2 % FCS and 30 mM sodium chlorate (NaClO₃). Cells were removed from their culture surface, distributed to 24-well tissue culture plates containing sterile coverslips and grown to confluence in the presence of various concentrations of NaClO₃ ranging from 5 to 200 mM. The confluent cells were then infected with EBs. Following incubation for 48 h, cells were fixed and stained with antichlamydial mAb (IMAGEN chlamydia, DAKO, UK); the number of inclusions was counted using fluorescence microscopy at x400 magnification. The infectivity of chlorate-treated cells was compared to an untreated control.

To investigate whether sulphated GAGs are present on the chlamydia, we examined the effect of chlorate treatment on the EBs. C. trachomatis serovar LGV1 was grown in the presence of various concentrations of NaClO₃ (1070 mM) for 48 h at 37 °C in 5 % CO_2. The EBs were isolated by centrifugation at 30 000 g for 1 h at 4 °C and the resulting pellet was resuspended in 10 µl PBS. The EB suspension was used to infect confluent cell monolayers which were incubated at 37 °C in 5 % CO₂ for 48 h. The infected cell monolayers were fixed and stained as mentioned above.

Restoration of infectivity.
For the reversal of sulphation inhibition, chlorate-treated HeLa and Hec1B cells were supplemented with different concentrations of sodium sulphate, ranging from 10 to 50 mM, infected with EBs of serovars LGV1 and LGV2 and incubated as mentioned above.

Chlorate is known to be an in vitro inhibitor of ATPsulphurylase, the first enzyme in the biosynthesis of PAPS, the high-energy sulphate donor in biological reactions (Leong et al., 1995). In a previous work, Baeuerle & Huttner (1986) mentioned that treatment of various cell cultures with chlorate resulted in inhibition of protein sulphation. Chlorate did not inhibit protein synthesis and did not exhibit any other toxic effects, even after prolonged incubation periods. In addition, Keller et al. (1989) reported that cells grown in the presence of chlorate produced HS GAG chains containing only about 8 % of the sulphate normally present and which had lost the ability to bind fibronectin. Furthermore, the iduronic acid content of HS produced in the presence of chlorate was reduced to less than 7 % as compared to the 36 % from untreated cells. It was concluded that the use of chlorate could be valuable for the study of the biosynthesis and structurefunction relationships of sulphated GAGs.

The present study has shown that even low concentrations of chlorate treatment decreased the sulphation of GAGs of the host cells and resulted in a marked reduction of infectivity of both LGV serovars; a slight reduction of E64 infectivity occurred only in the presence of very high chlorate concentrations. These results show the crucial role of sulphated GAGs in the infectivity of LGV serovars of C. trachomatis. For the LGV1 serovar, infectivity progressively decreased with increasing chlorate concentration up to 100 mM (Fig. 1). The LGV2 serovar had an infectivity reduction pattern similar to LGV1 in response to the various concentrations of chlorate tested (data not shown). For the E64 serovar, there was no significant change in infectivity when a concentration of 70 mM chlorate was used, indicating that a decrease in sulphation had little effect on infectivity of this serovar. Even at a concentration of 100 mM chlorate, only a 30 % decrease of infectivity was obtained. Concentrations greater than 100 mM chlorate resulted in detachment of many of the cells and a rounded appearance of others and was probably due to excess salt concentration. This result showed that the binding of E64 to host cells is not as dependent on the degree of sulphation of the GAGs. Moreover, for all serovars tested, the inhibition of infectivity by various concentrations of chlorate was never complete, indicating that the interaction of the organism with host cells could be concomitantly mediated by other adherence factors.

(27K): Fig. 1. Effect on the infectivity of C. trachomatis serovars LGV1 (striped bars) and E64 (white bars) using sodium chlorate-treated HeLa cells. Control, untreated cells. Results are given as the mean of three experiments; error bars represent the standard error.

In contrast, growing the three tested C. trachomatis serovars following treatment of EBs with up to 70 mM chlorate did not show a considerable difference in infectivity compared to untreated EBs. The effect of chlorate on LGV1 EBs is included as an example in Fig. 2. This suggested a lack of GAGs on the chlamydia surface and supported the conclusions from previous work (Su et al., 1996; Taraktchoglou et al., 2001).

(32K): Fig. 2. Effect of different concentrations of sodium chlorate on the infectivity of C. trachomatis LGV1 EBs. Control (C), untreated cells. HeLa cells, solid bars; Hec1B cells, hatched bars. Results are given as the mean of three experiments.

Supplementing chlorate-treated host cells with sodium sulphate restored sulphation and led to a partial restoration of the infectivity. For the LGV1 serovar, 10 mM sodium sulphate was ideal for rescuing the infectivity (85 and 60 % of control in the presence of 30 and 50 mM chlorate, respectively, in HeLa cells), thus the effect of chlorate appeared to be reversible. Higher concentrations of up to 50 mM sulphate did not lead to any further significant restoration of infectivity of the chlorate-treated cells (Fig. 3). This result suggested that sulphation levels can play an important role in the infectivity of C. trachomatis. Notably, restoration of infectivity was seen less with serovar LGV2 than with serovar LGV1 in response to the addition of sodium sulphate to the chlorate-treated cells (data not shown). This might reflect a difference in the quantity and/or fine structure of GAGs required by each serovar to mediate infection.

(28K): Fig. 3. Effect of different concentrations of sodium sulphate on the infectivity of serovar LGV1 of C. trachomatis following 30 mM sodium chlorate treatment of host cells. HeLa cells, solid bars; Hec1B cells, hatched bars. Results are given as the mean of three experiments.

In conclusion, our results strongly suggest that C. trachomatis does not produce HS and that HS of the host cells is necessary and sufficient to mediate chlamydial infectivity. The study also highlighted the essential role played by the sulphate constituents of the host-cell GAGs in the infectivity of LGV serovars, and to a lesser extent of serovar E.