RESEARCH ARTICLE

A comparative analysis of the attachment of Leptospira interrogans and L. borgpetersenii to mammalian cells Gabrielle I. Andrade & Paul D. Brown

IMMUNOLOGY & MEDICAL MICROBIOLOGY

Basic Medical Sciences Department, Faculty of Medical Sciences, University of the West Indies, Kingston, Jamaica

Correspondence: Paul D. Brown, Basic Medical Sciences Department, Faculty of Medical Sciences, University of the West Indies, Mona Campus, Kingston 7, Jamaica. Tel.: +876 927 2290; fax: +876 977 7852; e-mail: [email protected] Received 14 December 2011; revised 15 February 2012; accepted 2 March 2012. Final version published online 4 April 2012. DOI: 10.1111/j.1574-695X.2012.00953.x Editor: Eric Oswald Keywords binding; mediators; lectins; HEp-2 cells.

Abstract Leptospirosis, the world’s most ubiquitous zoonosis, is caused by pathogenic Leptospira. As microbe–host interactions are specific in pathogenesis, it is likely that there are several molecules mediating the attachment of the Leptospira to mammalian cells. In this study, we analysed the attachment of Leptospira interrogans serovar Portlandvere and Leptospira borgpetersenii serovar Jules to untreated HEp-2 cells or HEp-2 cells treated with the various enzymes, lectins or sugars and to integrins aVb3 and a5b1, relative to control wells. We found that both serovars bound equally well to HEp-2 cells; however, serovar Jules showed a higher level of attachment to integrins. Both serovars showed an increase in attachment to HEp-2 cells coated with lectins peanut agglutinin, Ulex europaeus agglutinin, soybean agglutinin and Erythrina cristagalli agglutinin (p < 0.05); in the case of Concanavalin A, Jules showed an increase, while Portlandvere showed a significant decrease in attachment. Trypsinizing monolayers resulted in a decrease in attachment for both serovars, while when chondroitinase, neuraminidase and heparinase were used an increase in attachment was recorded. Leptospires coated with sugars showed a decrease in attachment. These results show that serovar Jules’ general greater affinity for the mediators examined may suggest a greater potential for virulence over serovar Portlandvere.

Introduction Since its characterization as an infectious disease in humans by Adolf Weil in 1886, annual cases of leptospirosis have reached 500 000, making it the world’s most common zoonosis with a mortality rate of 5–20% (WHO, 2003). The numerous serovars that make the development of suitable and affordable human and animal vaccines challenging for developing countries and the lack of rodent pest control in slums and overpopulated urban areas, as well as the dependency on locally produced agricultural products like rice and sugarcane, leave Third World countries at a disadvantage (Levett, 2001; Bharti et al., 2003). Jamaica’s first reported case of leptospirosis occurred in 1953 (Bras, 1955). Since then, the disease reached epidemic proportions in 1979 after heavy June floods (PAHO, 1979), and a decade later, serological studies showed predominant serovars to be Portlandvere, FEMS Immunol Med Microbiol 65 (2012) 105–115

Jules, Canicola and Icterohaemorrhagiae. At the time, Jules had only been reported in Jamaica and more recent unpublished data show it has a seroprevalence of 25%. Portlandvere has a seroprevalence of 12%. Disease incidence in Jamaica is 6 per 100 000 population with a peak in reported cases during the second rainy season of the year, between October and December (Brown et al., 2011). Leptospira are not intracellular organisms, and attachment has been implicated to be the basis of mediating pathogenic activities (Merien et al., 2000). In fact, pathogenic Leptospira attach with greater affinity to mammalian cells than their saprophytic counterparts (Thomas & Higbie, 1990), and several mediators have been identified, including lipopolysaccharide (Chirathaworn et al., 2007), leptospiral immunoglobulin-like (Lig) proteins (Matsunaga et al., 2003; Choy et al., 2007; Lin & Chang, 2008). It has been shown by the Ito and Yanagawa study in 1987 ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

106

that molecules like fibronectin which bind to Leptospira through the several surface proteins of the spirochaete and which bind to integrins form some sort of a bridge between the two cells aiding in their union (Ito & Yanagawa, 1987). This attachment is mediated through a precise union between several bacterial surface proteins and the host cell’s extracellular matrix (ECM) components (Barbosa et al., 2006; Chirathaworn et al., 2007; Choy et al., 2007) and has been shown to be an active process as heat- and formalin-treated leptospires do not attach to molecules of the ECM (Ito & Yanagawa, 1987). Although the interaction between the spirochaete Borrelia burgdorferi and integrins has been more extensively studied (Alugupalli et al., 2001; Defoe & Coburn, 2001), little has been shown for the interaction between Leptospira and these host cell surface molecules. An indication of the possible interaction between leptospires and integrins was posited by Cinco et al. (2002), who reported that the I-domain of Mac-1 (a CR3 integrin involved in promoting phagocytosis that had been shown to bind iC3b, ICAM-1, fibrinogen and fibronectin) was involved in the interaction between Leptospira and neutrophils. This group also noted that the monosaccharide D-mannose could bind to both the lectin-like domain of Mac-1 and the spirochaetes, a mechanism that would aid in the docking of the spirochaete. It was previously shown that Mac-1 could bind to beta-glucans, mannose, N-acetyl-D-glucosamine and lipopolysaccharide via its lectin-like domain (Diamond & Springer, 1993; Thornton et al., 1996). Proteoglycans consist of glycosaminoglycans (GAGs), which are perpendicularly and covalently bonded to core proteins. GAGs are long unbranched polymers of repeating disaccharide units and are located on the cell membrane. Some mammalian GAGs have been associated with binding of bacteria such as heparin with Pseudomonas aeruginosa (Plotkowski et al., 2001) and Listeria monocytogenes (Alvarez-Dominguez et al., 1997). Generally, to test the importance of a certain GAG in attachment, an enzyme is used to cleave the molecule from a host cell and then an attachment assay is conducted to measure the difference between the enzyme-treated cells and untreated cells. To date, only one other study has analysed any role GAGs play in leptospiral attachment to host cells. It was shown that the use of chondroitinase resulted in a significant decrease in attachment of Leptospira interrogans serovar Copenhageni to HMEC and HEp-2 cells (Breiner et al., 2009). In this study, we analysed the attachment of L. interrogans serovar Portlandvere and Leptospira borgpetersenii serovar Jules to untreated HEp-2 cells or HEp-2 cells treated with the enzymes (trypsin, chondroitinase ABC, heparinise and neuraminidase) or lectins [Ulex europaeus (UEA), Glycine max (SBA), Arachis hypogaea (PNA), ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

G.I. Andrade & P.D. Brown

Erythrina cristagalli (ECA) and Canavalina ensiformis Type IV (Concanavalin A; ConA)] and to integrins aVb3 and a5b1. In addition, the binding of HEp-2 cells by leptospires coated with the sugars N-acetylgalactosamine, N-acetylglucosamine, L-fucose, L-mannose and a-D-methylglucoside was also assessed.

Methods Bacterial strains and cell line

Leptospira interrogans serovar Portlandvere strain MY1039 and L. borgpetersenii serovar Jules strain Jules were grown at 30 °C in EMJH medium (Difco) enriched with 10% rabbit serum. Viability and enumeration were carried out using dark-field microscopy and Petroff-Hausser chamber. Strains had been laboratory-passaged for 6 months prior to the start of the experiments. The HEp-2 cell line was a generous gift from Carol Khan (Microbiology Department, University of the West Indies, Mona, Kingston, Jamaica) and was cultured at 37 °C and 5% CO2 in DMEM supplemented with 10% FBS, 100 lg mL 1 streptomycin and 100 units mL 1 penicillin. Viability of HEp-2 cells was assessed by trypan blue staining. To coat the wells of the microtitre plate, HEp-2 cells were made up to 1 9 105 cells mL 1 in complete media without antibiotics and incubated overnight then checked for adherence and confluence using an inverted microscope (Aus Jena). Primary and secondary antibodies

Acute human anti-Leptospira sera with Portlandvere or Jules as the sole or main sera reactor were a generous gift from Tricia Fraser (Ministry of Agriculture and Fisheries, Hope Gardens, Jamaica) and were used at a dilution of 1 : 5000, based on empirical observations. Reserve APTM phosphatase-labelled goat antibody to human IgM was obtained from Kirkegaard & Perry Laboratories Inc. (KPL, Gaithersburg, MD). Goat antihuman integrin aV IgG antibody and antihuman integrin a5 IgG antibody were obtained from R & D systems (NE Minneapolis, MN). Reserve APTM phosphatase-labelled antibody to goat IgG, produced in rabbit, was obtained from KPL. Attachment of Leptospira to mammalian cells

The adherence assay of Merien et al. (2000) was consulted with the following modifications. All washes were carried out four times with phosphate-buffered saline– Tween 20 (PBST). Monolayers of HEp-2 cells in sterile Falcon Microtest flat-bottom 96-well plates were incubated with 50 lL Leptospira serovars at a concentration of 106 cells mL 1 for 1 h at 37 °C. Wells were washed to FEMS Immunol Med Microbiol 65 (2012) 105–115

Attachment of Leptospira to mammalian cells

remove unbound spirochaetes. To measure attachment, 50 lL of human anti-Leptospira serum was added and left to incubate for 1 h at 37 °C. Plates were washed and 50 lL goat antihuman IgM-alkaline phosphatase conjugate was added, and the wash repeated. Fifty microlitres of para-nitrophenylphosphate substrate in DEA buffer (pH 9.8) was then added. Plates were left to incubate for 30 min, and the reaction was stopped by adding 50 lL of 0.75 M NaOH. Plates were then read with a Wallac 1420 multilabel microplate reader at 405 nm. Attachment of Leptospira to fibronectin

The following shows the changes made to the established attachment assay of Guo et al. (1995). Microtitre plates were coated overnight at 4 °C with 100 lL of a 0.5, 1 and 10 lg mL 1 solution of human fibronectin (R&D Systems). Uncoated wells were used as blanks. Plates were gently washed four times with PBST (all washes were carried out in this manner) and blocked with 200 lL of 0.05% skimmed milk (Lasco, Jamaica) and incubated for 1 h at room temperature; then, the wells were washed again. Fifty microlitres of each Leptospira serovar at a concentration 106 cells mL 1 was added to wells and left to incubate for 1 h at 37 °C. Plates were then washed again to remove unbound bacteria, and attachment was measured as mentioned as described previously. Attachment of Leptospira to integrins

On the basis of preliminary experiments, it was decided to use a concentration of 106 cells mL 1 Leptospira and 0.8 lg mL 1 of each integrin (avb3 and a5b1) as these proved to be the minimum concentrations necessary to give maximum absorbance. Fifty microlitres of purified integrins (0.8 lg mL 1) was immobilized by adding the solution to prechilled 96-well microtitre plates and incubating overnight at 4 °C. Plates were gently washed four times with PBST (all washes were carried out in this manner) and blocked with 200 lL of 0.05% skimmed milk and incubated for 1 h at room temperature; then, the wells were washed again. Fifty microlitres of each Leptospira (106 cells mL 1) was added to wells and left to incubate for 1 h at 37 °C. Plates were then washed again to remove unbound bacteria, and the previously described ELISA was carried out to assess the extent of binding. To test the efficacy of coating, the integrin-coated plates were also subjected to a similar ELISA with 50 lL of purified goat antihuman aV and a5 molecules used to attach to the plates coated with integrins and blocked before washing and adding 50 lL AP-labelled rabbit antigoat IgG (0.5 mg mL 1). Colour development and measurement were performed as before. FEMS Immunol Med Microbiol 65 (2012) 105–115

107

Inhibition of binding of Leptospira to cells treated with exogenous lectins

Prior to each assay, cell growth media was removed from the wells of the microtitre plate and cells were washed with PBST. Aliquots of 100 lL of 10 lg mL 1 solution in PBS of the following lectins were added: U. europaeus (UEA), G. max (SBA), A. hypogaea (PNA), E. cristagalli (ECA) and C. ensiformis Type IV (ConA). For this analysis, seven wells of each lectin per strain were included, with untreated wells serving as controls, and the plates were left to incubate 1 h at 37 °C. Plates were washed again with PBST before adding 50 lL of each Leptospira (106 cells mL 1) to wells, which were then left to incubate for 1 h at 37 °C and attachment measured as before. Assessment of binding of Leptospira preincubated with sugars to HEp-2 cells

Leptospires (1 mL each; 106 cells) were pelleted and resuspended in 1 mL of a 100 lg mL 1 solution of each of N-acetylgalactosamine, N-acetylglucosamine, L-fucose, L-mannose and a-D-methylglucoside. They were left to incubate for 1 h at room temperature. Cell growth media was removed from the wells of the microtitre plate containing confluent monolayers of HEp-2 cells and they were washed four times with PBST. Fifty microlitres of each treated Leptospira suspension was added to the wells, and the plates were left to incubate 1 h at 37 °C. Untreated leptospires added to HEp-2-coated wells served as controls. Plates were gently washed four times with PBST, and the attachment assay was continued as previously described. Assessment of binding of Leptospira to enzyme-treated HEp-2 cells

The following assay was modified after Thomas & Higbie (1990). Cells were washed with PBST, and 50 lL of each of the following enzymes (chondroitinase ABC, trypsin and heparinase I at 125 lg mL 1, and neuraminidase at 12.5 lg mL 1) was added to a minimum of seven wells per strain and incubated for 1 h at 37 °C. Untreated HEp-2-coated wells served as controls. Plates were washed as before and 50 lL of each Leptospira strain at a concentration of 106 cells mL 1 was added to wells and left to incubate for 1 h at 37 °C and the attachment protocol followed to measure binding. Statistical analyses

Each parameter was replicated in seven wells, and each experiment repeated on three separate occasions to confirm reproducibility. The results are presented in paired ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

108

(a)

0.350 0.300

Empty wells E H Hep-2 coated wells

**

**

0.250 A405 nm

graphs. The first of the pair shows OD values of attachment of the spirochaetes to treated cells (or wells) and they were compared to untreated cells (or wells) to analyse the significance. The second of the pair presents the percentage increase in attachment and a comparison between the serovars’ change in attachment. This percentage change was calculated for each parameter by subtracting the OD of the untreated well from the OD of the treated well and dividing this result by the OD of the untreated well and multiplying by 100. For each of the paired graphs, the standard error of the mean was then calculated and the P values were obtained using Student’s t-test with two tails. Statistical significance was regarded as those values of P < 0.05. Values of P < 0.01 were deemed very statistically significant.

G.I. Andrade & P.D. Brown

0.200 0.150 0.100 0.050 0.000 Julees

(b)

Portlan ndvere

160 140

Assessing the binding of Leptospira to HEp-2 cells

Leptospira borgpetersenii serovar Jules attached to the cells with a slightly higher affinity than L. interrogans serovar Portlandvere. The optical density values and the percentage increase in attachment of the serovars when made to attach to cells compared to the empty wells were calculated and represented in Fig. 1a and b, respectively. The increase in binding was found to be significant for both serovars (Fig. 1a). Portlandvere showed a slightly higher percentage increase in attachment, but there was no significant difference between the two strains (Fig. 1b). Assessing the binding of Leptospira to fibronectin

Both serovars showed an increase in attachment to fibronectin up to a concentration of 1 lg mL 1; then, a slight decrease was seen thereafter (Fig. 2a). When compared to wells with no fibronectin, these increases were all significant (P < 0.01). Serovar Jules showed a greater percentage increase in its level of attachment to fibronectin when compared to serovar Portlandvere throughout the assay, and the differences between the two serovars’ increase in attachment were shown to be significant (P < 0.01) at every concentration (see Fig. 2b). Assessment of binding of Leptospira to integrins

There was an increase in the attachment of serovars Portlandvere and Jules to wells that had been coated with the integrins tested. The highest readings were seen for the attachment of serovar Jules to aVb3, which showed a ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

% increase

120

Results

100 80 60 40 20 0 Jules

P Portlandvere

Fig. 1. The attachment of serovars Jules and Portlandvere to wells coated with HEp-2 cells. (a) After coating of wells with the HEp-2 cells as described in materials and methods, they were infected with each spirochaete, attachment was measured using human antiLeptospira serum, and then goat antihuman IgM antibodies linked to alkaline phosphatase. Para-nitrophenylphosphate was the substrate used to give the colour change, which was measured by the plate reader. The statistical analysis of the OD readings was completed, and the means (±SE) represented in the graph and a two-tailed Student’s t-test used to calculate P. Results showed there was a significant increase in attachment of both serovars to the cells when compared to their attachment to empty wells (P < 0.01 for both serovar Jules and Portlandvere). (b) The procedure was completed as described for (a). The percentage increase in attachment was calculated, the statistical analysis was completed, and the means (±SEM) represented in the graph and a two-tailed Student’s t-test used to compare the two serovar’s percentage increase. Results showed no statistical difference between the percentage increase in attachment between the two serovars (P = 0.814).

38% increase in attachment when compared to the attachment of the spirochaete to uncoated wells. This was statistically significant (P < 0.01). Jules also showed a 27% increase in attachment in wells coated with a5b1 when compared to empty wells (P < 0.05) (Fig. 3a). There was only a major difference between the serovars’ attachment to aVb3, where Jules showed a significantly greater percentage increase in attachment to the integrin when compared to Portlandvere’s increase (P < 0.05) (Fig. 3b). FEMS Immunol Med Microbiol 65 (2012) 105–115

Attachment of Leptospira to mammalian cells

109

(a) 0.250

(a) 1.6

**

1.4

** **

1.0

**

**

0.8 0.6 0.4

0

0.5

1

α5β1 coated well

0.150 0.100

Jules

10 60

(b)

Conc. fibron nec n μg/ml

**

* **

**

700 600

% increase

50

% increase

αVβ3 coated well

0.000

0

800

*

0.050

Jules Portlan ndvere

0.2

(b) 900

**

0.200 A405 nm

A405 nm

1.2

Empty well

**

*

Portlandvere Jules Portlandvere

40 30 20 10 0

500

–10

400 300 200

Jules Portlan ndvere

100 0 0.5

1 Conc. fibron nec n μg mL–1

10

Fig. 2. The attachment of serovar Jules and Portlandvere to wells coated with increasing concentrations of fibronectin. (a) Varying concentrations (0.5, 1 and 10 lg mL 1) of fibronectin were used to coat wells, and each leptospire was added followed by the reveal ELISA described previously. The means with standard errors of the OD measurements of attachment for each serovar when compared to the control wells without fibronectin. Results were plotted in a line graph to show the trend an increase in fibronectin concentration made. The two-tailed Student’s t-test was used to calculate the P values. They showed a significant increase in attachment for serovar Jules and Portlandvere at all concentrations of fibronectin analysed; 0.5, 1 and 10 lg mL with P < 0.01 in all cases. (b) The method was carried out as described previously for (a). The means with standard errors of the percentage increase in attachment for each serovar when compared to the control wells without fibronectin. Results were plotted in a line graph to show the trend an increase in fibronectin concentration made. The two-tailed Student’s t-test was used to calculate the P values of the comparison between the serovars’ increase in attachment at each concentration. They showed a significantly higher percentage increase in attachment for serovar Jules when compared to Portlandvere at all concentrations of fibronectin analysed; 0.5, 1 and 10 lg mL 1 with P < 0.01 in all cases.

Lectin inhibition assays

In general, both serovars responded with little change in binding when incubated with any of the lectin-coated HEp-2 cells except for ConA (P < 0.01), SBA (P < 0.05 for FEMS Immunol Med Microbiol 65 (2012) 105–115

αVβ3

α5β1

Fig. 3. The attachment of serovars Jules and Portlandvere to aVb3 and a5b1. (a) The graph depicts the ODs of wells coated with 0.8 lL mL 1 of the integrin aVb3 and wells coated with 0.8 lL mL 1 of a5b1 then infected with serovars Jules and Portlandvere following the aforementioned protocol with blank blocked wells as controls. There was a statistically significant increase in attachment exhibited by Jules to aVb3 (P < 0.01) and a5b1 (P < 0.05). (b) The graph depicts the percentage increase in attachment of serovars Jules and Portlandvere to wells coated with 0.8 lL mL 1 of the integrin aVb3 and wells coated with 0.8 lL mL 1 of a5b1 when compared with blank blocked wells as controls. The two-tailed Student’s t-test was performed to find the P values when comparing serovar Jules’ percentage increase to that of serovar Portlandvere. Serovar Jules demonstrated a significantly higher percentage increase in attachment to aVb3 than serovar Portlandvere (P < 0.05).

Jules and P < 0.01 for Portlandvere) and in the case of Jules, ECA (P < 0.05), where both serovars showed significant increases in binding (Fig. 4a). When comparisons were made between the two serovars, Jules had a significantly higher increase than Portlandvere in its attachment to ConA (P < 0.05), and there were significant differences in how they responded to ECA and SBA (P < 0.01) (Fig. 4b). Serovar Jules showed a percentage increase in binding to ECA and SBA lectin-coated cells, while Portlandvere showed a decrease. The only decrease in binding was seen with serovar Portlandvere and HEp-2 cells incubated with PNA, ECA, UEA and SBA. Both serovars Jules and Portlandvere showed a decrease in binding after incubation with the various sugars (Fig. 5a), in particular with L-fucose (P < 0.05) and in the case of serovar Jules, with L-mannose (P < 0.01). There was no significant difference between the serovars’ response to the treatment with any of these particular sugars (Fig. 5b). ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

110

G.I. Andrade & P.D. Brown

**

0.900

**

0.800 A405 nm

0.700

Untreated ConA PNA ECA UEA SBA

0.600 0.500

*

0.400

*

**

(a) 0.300 0.250

L -fucose

α- D -meth

Glu N Ac

Gal N Ac

L -mannose

*

** *

0.150 0.100

0.300 0.200

0.050

0.100

0.000

0.000 Jules

Jules

Portlandvere Jules Portlandvere

(b) 250 200

Untreated

0.200 A405 nm

(a) 1.000

L -fucose

(b)

Portlandvere

α- D -meth

Glu N Ac

Gal N Ac

L -mannose

2 0

*

–2 % increase

% increase

–4

150 100

–6 –8 –10

50 **

**

–12 –14

0 ConA

PNA

ECA

UEA

SBA

Jules

Portlandvere

–16

–50 Fig. 4. The attachment of serovars Portlandvere and Jules to lectincoated HEp-2 cells. (a) After immobilizing the HEp-2 cells, they were coated with 100 lL of ConA TypeIV and lectins from Arachis hypogaea (PNA), Erythrina cristagalli (ECA), Ulex europaeus (UEA) and Glycine max (SBA), which were made to a concentration of 10 lg mL 1. The same ELISA was used to measure attachment of the serovars. The mean and standard error of each serovar with each lectin is shown and P values calculated as before, comparing attachment of the spirochaete to lectin-treated HEp-2 confluent well with their attachment to wells with untreated HEp-2 cells. Jules showed an increase in attachment with all assays, though only significant for ConA, ECA and SBA (P < 0.01, P < 0.05 and P < 0.05, respectively), while Portlandvere only showed a significant increase with ConA assays (P < 0.01), and only one significant decrease in attachment to wells treated with SBA (P < 0.01). (b) After immobilizing the HEp-2 cells, they were coated with 100 lL of ConA, PNA, ECA, UEA and SBA that were made to a concentration of 10 lg mL 1. The same OD measurements of attachment of the serovars were used to calculate the percentage increase in attachment. The mean and standard error of the serovars’ percentage increase in attachment were calculated using the OD values obtained from the protocol described previously in (a), and for each lectin, the P values comparing Jules’ percent increase to Portlandvere’s were calculated as before. Serovar Jules exhibited a higher percentage increase in attachment to monolayers coated with ConA than serovar Portlandvere (P < 0.05). Jules showed a percentage increase in attachment to monolayers coated with ECA and SBA, while Portlandvere showed a decrease. This difference was found to be significant for the ECA and SBA assays (P < 0.01 in each case).

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

Fig. 5. The effect of preincubation with sugars on attachment of serovars Portlandvere and Jules to HEp-2 cells. (a) One milliter of each of the leptospire samples were aliquoted into five separate tubes, were spun down and resuspended in separate solutions of L-fucose, a-D-methylglucosamine (a-D-meth), N-acetylglucosamine (Glu N Ac), N-acetylgalactosamine (Gal N Ac) and L-mannose at a concentration of 100 lg mL 1. The sample was left to incubate for 1 h at 37 °C before the attachment assay and ELISA were performed as described previously. The means and standard errors of OD values are represented in the graph. The Student’s t-test was used to calculate P values, which compared the sugar treated with untreated spirochaetes. Although there was a decrease in attachment seen with both serovar for all sugars analysed, serovar Jules showed a significant decrease in its attachment to HEp-2 cells after its incubation with L-fucose and L-mannose (P < 0.05 and P < 0.01, respectively), while serovar Portlandvere showed a significant decrease with L-fucose only (P < 0.05). (b) The OD measures from the aforementioned protocol (a) were used to calculate the percentage increase in attachment of the serovars to the monolayers after the spirochaetes incubation with the sugars when compared to untreated leptospires and the means and standard errors represented in the graph. In this instance, there was no significant difference between the percentage increase in attachment of the serovars (P > 0.5).

Enzymatic cleaving of potential receptors

As illustrated in Fig. 6a, the two strains showed differing affinities for the cleaved moieties on the cell surface. Serovar Jules showed a decrease in the level of attachment to trypsin-treated cells and significant increases in its FEMS Immunol Med Microbiol 65 (2012) 105–115

Attachment of Leptospira to mammalian cells

(a) 0.500

Untreated Chondroi nase

111

Trypsin Neuraminidase Heparinase

** *

0.450 0.400 0.350 A405 nm

0.300 0.250 0.200

Discussion

0.150 0.100 0.050 0.000 Jules 80 70 60 50 40 30 20 10 0 –10 –20 –30

Jules

Portlandvere

Portlandvere

% increase

(b)

The pattern of the degree of attachment for Portlandvere was chondroitinase-treated cells > neuraminidase-treated cells > heparinase-treated cells > trypsin-treated cells. There were no significant differences between the percentage increase in attachment of serovars Jules and Portlandvere to enzymatically treated cells (Fig. 6b).

Trypsin

Neuraminidase Chondroi nase

Heparinase

Fig. 6. The attachment of Jules and Portlandvere to enzyme-treated HEp-2 monolayers. (a) Confluent HEp-2 monolayers were washed with PBST. Each enzyme solution (made at a concentration of 12.5 lg mL 1 for neuraminidase and 125 lg mL 1 for trypsin, chondroitinase and heparinise) was added separately to the cells and left to incubate for an hour at 37 °C. The attachment assay and ELISA were carried out as previously described with each of the serovars. OD values were averaged and represented in the graph with the standard error. The Student’s t-test was used to calculate P values comparing treated wells to untreated wells. Serovar Jules showed a significant increase in attachment to wells treated with neuraminidase and heparinise (P < 0.01 and P < 0.05, respectively). (b) The OD values from the assay represented in (a) were used to calculate the percentage increase in attachment of the serovars to the wells treated with the various enzymes (trypsin, neuraminidase, chondroitinase and heparinise). A Student’s t-test was used to compare Jules’ percentage increase with Portlandvere’s for each enzyme. There were no significant differences between the two serovars’ changes in attachment to the cells before and after the use of each enzyme (P > 0.1).

attachment to cells treated with neuraminidase (P < 0.01) and chondroitinase (P < 0.05). The pattern of the degree of attachment for serovar Jules was neuraminidase-treated cells > heparinase-treated cells > chondroitinase-treated cells > trypsin-treated cells. Similarly, serovar Portlandvere also showed a decrease in attachment to cells treated with trypsin, but uniquely it did not attach well to cells treated with heparinase. The increase in Portlandvere’s attachment to neuraminidaseand chondroitinase-treated cells was insignificant as well. FEMS Immunol Med Microbiol 65 (2012) 105–115

For extracellular pathogens, attachment is often a crucial initial stage in pathogenesis. Pathogenic and saprophytic Leptospira serovars have shown varying affinities for ECM proteins (Ito & Yanagawa, 1987; Barbosa et al., 2006; Choy et al., 2007; Breiner et al., 2009). Also, the more pathogenic strains of Leptospira spp. attached to host cells with six times greater affinity than the saprophytic strains (Thomas & Higbie, 1990). Further, strains rely on different molecules to greater or lesser degrees based on their pathogenicity (Merien et al., 2000). The 36-kDa outer membrane protein of L. interrogans serovar Icterohaemorrhagiae has been shown to bind to the gelatinbinding domain of fibronectin. The molecule is only present in pathogenic strains and is lost in highly passaged pathogenic strains (Merien et al., 2000). In this study, Jules had a greater (35%) affinity for binding to fibronectin, a proven characteristic of virulence in the spirochaete, than Portlandvere (Fig. 4). This particular study showed that serovar Jules showed a higher level of attachment to the HEp-2 cell lines but it was not significant. As the integrin a5b1 is a receptor for fibronectin, analysing how the serovars’ binding pattern to fibronectin differs from their binding pattern to a5b1 was of interest. Several molecules have been implicated in the attachment of Leptospira to host cells during pathogenesis, including molecules of the spirochaete (Matsunaga et al., 2003; Barbosa et al., 2006; Choy et al., 2007; Lin & Chang, 2008), the ECM (Ito & Yanagawa, 1987; Chirathaworn et al., 2007) and host cells (Cinco et al., 2002; Breiner et al., 2009). Among the molecules studied, leptospires have been reported to bind to the integrin CR3 (also known as Mac-1), a receptor for the complement protein 1C3b, which had been shown to have the function of attachment and phagocytosis of bacteria such as B. burgdorferi (Cinco et al., 2002). In addition, Coburn et al. (1998) noted that virulent and avirulent strains of B. burgdorferi attach to aVb3 and a5b1. These ubiquitous integrins, although they may only be involved partially in attachment, may also possibly have a role in the colonization of target organs. For the first time, we report that serovars Jules and Portlandvere also attach to aVb3 (vitronectin receptor) and a5b1 (fibronectin receptor). The strains analysed, even though laboratory-passaged, showed strong affinity for both integrins, with serovar Jules showª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

112

ing the stronger affinity, particularly for aVb3 (Fig. 3b). This suggests that serovar Jules might be better able to utilize the aVb3 integrin in attachment and may indicate a greater potential for virulence. Interestingly, these results mirror those of Coburn et al. (1998) who showed that B. burgdorferi and Borrelia garinii attach better to aVb3 than to a5b1. However, with only two serovars analysed, albeit from different species, it would not be wise to suggest that there is a relationship (genetic or otherwise) between an organism’s attachment to one particular integrin and its attachment to another. From these data, it was noted that the serovars responded similarly to both fibronectin and the integrin a5b1 (fibronectin receptor), although more serovars should be used to confirm any trend in binding. Fibronectin has a role as a sort of bridging molecule between the cell and the spirochaete (Ito & Yanagawa, 1987) and it is unlikely that the fibronectin’s integrin/ cell-binding domain that was involved in adhesion (Merien et al., 2000). Leptospira borgpetersenii (which includes serovar Jules) has a genome 16% smaller with more pseudogenes than L. interrogans (which includes serovar Portlandvere) (Bulach et al., 2006) and it might be suggested that Jules would respond with lower levels of attachment than serovar Portlandvere. Evidently, this was not the case. Since the sequencing of L. interrogans serovar Lai (Ren et al., 2003) and Copenhageni (Nascimento et al., 2004), two genome studies of Leptospira have reported that L. borgpetersenii’s need for a host for survival is owing to its reduced genome, particularly because of lost or missing genes that enable sensory transduction and defence mechanisms (Bulach et al., 2006; Picardeau et al., 2008). It is likely that Jules has better conserved its remaining genes and thus has better developed mechanisms for attaching to host cells and this may explain why its level of binding to host cell molecules is higher. This study also examined the effect of lectins and sugars on the level of attachment of the spirochaete to the sugars on the cell surface of the HEp-2 cells. We noted that for the most part, there was an increase in the attachment of serovar Jules to the cells coated with lectins, and ConA in particular. In contrast, serovar Portlandvere showed a decrease in binding except with ConA, where a significant increase in attachment was also seen (Fig. 4a). It appeared that ConA was binding to both the HEp-2 cells and the leptospires and acting as a kind of bridging molecule. However, when the serovars were preincubated with sugars, there was reduced binding to cells (Fig. 5a). The use of L-mannose caused a decrease in attachment, while other authors reported an increase with D-mannose (Cinco et al., 2002) where it has been hypothesized that ‘mannose’ (the author used D-mannose) was ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

G.I. Andrade & P.D. Brown

recognized by the CR3 receptor of polymorphonuclear neutrophils and Chinese hamster ovary cells and to another unknown molecule on the surface of Leptospira. It has been shown that there seems to be an increase in the severity of illness when infected individuals exhibit high concentrations of a mannose-binding lectin (MBL) (Miranda et al., 2009). These results suggest that Leptospira also may use a MBL to attach to the host. This has also been seen in cases of lepromatous forms of leprosy (Dornelles et al., 2006). It may then seem probable that ConA, an MBL, may also potentiate virulence through an increased level of attachment. Pattern recognition receptors (PRRs) are molecules capable of identifying very particular pathogen-associated molecular patterns. The DC-SIGN molecule on the surface of dendritic cells belongs to the C-type lectin group of PRRs (Geijtenbeek et al., 2000). Molecules containing L-fucose and those rich in mannose have been shown to bind to the DC-SIGN receptor on dendritic cell surfaces. It has been suggested that L. interrogans uses their mannose-rich surfaces to attach to the molecule (Gaudart et al., 2008). The findings of this study suggest that both mannose and fucose possibly have an inhibitory effect on binding and thus may support the role of a DC-SIGNlike molecule in attachment of the bacteria to HEp-2 cells and also, it warrants further study of the sugars as possible forms of treatment for leptospiral infection in the same way D-mannose has been shown to be effective against urinary tract infections caused by Escherichia coli (Howell, 2008). In this study, serovar Jules showed no affinity for GAGs cleaved by chondroitinase ABC (Fig. 3b), and in fact registered a significant increase in attachment after the moieties were removed by the enzyme (P < 0.01). These moieties would have included chondroitin sulphates A, B and C, chondroitin and to a lesser extent hyaluronic acid. To date, only one report (Breiner et al., 2009) has investigated in particular the role of GAGs in the attachment of Leptospira. The authors noted that attachment to the molecules seemed to be largely dependent on their sulfation and polymer size, which possibly provided more binding sites for the bacteria and that chondroitin sulphate B and C in particular were greatly implicated in the attachment of serovars Copenhageni and Canicola (Breiner et al., 2009). This difference may be due to the fact that the group used the serovar L. interrogans serovar Copenhageni Fiocruz L1-130 and incubated their cell lines with enzymes for one more hour than performed in this study. The increase in the attachment seen in the enzyme assays (with the exception of trypsin) may be due to the cleaving of moieties on the cells surface by the enzymes which have exposed receptors to which the bacteria have FEMS Immunol Med Microbiol 65 (2012) 105–115

Attachment of Leptospira to mammalian cells

greater affinity. The decrease in attachment attributed by trypsin was also reported by Thomas & Higbie (1990), who used L. interrogans serovars Canicola, Pomona and Grippotyphosa. Trypsin, a digestive protein, cleaves peptides on the C-terminal side of lysine and arginine amino acid residues. It is likely that there are several receptors on the cell surface, which are modified by this enzyme, and that these vary quantitatively and qualitatively from cell line to cell line. This could explain the greater fall shown in that earlier study, which used MDCK, PK-15 (kidney cells) and human umbilical vein endothelium cells. Although the group reported no testing with chondroitinases or heparinases, they did reveal a statistically insignificant decrease in attachment with neuraminidase, while this study showed a significant increase with the two serovars analysed. The two serovars themselves differed from each other in their responses to the different enzymes (Fig. 6b). Differences in genes for the production of virulence proteins exist (Bulach et al., 2006). Unique to L. borgpetersenii are two N-acetylneuraminic acid synthetases used to produce sialic acid which the spirochaete could use to coat the cell as an escape mechanism as the coating is used to conceal mannose. As L. borgpetersenii serovar Jules showed a significant increase in attachment when neuraminidase was used to cleave sialic acid from the cells, it seems plausible that the sialic acid on the surface of serovar Jules may be able to replace the cleaved sialic acid on the mammalian cell surface by binding to the glycoprotein or glycolipid chain left behind. Interestingly, these sialic acid-producing genes are absent in L. interrogans (Bulach et al., 2006) and very little change in the attachment of Portlandvere was seen when the sialic acid molecules were cleaved with neuraminidase (this study). Another possible hypothesis is that the cleaving of sialic acid was able to produce an increase in its affinity for a member of the ECM to which the serovar Jules has a greater affinity when compared to serovar Portlandvere. It has been reported that tumour cell lines have a greater affinity for fibrinogen and collagen type IV after treatment with neuraminidase (Dennis et al., 1982). Another possibility may be that the removal of the negative charge of the sialic acid would lead to better probabilities for attachment of the bacteria (whose cell surface is rich in lipopolysaccharide, which confers a negative charge) to the mammalian cell surface, or even changes in the conformation of surface molecules. The heparin molecule is distributed within the cell, not on the cell surface like heparan sulphate, and that would explain why treatment with heparinase showed no significant changes in the attachment of either serovar to the cells. As it pertains to heparinase, similar studies have demonstrated that L. interrogans serovar Pomona showed no significant attachment to heparin-coated plates when FEMS Immunol Med Microbiol 65 (2012) 105–115

113

compared to BSA (blocking agent) (Lin & Chang, 2008), supporting this study.

Conclusions Leptospira interrogans serovar Porlandvere and L. borgpetersenii serovar Jules showed similar attachment to HEp-2 cell line monolayers. Both serovars showed significant increases in attachment to wells coated with fibronectin, and serovar Jules had a significantly greater affinity which may imply that Jules has a greater potential for virulence as attachment to fibronectin has been linked to virulence. Although both serovars showed increased attachment to the integrins tested, only serovar Jules showed a significantly higher level of attachment to integrins aVb3 and a5b1 when compared to empty blocked wells and in the case of aVb3, Jules showed a greater level of binding. Serovars Jules and Portlandvere exhibited significantly increased binding to cells, which have been coated with ConA but Portlandvere showed a decrease in its level of attachment to cells coated with PNA, ECA, UEA and SBA. The difference between the two serovars was significant for ECA and SBA lectins as serovar Jules showed significantly higher level of binding to those particular molecules. Further investigations are required to examine the role, if any, that ConA may play in attachment. Both serovars, when preincubated with various sugars, showed a decrease in attachment to the cells. Further, serovar Jules showed a significant increase in attachment to cell surfaces, which had been treated with neuraminidase and chondroitinase ABC and seemed to require molecules cleaved by trypsin for attachment. Serovar Portlandvere, while also requiring these trypsin-cleaved moieties, did not seem to rely heavily on those cleaved by neuraminidase for its attachment. Taken together, these results indicate that serovar Jules tended to exhibit traits of a more virulent strain when compared to Portlandvere based on its higher level of attachment to monolayers, fibronectin (a known virulence factor), integrins aVb3 and a5b1 (similar to B. burgdorferi), and several lectins tested. When analysed with local epidemiological data, this may explain the resurgence of serovar Jules as a leading aetiological agent of leptospirosis in Jamaica. While many will argue that clinically there is no species relatedness, epidemiologically, it is critical to be able to understand trends or variations in the level of the aetiological agent in the environment and in clinical cases.

Acknowledgements The authors acknowledge the grants provided by the Office of Graduate Studies and Research of the University ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

114

of the West Indies, Mona Campus, and the National Health Fund, Jamaica.

Competing interests The authors declare that they have no competing interests.

Authors’ contribution G.I.A. assisted in the design of the study, carried out the attachment immunoassays, performed statistical analyses and drafted the manuscript. P.D.B. conceived of the study, its design and coordination and edited the manuscript. Both authors read and approved the final manuscript.

References Alugupalli KR, Michelson AD, Barnard MR, Robbins D, Coburn J, Baker EK, Ginsberg MH, Schwan TG & Leong JM (2001) Platelet activation by a relapsing fever spirochaete results in enhanced bacterium-platelet interaction via integrin aIIbb3 activation. Mol Microbiol 39: 330–340. Alvarez-Dominguez C, Vazquez-Boland J, Carrasco-Marin E, Lopez-Mato P & Leyva-Cobia F (1997) Host cell heparan sulfate proteoglycans mediate attachment and entry of Listeria monocytogenes, and the listerial surface protein ActA is involved in heparan sulfate receptor recognition. Infect Immun 65: 78–88. Barbosa AS, Abreu PAE, Neves FO, Atzingen MV, Watanabe MM, Viera ML, Morais ZM, Vasconcellos SA & Nacimiento ALTO (2006) A newly identified leptospiral adhesion mediates attachment to laminin. Infect Immun 74: 6356–6364. Bharti AR, Nally JE, Ricaldi JN et al. (2003) Leptospirosis: a zoonotic disease of global importance. Lancet Infect Dis 3: 757–771. Bras G (1955) Leptospirosis in Jamaica; case report. West Indian Med J 1955(4): 126–128. Breiner DD, Fahey M, Salvador R, Novakova J & Coburn J (2009) Leptospira interrogans binds to human cell surface receptors including proteoglycans. Infect Immun 77: 5528–5536. Brown PD, McKenzie M, Pinnock M & McGrowder D (2011) Environmental risk factors associated with leptospirosis among butchers and their associates in Jamaica. Int J Occup Environ Med 2: 47–57. Bulach DM, Zuerner RL, Wilson P et al. (2006) Genome reduction in Leptospira borgpetersenii reflects limited transmission potential. P Natl Acad Sci USA 103: 14560–14565. Chirathaworn C, Patarakul K, Saksit V & Poovorawan Y (2007) Binding of Leptospira to extracellular matrix proteins. J Med Assoc Thai 90: 2136–2142. Choy HA, Kelley MM, Chen TL, Moller AK, Matsunaga J & Haake DA (2007) Physiological osmotic induction of Leptospira interrogans adhesion: LigA and LigB bind

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved

G.I. Andrade & P.D. Brown

extracelluar matrix proteins and fibrinogen. Infect Immun 75: 2441–2450. Cinco M, Cini B, Perticarari S & Presani G (2002) Leptospira interrogans binds to the CR3 receptor on mammalian cells. Microb Pathog 33: 299–305. Coburn J, Magoun L, Bodary S & Leong J (1998) Integrins avb3 and a5b1 mediate attachment of lyme disease spirochetes to human cells. Infect Immun 66: 1946–1952. Defoe G & Coburn J (2001) Delineation of Borrelia burgdorferi p66 sequences required for integrin aIIbb3 recognition. Infect Immun 69: 3455–3459. Dennis J, Waller C, Timpl R & Schirrmacher V (1982) Surface sialic acid reduces attachment of metastatic tumor cells to collagen type IV and fibronectin. Nature 300: 274–276. Diamond MS & Springer TA (1993) A subpopulation of Mac-1 (CDllb/CD18) molecules mediates neutrophil adhesion to ICAM-1 and fibrinogen. J Cell Biol 120: 545–556. Dornelles LN, Pereira-Ferrari L & Messias-Reason I (2006) Mannan-binding lectin plasma levels in leprosy: deficiency confers protection against the lepromatous but not the tuberculoid forms. Clin Exp Immunol 145: 463–468. Gaudart N, Ekpo P, Pattanapanyasat K, van Kooyk Y & Engering A (2008) Leptospira interrogans is recognized through DC-SIGN and induces maturation and cytokine production by human dendritic cells. FEMS Immunol Med Microbiol 53: 359–367. Geijtenbeek TBH, Torensma R, van Vliet SJ, van Duijnhoven GCF, Adema GJ, van Kooyk Y & Figdor CG (2000) Identification of DC-SIGN, a novel dendritic cell–specific ICAM-3 receptor that supports primary immune responses. Cell 100: 575–585. Guo B, Norris S, Rosenberg L & Hook M (1995) Adherence of Borrellia burgdorferi to the proteoglycan decorin. Infect Immun 63: 3467–3472. Howell A (2008) D-mannose consumption and bacterial antiadhesion activity in human urine. J Fed Amer Soc Exper Biol 22: 702–732. Ito T & Yanagawa R (1987) Leptospiral attachment to four structural components of extracellular matrix. Jap J Vet Sci 49: 875–882. Levett P (2001) Leptospirosis. Clin Microbiol Rev 14: 296–326. Lin YP & Chang YF (2008) The C-terminal variable domain of LigB from Leptospira mediates binding to fibronectin. J Vet Sci 9: 133–144. Matsunaga J, Barocchi MA, Croda J et al. (2003) Pathogenic Leptospira species express surface-exposed proteins belonging to the bacterial immunoglobulin superfamily. Mol Microbiol 49: 929–945. Merien F, Truccolo J, Baranton G & Perolat P (2000) Identification of a 36-kDa fibronectin-binding protein expressed by a virulent variant of Leptospira interrogans serovar icterohaemorrhagiae. FEMS Microbiol Lett 185: 17–22. Miranda KA, Vasconcelos LRS, Coelho LCBB, Lima Filho JL, Cavalcanti MSM & Moura P (2009) High levels of serum mannose-binding lectin are associated with the severity of clinical signs of leptospirosis. Braz J Med Biol Res 42: 353–357.

FEMS Immunol Med Microbiol 65 (2012) 105–115

Attachment of Leptospira to mammalian cells

Nascimento ALTO, Verjovski-Almeida S, Van Sluys MA et al. (2004) Genome features of Leptospira interrogans serovar Copenhageni. Braz J Med Biol Res 37: 459–477. Pan American Health Organization (1979) Disaster Reports Number 2: Jamaica, St. Vincent, and Dominica. Retrieved from New Zealand Digital Library [http://nzdl.sadl.uleth.ca/ cgi-bin/library?e=q-00000-00—off-0paho–00-0–0-10-0—0— 0prompt-10—4—-st0–0-1l–11-en-50—20-about-Leptospirosis+epidemics+Jamaica–00-0-1-00-0-0-11-1-0utfZz-8-00&a= d&c=paho&srp=0&srn=0&cl=search& d=HASH012be74bf12f4a34523c2510.4.4.1]. Picardeau M, Bulach DM, Bouchier C et al. (2008) Genome sequence of the saprophyte Leptospira biflexa provides insights into the evolution of Leptospira and the pathogenesis of leptospirosis. PLoS One 3: 1607. Plotkowski MC, Costa AO, Morandi V, Barbosa HS, Nader HB, DeBentzmann S & Puchelle E (2001) Role of heparan

FEMS Immunol Med Microbiol 65 (2012) 105–115

115

sulphate proteoglycans as potential receptors for non pilliated Peudomonas aeruginosa adherence to non-polarized airway epithelial cells. J Med Microbiol 50: 183–190. Ren SX, Fu G, Jiang XG et al. (2003) Unique physiological and pathogenic features of Leptospira interrogans revealed by whole-genome sequencing. Nature 422: 888–893. Thomas DD & Higbie LM (1990) In vitro association of leptospire with host cells. Infect Immun 58: 581–585. Thornton BP, Vetvika V, Pitman M, Goldman RC & Ross GD (1996) Analysis of the sugar specificity and molecular location of the b-glucan-binding lectin site of complement receptor type 3 (CD11b/CD18). J Immunol 156: 1235– 1246. World Health Organization (2003) Human leptospirosis: Guidance for diagnosis, surveillance and control. (NLM classification: WC 420) Retrieved from http://whqlibdoc. who.int/hq/2003/WHO_CDS_CSR_EPH_2002.23.pdf.

ª 2012 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved