QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wallach-Dayan, S. B. Articles by Naor, D. Search for Related Content PubMed PubMed Citation Articles by Wallach-Dayan, S. B. Articles by Naor, D.

CD44-dependent lymphoma cell dissemination: a cell surface CD44 variant, rather than standard CD44, supports in vitro lymphoma cell rolling on hyaluronic acid substrate and its in vivo accumulation in the peripheral lymph nodes

Shulamit B. Wallach-Dayan¹, Valentin Grabovsky², Jürgen Moll^3,*, Jonathan Sleeman³, Peter Herrlich³, Ronen Alon² and David Naor^1,

¹ The Lautenberg Center for General and Tumor Immunology, The Hebrew University-Hadassah Medical School, Jerusalem, 91120 Israel
² Department of Immunology, The Weizmann Institute, Rehovot, 76100 Israel
³ Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics, Karlsruhe D-76021, Germany
^* Present address: Pharmacia & Upjohn, via Pasteur 10, 1-20014 Nerviano (Milan), Italy

View larger version (17K): [in a new window]   Fig. 1. Generation of LB-TRv transfectants. (A) Schematic map of the CD44 gene. Arrowheads mark the positions of 5'-ORF and 3'-ORF primers (sequence as shown in B) used to isolate the CD44v4-v10 cDNA from mouse keratinocytes. Arrows mark the positions and directions (sense ; anti-sense ) of the primers (for sequence, see Rochman et al. (Rochman et al., 2000) used in the exon-specific RT-PCR analysis of LB cells transfected with isolated CD44v4-v10 cDNA (LB-TRv cells). Hatched bars denote 5' and 3' constant exons (excluding introns), open bars represent the variant exons, the lines connecting the exons depict intron sequences. Pc5 and pv1 to pv10 primers were used for exon-specific PCRs in combination with the 3' primer pc16, for analysis of mouse cDNA. (B) ORF-5' and ORF-3' primers.

View larger version (34K): [in a new window]   Fig. 2. LB cells transfected with CD44v4-v10 cDNA express v4 and v6-encoded epitopes and bind soluble HA. (A) Parental LB cells (LB), LB cells transfected with CD44s cDNA (TRs) as well as LB cells transfected with CD44v4-v10 cDNA (TRv) and two representative clones (TRv2 and TRv3) of six derived from two separate transfections, were analyzed by flow cytometry for CD44 isoform expression, as described in Materials and Methods. The cells were stained with KM81 anti-pan CD44 mAb (marked CD44), 10D1 anti-CD44v4 mAb (marked V4) and 9A4 anti-CD44v6 mAb (marked V6) followed by anti-rat IgG conjugated to FITC. FITC indicates cells stained with the second antibody alone. Note that the pan CD44 fluorescence intensity of the TRs cells is higher than that of the other cell lines. (B) LB, TRs and TRv cells, as well as representative TRv cloned cells, were incubated with fluorescein-labeled HA (20 µg/ml) in the absence (Fl-HA) or presence ((Fl-HA+HA) of unlabeled HA (1 mg/ml), which competitively inhibits the binding of Fl-HA, and analyzed by flow cytometry. Segregated histograms display HA binding; matched histograms indicate the inability to bind HA. (C) All cells were incubated with fluorescein-labeled HA (20 µg/ml) in the absence (Fl-HA) or presence (Fl-HA+KM81) of KM81 (IgG2b) anti-CD44 mAb (100 µg/ml) and analyzed by flow cytometry. KM81 anti-CD44 mAb blocks the binding of Fl-HA to TRv cells. The histograms of the LB-TRo cells were almost identical to those of the LB cells (not shown).

View larger version (31K): [in a new window]   Fig. 3. Expression of a CD44 variant extends the ligand binding capacity of the lymphoma cells. (A) Western blots of CPC-precipitated CD44 derived from a total cell population of LB cells transfected with CD44v4-v10 cDNA (LB-TRv) or standard CD44 (LB-TRs). LB-TRv and LB-TRs cells were lysed and mixed with KS, HA, CS, H (Heparin) and HS. After CPC precipitation of GAG-associated proteins and their resolution on 7.5% SDS-PAGE (under nonreducing conditions), they were identified by western blot with KM81 anti-CD44 mAb. The last lane in each panel (marked by C) shows control precipitation (i.e. without the prior addition of GAGs). Lanes marked by ‘lysate’ show western blotting of the total cell lysate (no CPC precipitation) with anti-CD44 mAb. CD44v of LB-TRv cells binds HA as well as additional GAGs. (B) Flow cytometry analysis of CS binding to total LB-TRv and LB-TRs cell populations. LB-TRv and LB-TRs cells were stained with fluoresceinated CS (50 µg/ml) in the absence (FL-CS) or presence of KM81 anti-CD44 mAb (100 µg/ml) (FL-CS+KM81). Separate histograms display CS binding; matched histograms show the inability to bind CS. AF, autofluorescence.

View larger version (41K): [in a new window]   Fig. 4. Binding of LB cell line to HA under wall shear stress is mediated by CD44v. (A) Accumulation and rolling of LB-TRs and LB-TRv cells on different immobilized GAGs under shear flow. LB-TRs and LB-TRv lymphoma cells were allowed to settle for 100 seconds on different substrates, each coated with 1 mg/ml of a purified GAG, as indicated. The cells were subjected to a shear stress of 0.25 dyn/cm2 for 5 seconds, followed by shear stresses of 0.5, 0.75 and 2 dyn/cm2, each lasting for 5 seconds. The number of cells remaining attached at 2 dyn/cm2 was counted at the end of the last interval of shear stress. (B) Effect of anti-CD44 mAbs on accumulation and rolling of CD44-transfected LB cells on HA. LB-TRV cells (107 cells/ml) were pretreated with medium alone or with 100 µg/ml isotype-matched control mAb (4D2), blocking KM81 anti-CD44 mAb or nonblocking KM114 anti-CD44 mAb (5 minutes 4°C) in binding medium. The cells were diluted 1:10 with medium without washing out the antibodies and the suspension was then perfused into a flow chamber coated with 0.5 mg/ml HA. The cells were allowed to settle on the substrate and then subjected to increasing shear stress, as described in A. The number of cells remaining attached at 2 dyn/cm2 was counted at the end of the last interval of shear stress. (C) Inhibition of attachment events (followed by rolling) of LB-TRv cells on immobilized HA with the following (1 mg/ml) soluble GAGs: HA, CS (chondroitin sulfate) and KS (keratan sulfate). Cells were perfused for 1 minute at a shear stress of 0.75 dyn/cm2 over an HA (1 mg/ml) substrate. The number of cells that attached to and persistently rolled on the CD44 ligand was determined. (D) Effect of increasing shear stress on the rate of accumulation of CD44 transfectants on HA-coated substrate. LB-TRs or LB-TRv cells were perfused in multiple runs over a HA-coated field (1 mg/ml). Each perfusion run was performed at a different shear stress and lasted for 1 minute. The number of rolling cells accumulated at the end of each run is shown for each indicated shear stress. Note that at physiological shear stresses (>1 dyn/cm2), only LB-TRv cells were capable of persistent rolling on the HA field. (E) Accumulation and resistance to detachment of different lymphoma cell lines on HA. The indicated cells were accumulated for 1 minute at a low shear stress (0.75 dyn/cm2) and then subjected to a sharp increase in shear stress (1.75 dyn/cm2) for 20 seconds. The number of cells bound per field at the end of the accumulation period (marked accumulated) and after 20 seconds of high shear stress (marked resistant to detachment) are shown. The number of rolling cells was counted in one (C,D) or, if required, in two (A,B,E) representative fields. When two fields were counted, their mean ± range of determination are shown. All panels depict one of three independent experiments showing similar results.

View larger version (36K): [in a new window]   Fig. 5. LB cells transfected with CD44v4-v10 cDNA mutated at the HA binding site lose their ability to bind HA and CS. (A) LB cells transfected with CD44v4-v10 cDNA (LB-TRv) (top panels) and LB cells transfected with CD44v4-v10 cDNA mutated at the HA binding site (LB-TRvM) (bottom panels) were indirectly analyzed by flow cytometry for expression of CD44 isoforms (pan CD44 and v6; left panels), as described in the legend to Fig. 2. In addition, the cells (right panels) were incubated with fluorescein-labeled HA (20 µg/ml) in the absence (FL-HA) or presence of unlabeled HA (1 mg/ml) (FI-HA+HA) or KM81 anti-CD44 mAb (100 µg/ml) (FI-HA+KM81) and analyzed by flow cytometry. Segregated histograms display HA binding; matched histograms indicate the inability to bind HA. (B) LB-TRv and LB-TRvM cell lines were subjected to flow cytometry to test their ability to bind FL-CS, as described in the legend to Fig. 3. LB-TRvM cells did not bind CS. AF, autofluorescence.

View larger version (35K): [in a new window]   Fig. 6. LB cells transfected with CD44v4-v10 cDNA show more efficient local tumor formation and lymph node accumulation than LB cells transfected with standard CD44 cDNA or LB cells transfected with mutated CD44v4-v10 cDNA. Green fluorescent protein (GFP) cDNA was transfected into LB-TRv (TRv-G), LB-TRs (TRs-G) (both derived from independent CD44 transfections different from those of the corresponding nonfluoresceinated lymphoma cells) and parental LB (LB-G) cells. A quantity of 3x106 cells from each transfected cell population was s.c. injected into BALB/c mice. The size of the local tumor (± s.e.m.) at the injection site (A) was measured at different time points after tumor inoculation (5 mice from each group were killed at each time point). Invasion of a peripheral lymph node (B) by the green lymphoma cells at different time points after tumor inoculation is indicated by the normalized mean (± s.e.m.) fluorescence intensity (MFI) of the whole organ (5 animals from each group were tested for lymph node invasion at each time point). In another experiment, BALB/c mice were s.c. inoculated with 3x106 green LB cells transfected with mutated CD44v4-v10 cDNA (TRvM-G), green LB cells transfected with wild-type CD44v4-v10 cDNA (TRv-G) or green parental LB cells (LB-G). At each time point, 5 mice from each group were killed. Local tumor size (C) (± s.e.m.) and the intensity of lymph node invasion (D) (± s.e.m.) were assessed at different time points after tumor inoculation. *P<0.05; **P<0.01 by Mann-Whitney, compared with LB-G (A,B) or TRv-G (C,D). Note that the asterisks in C and D mark the two lines (LB-G, TRvM-G).

View larger version (38K): [in a new window]   Fig. 7. The effect of hyaluronidase and anti-CD44 mAb on local tumor growth and lymph node invasion by LB-TRv lymphoma cells. BALB/c mice were s.c. inoculated with 3x106 green LB-TRv cells (TRv-G). After 6 days, PBS (TRv-G), 20 units heparinase (TRv-G + HEP’ase) or hyaluronidase (TRv-G + HA’ase) of equivalent specific activity was injected i.p. into the mice (A,B). Using the same injection protocol, but in a different experiment, other groups of mice received PBS (TRv-G), isotype-matched control mAb (TRv-G+4D2) or IM7.8.1 anti-CD44 mAb (TRv-G+anti-CD44) (C,D). PBS, enzymes and antibodies, were injected every other day for 12 days (a total of 6 injections). Five mice from each group were killed at each time point). Local tumor size (A,C) and the intensity of lymph node invasion (B,D) were assessed at different time points, as described in Fig. 6. Statistical analysis as indicated in Fig. 6; all groups were compared with the TRv-G group. Note that the intensity of lymph node invasion following herparinase (B) or antibody injection (D) was assayed on day 19 only.

View larger version (42K): [in a new window]   Fig. 8. Accumulation and proliferation of lymphoma cells in the peripheral lymph nodes following i.v. injection. The indicated fluorescent lymphoma cells were injected i.v. into BALB/c mice. Three-color flow cytometric analysis of their lymph node cell suspensions was performed on days 4, 8 and 12 postinjection to show the accumulation of the different lymphoma cell lines as a function of time. The upper edge of the autofluorescence spot of normal lymph node cells (not shown) was used to determine the horizontal line bordering the lymph node cells (below the line) and the green tumor infiltrating cells (above the line). The percentage of GFP-positive infiltrating tumor cells (above the line; A in each box) in the total lymph node cell suspension (the sum of cells below and above the horizontal line (i.e. 7000 events) was determined by plotting green-fluorescence intensity (on a logarthmic scale) versus forward scatter. The green lymphoma cells (above the horizontal line) were gated and the proportion of BrdU-labeled dividing cells in the S-phase (B in each box) was determined following staining with anti-BrdU antibody and propidium iodide, as indicated in Materials and Methods. The percentage of cells in the S-phase (shown for day 12) was determined by dividing the number of cells framed in G3 by the total cell number (G1+G2+G3) x100. A similar formula was used to calculate the proportion (percentage) of cells in the S-phase at days 4 and 8. Note that the proportion of dividing cells in the population of green cells (i.e. lymphoma cells) is similar in all four cell lines, indicating a similar proliferation rate.