Anaplasma phagocytophilum dense-cored organisms mediate cellular adherence through recognition of human P-selectin glycoprotein ligand 1

Abstract

Anaplasma phagocytophilum is an obligate intracellular bacterium that infects granulocytes to cause human granulocytic anaplasmosis. The susceptibilities of human neutrophils and promyelocytic HL-60 cells to A. phagocytophilum are linked to bacterial usage of P-selectin glycoprotein ligand 1 (PSGL-1) as a receptor for adhesion and entry. A. phagocytophilum undergoes a biphasic developmental cycle, transitioning between a smaller electron dense-cored cell (DC), which has a dense nucleoid, and a larger, pleomorphic electron lucent reticulate cell (RC), which has a dispersed nucleoid. The pathobiological roles of each form have not been elucidated. To ascertain the role of each form, we used electron microscopy to monitor bacterial binding, entry, and intracellular development within HL-60 cells. Only DCs were observed binding to and inducing uptake by HL-60 cells. By 12 h, internalized DCs had transitioned to RCs, which had initiated replication. By 24 h, large RC numbers were observed within individual inclusions. Reinfection had occurred by 36 h, as individual, vacuole-enclosed DCs and RCs were again observed. The abilities of DC- and RC-enriched A. phagocytophilum populations to bind and/or infect HL-60 cells or Chinese hamster ovary cells transfected to express PSGL-1 (PSGL-1 CHO) were compared. Only DCs bound PSGL-1 CHO cells and did so in a PSGL-1-blocking antibody-inhibitable manner. These results demonstrate that the respective roles of A. phagocytophilum DCs and RCs are consistent with analogous forms of other obligate intracellular pathogens that undergo biphasic development and hint that the PSGL-1-targeting adhesin(s) may be upregulated or optimally posttranslationally modified on DCs.

FIG. 1.

A. phagocytophilum DCs and RCs. A. phagocytophilum organisms were cultivated in HL-60 cells and examined by transmission electron microscopy. Representative images of DCs (A to H) and RCs (I to P). (A to H) DCs were 0.83 ± 0.24 μm wide, had a dense nucleoid and a ruffled outer membrane, and were spheroid. The arrow in panel A denotes an individual DC surrounded by a membranous projection. (I to P) RCs were 0.69 ± 0.36 μm by 1.04 ± 0.44 μm, had a dispersed nucleoid and an outer membrane that was somewhat smoother than that observed for DCs, and were pleomorphic. Panels F and N are magnifications of the boxed areas in panels E and M, respectively. Bars, 0.5 μm.

FIG. 2.

A. phagocytophilum adhesion to, invasion of, and replication within HL-60 cells. Host cell-free A. phagocytophilum organisms were incubated with HL-60 cells. After 40 min, unbound bacteria were removed and host cells were examined by transmission electron microscopy. (A to C) At 40 min, DC organisms were observed binding to and triggering their own uptake by HL-60 cells. (D to F) At 4 h, some DC organisms remained bound at the HL-60 cell surface, while others had been internalized into vacuoles that were in close proximity to the host cell surface. No RC organisms were observed interacting with HL-60 cell surfaces. Bars, 0.5 μm. Representative results of one of two separate experiments are shown.

FIG. 3.

Intracellular development of A. phagocytophilum in HL-60 cells. Host cell-free A. phagocytophilum organisms were incubated with HL-60 cells. After 40 min, unbound bacteria were removed and host cells were examined at 12 (A), 24 (B), 36 (C and D), 48 (E), and 72 h (F) postinfection by transmission electron microscopy. Arrowheads in panels B and E indicate morulae harboring A. phagocytophilum organisms that are transitioning from RCs to DCs. Arrows in panels D and F denote inclusions that contain DCs. Bars, 1.0 μm. Representative results of one of two separate experiments are shown.

FIG. 4.

Number of A. phagocytophilum (Ap) organisms observed per morula and the relative abundances of DCs and RCs observed over the course of infection. Values presented were determined by enumerating the numbers of A. phagocytophilum organisms within individual morulae observed by transmission electron microscopy (A) and the total numbers of RCs and DCs in Fig. 2 and 3 (B). (A) Numbers of organisms observed per morula between 12 and 72 h postinfection. Morulae containing 1 to 2, 3 to 5, 6 to 10, 11 to 15, and ≥16 organisms were enumerated and are presented as percentages of the total number of morulae. (B) Total numbers of DCs and RCs observed binding to or internalized within HL-60 cells. Representative results of one of two separate experiments are shown.

FIG. 5.

Binding and infection of RC- and DC-enriched A. phagocytophilum (Ap) organisms in HL-60 cells. Comparable amounts of host cell-free A. phagocytophilum organisms were isolated at 24 h (RC-enriched) or 72 h (DC-enriched) following a synchronous infection and incubated with naïve HL-60 cells for 40 min. (A) After removal of unbound bacteria, aliquots were examined by indirect immunofluorescence microscopy and the mean numbers of bound A. phagocytophilum organisms per cell were enumerated. (B) The remainders of each infected HL-60 culture were grown for 96 h, during which aliquots were removed at 24-h intervals and assessed for infection, as determined by light microscopic detection of morulae. Results shown are the means of four separate experiments. Statistically significant values are indicated (*, P < 0.05).

FIG. 6.

Analysis of PSGL-1 CHO and CHO (−) cells for PSGL-1 and sLe^x expression. PSGL-1 CHO and CHO (−) cells were incubated with anti-PSGL-1 MAb KPL-1 (A), anti-sLe^x MAb CSLEX1 (B), or isotype-matched controls. Alexa Fluor 488-conjugated secondary antibodies detected bound primary antibodies. Representative results of one of two separate experiments are shown.

FIG. 7.

Binding of RC- and DC-enriched A. phagocytophilum (Ap) organisms to PSGL-1 CHO and CHO (−) cells. Comparable amounts of host cell-free A. phagocytophilum organisms isolated at 24 h (RC-enriched) and 72 h (DC-enriched) post-synchronous infection were incubated with PSGL-1 CHO cells in the presence of anti-PSGL-1 MAb KPL-1 or isotype control or with CHO (−) cells. After removal of unbound bacteria, A. phagocytophilum binding was assessed by indirect immunofluorescence microscopy. Representative results of one of three separate experiments are shown. Statistically significant values are indicated (**, P < 0.01; ***, P < 0.001).

FIG. 8.

A. phagocytophilum (Ap) DCs, but not RCs, bind to PSGL-1 CHO cells. Comparable amounts of host cell-free DC-enriched or RC-enriched populations were incubated with PSGL-1 CHO cells. After removal of unbound bacteria, the PSGL-1 CHO cells were examined by transmission electron microscopy. (A) Percentage of bound A. phagocytophilum organisms that are in either the DC stage or other (membrane ghosts or transition) stages that were observed bound to the surfaces of PSGL-1 CHO cells. One hundred PSGL-1 CHO cell-associated A. phagocytophilum organisms were counted per population per experiment. Results shown are the means of two separate experiments. Statistically significant values are indicated (**, P < 0.01; ***, P < 0.001). (B to E) Representative transmission electron micrographs showing a DC (B and D), a membrane ghost (C), or a putative transition stage organism (E) interacting with the PSGL-1 CHO cell surface.