Induction of homotypic aggregation in the rainbow trout macrophage-like cell line, RTS11

Fish & Shellfish Immunology 22 (2007) 487e497 www.elsevier.com/locate/fsi

Induction of homotypic aggregation in the rainbow trout macrophage-like cell line, RTS11 S.J. DeWitte-Orr*, H.C.H. Hsu, N.C. Bols 1 Department of Biology, University of Waterloo, 200 University Avenue West, Waterloo, ON, Canada N2L 3G1 Received 26 April 2006; revised 15 June 2006; accepted 16 June 2006 Available online 21 June 2006

Abstract The rainbow trout monocyte/macrophage-like cell line, RTS11, has been used to study homotypic aggregation (HA), which is a well-studied feature of leucocytes in mammals but less understood in fish. HA is the aggregation of cells of the same cell type. RTS11 underwent HA in response to polyinosinic:cytidylic acid (poly IC), polyadenylic acid (poly A), lipopolysaccharide (LPS), zymosan, and phorbol 12-myristate 13-acetate (PMA). Poly IC was the best inducer of HA and did so in a dose- and time-dependent manner. The induction of RTS11 aggregation by poly IC required divalent cations but was not blocked by either an inhibitor of lymphocyte function-associated molecule-1 (LFA-1) or the tripeptide integrin adhesion recognition sequence, RGD. Poly ICinduced HA was inhibited by colchicine and latrunculin B, which act on microtubules and microfilaments, respectively, implying the necessity for an intact cytoskeleton. HA induction by poly IC did not occur at 4  C and was blocked by the transcriptional and translational inhibitors, actinomycin D and cycloheximide, respectively, suggesting the requirement for de novo protein synthesis. Poly IC-induced RTS11 aggregation was blocked by two inhibitors of dsRNA-dependent protein kinase (PKR). This is the first indication that PKR could have a role in the HA of leucocytes. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Homotypic aggregation; Monocyte/macrophage; RTS11; dsRNA; Integrin; Galectin; PKR

1. Introduction Cell to cell interactions are critical to immune responses and to the proliferation, migration and differentiation of leucocytes. The interactions can be between different cell types, heterotypic aggregation, or the same type, homotypic aggregation. Evidence suggests that homotypic aggregation (HA) is particularly important for leucocyte trafficking [1] and is a characteristic of activated immune cells [2]. HA has also been shown to affect cytokine expression and cellular proliferation in leukocytes [3]. Cellular adhesion can be most conveniently studied in vitro, and human myeloid cell lines, such as U937, have been especially useful for studying HA [1,4e9]. Several treatments have * Corresponding author. Tel.: þ1 519 888 4567; fax: þ1 519 746 0614. E-mail addresses: [email protected] (S.J. DeWitte-Orr), [email protected] (H.C.H. Hsu), [email protected] (N.C. Bols). 1 Tel.: þ1 519 888 4567x3993; fax: þ1 519 746 0614. 1050-4648/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2006.06.008

488

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

been found to induce the HA of U937, but perhaps the most studied inducer has been phorbol 12-myristate 13-acetate (PMA) [6,8,10]. Some forms of protein kinase C and possibly protein tyrosine kinases also mediate HA of U937 [1]. The surface molecules mainly responsible for HA appear to be the integrin leucocyte function-associated antigen-1 (LFA-1) interacting with its counter-receptor, the Ig superfamily member intercellular adhesion molecule-1 (ICAM-1) [5,6,11]. For fish leucocytes, relatively little is known about their aggregation or their surface molecules, such as integrins. Aggregation has been observed in channel catfish neutrophils in response to phorbol dibutyrate [12] and in gilthead seabream macrophages in response to a monoclonal antibody, anti-Aggregatin [13]. The distinctive expression of integrin subunits on mammalian leucocytes consists of a b2 (CD18) chain paired with one of four a chains to give rise to CD11a/CD18 (aLb2 or LFA-1), CD11b/CD18 (aMb2, Mac-1 or CR3), CD11c/CD18 (aXb2 or p150, 95), and CD11d/CD18 (aDb2) [14]. For fish, only limited information is available. CD18 or partial sequence for CD18 has been identified in channel catfish [15], carp [16], and in rainbow trout [17]. Channel catfish neutrophils and common carp granulocytes and monocytes/macrophages appear to contain CD11b/CD18 or CD11b/CD18-like receptors [12,16]. In this study, the rainbow trout monocyte/macrophage cell line, RTS11, has been used to explore HA in fish. RTS11 cells express genes for IL-1b and TGF-b [18], IL-8 [19,20], IL-11 [21], IL-18 [22], TNFa [23] as well as chemokines [24,25] and CD18 [26]. RTS11 appear to be especially sensitive to the actions of the synthetic double stranded (ds) RNA, polyinosinic:cytidylic acid (poly IC) [27]. For the first time in any leucocytes or immune cell lines, poly IC is shown to strongly induce HA, and the poly IC-induced HA of RTS11 appears to be mediated by dsRNA-dependent protein kinase (PKR). 2. Materials and methods 2.1. Cell culture RTS11, a rainbow trout monocyte/macrophage cell line developed in our laboratory [28], was grown in L-15 medium (Sigma, St. Louis, MO) supplemented with 15% foetal bovine serum (FBS), 150 U mL1 penicillin G, and 150 mg mL1 streptomycin sulphate (Sigma). RTS11 cells were grown at 18  C in 25 cm2 tissue culture treated flasks (Nunc, Roskilde, Denmark) and passaged every three to four weeks, the culture and conditioned medium was split into two flasks, and an equal volume of fresh medium was added to each flask. 2.2. Cell treatments RTS11 were seeded into 96-well plates (Falcon/BectoneDickinson, Franklin Lakes, NJ) in full growth medium at 1.5  105 cells per well and allowed to attach overnight at 18  C. Cells were then treated with the stimuli. Unless specified, cell aggregates and cell viability were generally measured after 24 h of incubation at 18  C. 2.2.1. Chemical stimuli Polyinosinic:cytidylic acid (poly IC; 0.05e500 mg mL1), polyadenylic acid (poly A; 0.05e500 mg mL1), lipopolysaccharide (LPS; 0.1e100 mg mL1), zymosan (0.005e50 mg mL1), and phorbol 12-myristate 13-acetate (PMA; 0.2 ng to 2 mg mL1) were added to the cells in increasing concentrations, with six replicates per concentration. All stimuli were purchased from Sigma with the exception of zymosan, which as purchased from InvivoGen (San Diego, CA). Cells were also treated with the carrier alone, in cases when the solvent was not L-15. In no cases was the solvent used at a concentration that was cytotoxic. Generally, cells were treated with the compounds for 24 h before cell viability and HA were measured. 2.2.2. Inhibitors All chemical inhibitors were added just prior to the addition of the stimulant. The inhibitors used in this study include cycloheximide (CHX; 0.1 and 1 mg mL1), actinomycin D (AMD; 0.05 and 0.5 mg mL1), colchicine (Col; 5e500 mg mL1), and latrunculin B (Lat; 0.05 and 0.5 mg mL1). Two PKR inhibitors were used in this study, 2-aminopurine (2-AP; 1.5 mM) and a commercially available oxindole functionalized with an imidazole, referred to as compound 16 [29] (0.5 mg mL1) with its negative control (0.5 mg mL1). Two integrin inhibitors were also used

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

489

as follows: RGD (5 mg mL1) and o-bromobenzoyl L-tryptophan, an LFA-1 inhibitor (2.5e25 mg mL1). All of the inhibitors were purchased from Sigma, with the exception of compound 16, its negative control, RGD and LFA-1 inhibitors, which were purchased from Calbiochem (La Jolla, CA). As mentioned above, control cells were also treated with the carrier alone, in cases when the solvent was not L-15. In no cases was the solvent used at a concentration that was cytotoxic. When low temperature was used as an inhibitor, the cultures were dosed with poly IC and immediately placed at 4  C. The cultures were kept at this temperature for 24 h. 2.2.3. Ca2þ/Mg2þ-free conditions When measuring divalent cation dependence, the cells were washed twice in Ca2þ/Mg2þ-free Hank’s buffered saline solution (HBSS; Sigma) before being plated at 1.5  105 cells per well in Ca2þ/Mg2þ-free HBSS. No FBS was added to these cultures to ensure Ca2þ/Mg2þ-free conditions. RTS11 was also seeded using HBSS with Ca2þ and Mg2þ as a control. Divalent cation dependence was also evaluated by treating RTS11 with poly IC in combination with 5 mM EDTA in serum free L-15 growth medium. No cell death was detected under any circumstance as measured using the two fluorescent indicator dyes, Alamar Blue and CFDA-AM, as described in Section 2.4 (data not shown). 2.3. Quantifying homotypic aggregation (HA) RTS11 cells in 96-well plates were examined for homotypic cell aggregation under a phase-contrast microscope (Nikon, Japan). Representative pictures of each well were taken at approximately the same location within each well at a magnification of 200 therefore six pictures were taken per treatment. Using Adobe Photoshop, aggregates larger than 0.04 mm by 0.04 mm were counted. In general, this particular size marker would include between 15 and 25 cells per aggregate. Any aggregate that did not fill the square was not included in the count. At least three separate experiments were performed for each treatment. 2.4. Measuring cytotoxicity Two fluorescent indicator dyes, Alamar BlueÔ (Medicorp, Montreal, PQ) and 5-carboxyfluorescein diacetate acetoxymethyl ester (CFDA-AM) (Molecular Probes, Eugene, OR), were used to evaluate cell viability. Viable cells cause the reduction of Alamar BlueÔ dye resulting in a chemical change from a non-fluorescent blue form (resazurin) to a fluorescent red form (resorufin) [30]. CFDA-AM, which is also taken up by live cells, is hydrolyzed by non-specific intracellular esterases to yield the fluorescent form, carboxyfluorescein. These esterases are only active within the confines of a cell; therefore, CFDA-AM can be used to monitor cell membrane integrity [31]. The levels of fluorescence were measured using a microplate spectrofluorometer. Alamar Blue and CFDAAM assays were performed as previously described [32]. Briefly, the cells were treated as described in Section 2.2. After incubation for the described amount of time, the media were removed and the two fluorescent dyes were added to the wells, diluted in a minimal salt solution called L-15ex. The cells were incubated with the dyes for 60 min before the microwells were read with a fluorometric plate reader (Molecular Devices SpectraMax). Excitation and emission wavelengths were 530 and 595 nm for Alamar BlueÔ and 485 and 530 nm for CFDAAM, respectively. 2.5. Data analysis For Alamar Blue and CFDA-AM assays, cytotoxicity was indicated by a decline in fluorescence units (FUs) for experimental cultures relative to control cultures. FUs for culture wells with no cells were constant and subtracted from FUs for experimental and control cultures. For graphical presentation, the results were plotted using SigmaPlot (Jandel Scientific). When comparing the effect of the inhibitors on aggregate formation, the average number of aggregates per treatment was compared to the control cultures with poly IC alone using an unpaired t-test. All statistical analyses were done using GraphPad InStat (version 3.00 for Windows 95, GraphPad Software, San Diego, CA, USA, www.graphpad.com).

490

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

3. Results 3.1. Induction of morphological changes in RTS11 cultures Several stimuli caused noticeable aggregation of RTS11 cultures 24 h after the addition of the stimuli (Fig. 1). Aggregates were seen with poly IC (5 mg mL1), poly A (5 mg mL1), phorbol 12-myristate 13-acetate (PMA, 2 mg mL1), lipopolysaccharide (LPS, 100 mg mL1) and zymosan (50 mg mL1). These concentrations had no detrimental effect on cell viability as measured with Alamar Blue for energy metabolism and CFDA-AM for cell membrane integrity (Fig. 2). The overall morphology of the aggregates differed slightly between the stimuli. With poly IC, LPS, and zymosan, the aggregates were stellate whereas with poly A and PMA they were more globular (Fig. 1). With treatment times of longer than 24 h, aggregate formation also differed between the stimuli. As determined by observation, poly IC-induced aggregates became larger with longer exposure times but the number of aggregates did not appear to increase, whereas with the other stimuli, longer incubations did appear to lead to more aggregates (data not shown). As more factors and mechanisms, such as cell proliferation, might contribute to the aggregation over longer exposures, further analysis was confined to 24 h treatments or less and the phenomenon was termed as homotypic aggregation (HA). 3.2. Quantification of homotypic aggregation (HA) When the number of aggregates was quantified, the stimuli differed in their capacity in inducing HA. For 24 h exposures, the best inducer of HA was poly IC. At the highest concentrations, poly IC induced approximately three to four times more aggregates than poly A, PMA or LPS and twice as many as zymosan (Fig. 3). The strong response with poly IC led to this stimulus being investigated further. Poly IC induced HA in a time- and concentration-dependent manner (Fig. 4). HA was induced at least as early as 8 h after exposure (Fig. 4). With 8 and 12 h exposures, increasing poly IC concentrations between 5 and 500 mg mL1 brought about an increasing number of aggregates. With 24 h exposures, the number of aggregates was increased by a concentration as low as 0.5 mg mL1 and dramatically increased by 5 mg mL1. Further increases in poly IC concentrations brought about only a slight increase in the number of aggregates after 24 h. All further experiments were performed to investigate the basic requirements for the induction of HA by poly IC.

Fig. 1. Induction of homotypic aggregation (HA) in RTS11 cultures. After 24 h, aggregates were observed in RTS11 cultures treated with 5 mg mL1 poly IC, 5 mg mL 1 poly A, 100 mg mL1 LPS, 2 mg mL1 PMA, or 50 mg mL1 zymosan for 24 h in full growth media. A box in each picture highlights a sample aggregate. Aggregate morphology varied depending on the stimulus, for example, poly IC induced ‘stellate’-like aggregates (indicated by a single arrow), while PMA induced ‘globular’-like aggregate morphology (double arrow). After 24 h, poly IC treated cultures showed the most dramatic difference in culture morphology; the number of aggregates in each treatment is quantified in Fig. 3. Pictures were taken at 200 magnification.

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

491

200 Alamar Blue CFDA

% control

150

100

50

0 poly IC

poly A

PMA

LPS

zymosan

Fig. 2. Effect of treatments causing HA on RTS11 cell viability. RTS11 cultures were treated with 5 mg mL1 poly IC, 5 mg mL 1 poly A, 2 mg mL1 PMA, 100 mg mL1 LPS, or 50 mg mL1 zymosan for 24 h after which cell viability was measured using Alamar Blue and CFDAAM. All values are expressed as percentage of carrier treated controls, and are plotted on y-axis.

3.3. Dependence of HA on macromolecular synthesis Inhibiting macromolecular synthesis inhibited the induction of HA by poly IC. At low temperatures, macromolecular synthesis in cultures of rainbow trout cell lines was severely reduced and at 4  C the induction of HA by poly IC was strongly inhibited (Fig. 5). The transcriptional inhibitor, actinomycin D, and translational inhibitor, cycloheximide, were used at concentrations that were not cytotoxic to RTS11 but inhibited the incorporation by RTS11 of 3 H-uridine by over 90% in the case of actinomycin D and of 3H-leucine by over 80% in the case of cycloheximide [27]. Actinomycin D at 0.5 mg mL1 inhibited HA induction by approximately 85% and cycloheximide at 0.1 and 1 mg mL1 completely blocked HA induction by poly IC (Fig. 5).

3.4. Dependence of HA on cytoskeleton integrity Cytoskeleton integrity appeared essential for HA. Colchicine, which disrupts microtubules, and latrunculin B, which impairs actin polymerization, were used at concentrations that were not cytotoxic to RTS11 as determined by Alamar Blue and CFDA-AM. Colchicine at 500 mg mL1 and latrunculin B at 0.05 and 0.5 mg mL1 completely inhibited HA induced by poly IC (Fig. 5). 3.5. Dependence of HA on Ca2þ/Mg2þ HA appeared to require Ca2þ/Mg2þ. Under normal growing conditions for assaying HA, divalent cations would be supplied by the L-15 and FBS. The addition of 5 mM EDTA reduced the induction of HA by poly IC in L-15 alone (data not shown). However, HA was also assayed in Hank’s buffered saline solution (HBSS) in order to more easily manipulate divalent cation levels. Poly IC was able to induce HA in HBSS; however, in HBSS without Ca2þ/Mg2þ no induction was observed (Fig. 6).

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

492 40

40

A

30

20

10

30

20

10

0

0 0

0.05

5

0.5

50

500

0

nucleotide concentration (µg mL-1)

2e-4

2e-3

0.02

0.2

2

PMA concentration (µg mL-1) 40

40

Average # of aggregates

30

20

10

D

Zymosan

C

LPS

Average # of aggregates

B

PMA

Average # of aggregates

Average # of aggregates

poly IC poly A

30

20

10

0

0 0

0.1

1

10

100

LPS concentration (µg mL-1)

0

5e-3

0.05

0.5

5

50

Zymosan concentration (µg mL-1)

Fig. 3. Quantification of HA induction by different inducers. RTS11 cultures were exposed for 24 h to synthetic dsRNA, poly IC (A; 500 mg mL1), a synthetic ssRNA, poly A (A; 500 mg mL1), PMA (B; 2 mg mL1), LPS (C; 100 mg mL1), and zymosan (D; 50 mg mL1) and were evaluated for HA as described in Section 2.

3.6. Integrin involvement in HA As the divalent cation requirement suggested the possible involvement of integrins, the effect of two integrin inhibitors on HA induction by 5 and 50 mg/ml poly IC was investigated. However, poly IC-induced HA was not impaired by either RGD (5 mg mL1), which is recognized by a subset of integrins and can block their binding to their normal ligands, or LFA-1 inhibitor (2.5e25 mg mL1), which binds a specific integrin, LFA1 and interferes with LFA-1 binding to its major ligand, intracellular adhesion molecule-1 (ICAM-1) (data not shown).

3.7. Involvement of PKR in HA As many actions of dsRNA are mediated by PKR, the effects of two PKR inhibitors were investigated (Fig. 7). The classic PKR inhibitor, 2-aminopurine, completely blocked the induction of HA by all concentrations of poly IC. The newer and more specific PKR inhibitor, compound 16 [29], completely blocked HA induction by 5 mg mL1 of poly IC and reduced the number of aggregates at higher poly IC concentrations.

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

493

40 24h pIC 12h pIC 8h pIC

Average # of aggregates

30

20

10

0 0.05

0

0.5

5

50

500

poly IC concentration (µg mL-1) Fig. 4. Effect of time and poly IC concentration on HA. RTS11 cells were treated with increasing concentrations of poly IC and were monitored over time for homotypic aggregation.

4. Discussion

35

35

30

30

25 20 15

**

10

** 5

0 g/mL

** 5

25

10

1

0 g/mL

CHX

0.5

AMD

Metabolic Inhibitors

*

15

5

0.05

*

20

**

0.1

pIC 4°C

Average # of aggregates

Average # of aggregates

RTS11 join a short list of myeloid cell lines capable of undergoing homotypic aggregation (HA) and is the first one to be described from a lower vertebrate. HA has been studied most intensively with U937 [4e8,10] and to a lesser extent with THP-1 [33], both of which are human monocytic leukaemia cell lines [34,35]. Other human cell lines

**

5

5

pIC

50

Col

500

**

**

0.05

0.5

Lat

Cytoskeleton Inhibitors

Fig. 5. Effect of metabolic and cytoskeleton inhibitors on poly IC-induced HA. RTS11 were exposed to 5 mg mL1 poly IC in combination with the inhibitors for 24 h in full growth media. The metabolic inhibitors included low temperatures, where cells were treated with poly IC and immediately placed at 4  C for 24 h. RTS11 were also treated with translation and transcription inhibitors, cycloheximide (CHX) and actinomycin D (AMD), respectively. Two cytoskeleton inhibitors, colchicine (Col) and latrunculin B (Lat) which disrupt microtubule and microfilament-mediated processes, were also used in combination with poly IC treatment. Treatments were compared using an unpaired t-test, all values which differed significantly from controls are indicated by two asterisks (P < 0.0001) or by one asterisk (P < 0.001).

494

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497 18 With Ca2+ & Mg2+ 16

Without Ca2+ & Mg2+

Average # of aggregates

14

12

10

8

6

4

2

0 0

0.05

0.5

5

50

500

poly IC concentration (µg mL-1) Fig. 6. Effect of calcium and magnesium on poly IC-induced HA. RTS11 were seeded in Hank’s balanced salt solution (HBSS) with or without 0.185 g L1 CaCl$2H2O and 0.098 g L1 MgSO4 (anhyd.) and dosed with increasing concentrations of poly IC.

that have been found to undergo HA are the myeloblastic cell line, KG1A [36], and the macrophage cell line, K1m [37]. Less research has been done on the aggregation of murine macrophage cell lines, like J774 [1,38]. RTS11 might be unique among the monocyte/macrophage cell lines in undergoing HA in response to a range of stimuli: synthetic RNAs, which are viral mimics; LPS, which is a cell wall component of gram-negative bacteria, and 40 pIC 2AP compound 16

Average # of aggregates

30

20

10

0 0

0.05

0.5

5

50

500

poly IC concentration (µg mL-1) Fig. 7. Effect of PKR inhibitors on poly IC-induced HA. Two PKR inhibitors, 2-aminopurine (2-AP) and compound 16, blocked poly IC-induced aggregation in RTS11 after a 24-h exposure. The negative control for compound 16 did not affect poly IC-induced HA (data not shown).

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

495

zymosan, which is derived from yeast cells walls, as well phorbol 12-myristate 13-acetate (PMA), which activates immune cells. The most studied inducers of HA in human myeloid cell lines have been PMA [7,8,10] and the ligation of specific clusters of differentiation (CD) [4,9]. There are fewer incidents of LPS and specific bacteria being shown to induce HA in U937 [5,6] but HA in response to synthetic RNA and zymosan appears to be a novel finding. As HA often accompanies the activation of immune cells [2], RTS11 might be a useful tool to screen for immunological stimuli in fish. However, RTS11 HA could vary with the stimulus as the overall morphology of the aggregates was ‘globular’ in response to PMA and poly A, but ‘stellate’ in response to poly IC, LPS and zymosan. These two morphologies have also been observed for HA of human macrophage cell lines, but not in the same cell line. The aggregates induced in U937 by PMA also were described as ‘globular’ [7], whereas ‘stellate’ aggregates were seen in K1m after ligation of HLA (class II) [37]. This is the first report of HA induction by dsRNA and the possible involvement of dsRNA-dependent protein kinase (PKR). PKR is activated by dsRNA and is responsible for many actions in mammalian macrophages [39]. Several observations suggested that poly IC-induced HA in RTS11 was mediated by PKR. Firstly, RTS11 express PKR, as evidenced by the ability of poly IC to stimulate phosphorylation of eIF2a (in preparation), which is a common substrate for PKR. Secondly, HA induction by poly IC was impeded by two inhibitors of PKR, 2-aminopurine (2-AP), which could possibly inhibit other protein kinases [40]; and compound 16, which appears to be a more specific inhibitor [29]. 2-AP completely inhibited HA induction and compound 16 dramatically reduced it. Thirdly, the much stronger induction of HA by poly IC than by poly A supported the involvement of PKR. In human monocyte/macrophage cell lines, the induction of HA by protein kinase C (PKC) agonists (PMA and indolactam V) and antagonists (staurosporine) and by phosphatase inhibitors (okadaic acid and calyculin A) suggests that multiple protein kinase pathways can be involved in HA [41,42,1,9]. The results with the rainbow trout monocyte/macrophage cell line suggest that PKR should be added to the list. Some characteristics of poly IC-induced HA of RTS11 have also been found in the HA of mammalian myeloid cell lines and of fish leucocytes, suggesting some common underlying mechanisms. Low temperature and disrupting the cytoskeleton inhibited HA of RTS11, seabream macrophages [13], and the human U937 line [1,4,11]. Divalent cations were required for HA of RTS11, seabream macrophages [13], and channel catfish neutrophils [12]. For U937, HA was cation dependent in response to some stimuli but not others [1,4,11]. Finally, the inhibition by cycloheximide suggested that HA induction required the synthesis of new proteins in RTS11, as well as in the HA of U937 in response to PMA [1]. In the case of poly IC as a stimulus, translation might be expected to be reduced due to eIF2a phosphorylation, but for mammalian cells, translation can be sustained for some messages, including those for intercellular adhesion molecules such as ICAM-1, VCAM-1 and E-selectin [43,44]. Whether the induction of RTS11 HA by poly IC is mediated by the induction and/or release of one or more cytokines is unknown. For mammalian mast cells, HA is induced by interleukin-4 (IL-4) [45]. IL-4 also appeared to have a role in human monocyte aggregation [46], but the role of this or other cytokines in the HA of mammalian macrophage cell lines is poorly understood. For RTS11, poly IC induced the expression of IL-8 and g-interferon inducible protein (gIP) within the time frame of HA [19,47]. As well, medium from RTS11 cultures previously exposed to poly IC contained interferon-like activity [48]. Yet such medium failed to induce HA (data not shown). This does not completely rule out a role for cytokines in RTS11 HA, as the stability and complexity of the cytokines in conditioned medium might make their biological activity less obvious compared to their continuous production or presentation singly. The surface molecules responsible for poly IC-induced HA of RTS11 can only be speculated upon at this time. For mammalian myeloid cell lines, LFA-1 and ICAM-1 are often responsible for HA [5,7]. Poly IC increased ICAM-1 expression in a subset of human dendritic cells [49], suggesting a possible role for these molecules in poly IC-induced HA of RTS11. However, a commercial inhibitor, o-bromobenzoyl L-tryptophan, shown to block the LFA-1/ICAM-1 association, failed to block RTS11 HA. Also, RGD did not inhibit RTS11 HA. Although ICAM-1 does not contain an RGD motif [50], LFA-1 appears to be able to bind proteins containing the RGD sequence [51]. Further knowledge regarding rainbow trout LFA-1 and ICAM-1 and their interactions with inhibitors is necessary before these experiments can be used to rule out their involvement in RTS11 HA. Another class of surface protein that might be considered is the galectins, which are b galactoside-binding proteins. In mammals, galectins participate in eosinophil HA, can be divalent dependent [52] and are up-regulated in endothelial cells by poly IC via PKR [53]. In rainbow trout, viral hemorrhagic septicemia virus (VHSV) induced galectin-9 like transcripts in leucocytes [54].

496

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

Acknowledgements This research was supported by scholarships to SJD-O and HCHH, as well as a grant to NCB from the Natural Sciences and Engineering Research Council of Canada. References [1] Cho JY, Katz DR, Chain BM. Staurosporine induces rapid homotypic intercellular adhesion of U937 cells via multiple kinase activation. Br J Pharmacol 2003;140:269e76. [2] Teixeira MM, Williams TJ, Au BT, Hellewell PG, Rossi AG. Characterization of eosinophil homotypic aggregation. J Leukoc Biol 1995;57:226e34. [3] Wake A, Tanake Y, Nakatsuka K, Misago M, Oda S, Moromoto I, et al. Calcium-dependent homotypic adhesion through leukocyte functionassociated antigen-1/intracellular adhesion molecule-1 induces interleukin-1 and parathyroid hormone-related protein production on adult T-cell leukemia cells in vitro. Blood 1995;86:2257e67. [4] Cho JY, Fox DA, Horejsi V, Sagawa K, Skubitz KM, Katz DR, et al. The functional interactions between CD98,b1-integrins, and CD147 in the induction of U937 homotypic aggregation. Blood 2001;98:374e82. [5] Moese S, Selbach M, Meyer TF, Backert S. cagþ Helicobacter pylori induces homotypic aggregation of macrophage-like cells by upregulation and recruitment of intracellular adhesion molecule 1 to the cell surface. Infect Immun 2002;70:4687e91. [6] Yamauchi Y, Kuroki M, Imakiire T, Uno K, Abe H, Beppu R, et al. Opposite effects of thrombospondin-1 via CD36 and CD47 on homotypic aggregation of monocytic cells. Matrix Biol 2002;21:441e8. [7] Yue L, Bao A, Li J. Phosphorylated form of MacMARCKS is essential to LFA-1-dependent cellecell adhesion of U937 monocytic cells. J Cell Physiol 1999;181:355e60. [8] Cabanas C, Lastres P, Bellon T, Aller P, Figdor CG, Corbi A, et al. Induction of Lfa-1-mediated homotypic adhesions in promonocytic U-937 cells occurs independently of cell-differentiation. Biochim Biophys Acta 1991;1092:165e8. [9] Cho JY, Chain BM, Vives J, Horejsi V, Katz DR. Regulation of CD43-induced U937 homotypic aggregation. Exp Cell Res 2003;290: 155e67. [10] Cho JY, Skubitz KM, Katz DR, Chain BM. CD98-dependent homotypic aggregation is associated with translocation of protein kinase Cg and activation of mitogen-activated protein kinases. Exp Cell Res 2003;286:1e11. [11] Kasinrerk W, Tokrasinwit N, Phunpae P. CD147 monoclonal antibodies induce homotypic cell aggregation of monocytic cell line U937 via LFA-1/ICAM-1 pathway. Immunology 1999;96:184e92. [12] Ainsworth AJ, Ye Q, Xue L, Hebert P. Channel catfish, Ictalurus punctatus Rafinesque, neutrophil adhesion to selected extracellular matrix proteins, lipopolysaccharide, and catfish serum. Dev Comp Immunol 1996;20:105e14. [13] Mulero V, Pelegrin P, Sepulcre MP, Munoz J, Meseguer J. A fish cell surface receptor defined by mAB mediated leukocyte aggregation and deactivation. Dev Comp Immunol 2001;25:619e27. [14] Harris ES, McIntyre TM, Prescott SM, Zimmerman GA. The leukocyte integrins. J Biol Chem 2000;275:23409e12. [15] Qian Y, Ainsworth AJ, Noya M. Identification of a beta 2 (CD18) molecule in a teleost species, Ictalurus punctatus Rafinesque. Dev Comp Immunol 1999;23:571e83. [16] Nakao M, Fujiki K, Kondo M, Yano T. Detection of complement receptors on head kidney phagocytes of the common carp Cyprinus carpio. Fish Sci 2003;69:929e35. [17] Goetz FW, Iliev DB, McCauley LA, Liarte CQ, Tort LB, Planas JV, et al. Analysis of genes isolated from lipopolysaccharide-stimulated rainbow trout (Oncorhynchus mykiss) macrophages. Mol Immunol 2004;41:1199e210. [18] Brubacher JL, Secombes CJ, Zou J, Bols NC. Constitutive and LPS-induced gene expression in a macrophage-like cell line from the rainbow trout (Oncorhynchus mykiss). Dev Comp Immunol 2000;24:565e74. [19] Laing KJ, Zou J, Wang T, Bols NC, Hirono I, Aoki T, et al. Identification and analysis of an interleukin 8-like molecule in rainbow trout Oncorhynchus mykiss. Dev Comp Immunol 2002;26:433e44. [20] Fujiki K, Gauley J, Bols NC, Dixon B. Genomic cloning of novel isotypes of the rainbow trout interleukin-8. Immunogenetics 2003;55: 126e31. [21] Wang TH, Holland JW, Bols N, Secombes CJ. Cloning and expression of the first nonmammalian interleukin-11 gene in rainbow trout Oncorhynchus mykiss. FEBS J 2005;272:1136e47. [22] Zou J, Bird S, Truckle J, Bols N, Horne M, Secombes C. Identification and expression analysis of an IL-18 homologue and its alternatively spliced form in rainbow trout (Oncorhynchus mykiss). Eur J Biochem 2004;271:1913e23. [23] Zou J, Peddie S, Scapigliati G, Zhang Y, Bols NC, Ellis AE, et al. Functional characterisation of the recombinant tumor necrosis factors in rainbow trout, Oncorhynchus mykiss. Dev Comp Immunol 2003;27:813e22. [24] Laing KJ, Secombes CJ. Trout CC chemokines: comparison of their sequences and expression patterns. Mol Immunol 2004;41:793e808. [25] Lally J, Al Anouti F, Bols N, Dixon B. The functional characterisation of CK-1, a putative CC chemokine from rainbow trout (Oncorhynchus mykiss). Fish Shellfish Immunol 2003;15:411e24. [26] Reinhart B, DeWitte-Orr SJ, Van Es SJ, Bols N, Lee LEJ. Cell adhesion characteristics of a monocytic cell line derived from rainbow trout, Oncorhynchus mykiss. Comp Biochem Physiol Part A Mol Integr Physiol 2006;144:437e43. [27] DeWitte-Orr SJ, Zorzitto JR, Sutton LP, Bols NC. Preferential induction of apoptosis in the rainbow trout macrophage cell line, RTS11, by actinomycin D, cycloheximide and double stranded RNA. Fish Shellfish Immunol 2005;18:279e95.

S.J. DeWitte-Orr et al. / Fish & Shellfish Immunology 22 (2007) 487e497

497

[28] Ganassin RC, Bols NC. Development of a monocyte/macrophage-like cell line, RTS11, from rainbow trout spleen. Fish Shellfish Immunol 1998;8:457e76. [29] Jammi NV, Whitby LR, Beal PA. Small molecule inhibitors of the RNA-dependent protein kinase. Biochem Biophys Res Commun 2003;308:50e7. [30] O’Brien J, Wilson I, Orton T, Pognan F. Investigation of the Alamar Blue (resazurin) fluorescent dye for the assessment of mammalian cell cytotoxicity. Eur J Biochem 2000;267:5421e6. [31] Schirmer K, Chan AGJ, Greenberg GM, Dixon DG, Bols NC. Methodology for demonstrating and measuring the photocytotoxicity of fluoranthene to fish cells in culture. Toxicol In Vitro 1997;11:107e19. [32] Ganassin RC, Schirmer K, Bols NC. Cell and tissue culture. In: Ostrander GK, editor. The handbook of experimental animals: the laboratory fish. London: Academic Press; 2000. p. 631e51. [33] Park H, Park SG, Lee JW, Kim T, Kim G, Ko YG, et al. Monocyte cell adhesion induced by a human aminoacyl-tRNA synthetase-associated factor, p43: identification of the related adhesion molecules and signal pathways. J Leukoc Biol 2002;71:223e30. [34] Auwerx J. The human leukemia-cell line, Thp-1 e a multifaceted model for the study of monocyteemacrophage differentiation. Experientia 1991;47:22e31. [35] Sundstrom C, Nilsson K. Establishment and characterization of a human histiocytic lymphoma cell line (U-937). Int J Cancer 1976;17: 565e77. [36] Tada J, Omine M, Suda T, Yamaguchi N. A common signalling pathway via Syk and Lyn tyrosine kinases generated from capping of the sialomucins CD34 and CD43 in immature hematopoietic cells. Blood 1999;93:3723e35. [37] Dialynas DP, Lee MJ, Shao LE, Tan PC, Yu J. Phenotypic and functional characterization of a new human macrophage cell line K1m demonstrating immunophagocytic activity and signalling through HLA class II. Immunology 1997;90:470e6. [38] Kaji K, Takeshita S, Miyake K, Takai T, Kudo A. Functional association of CD9 with the Fc gamma receptors in macrophages. J Immunol 2001;166:3256e65. [39] Maggi Jr LB, Heitmeier MR, Scheuner D, Kaufman RJ, Buller RML, Corbett JA. Potential role of PKR in double-stranded RNA-induced macrophage activation. EMBO J 2000;19:3630e8. [40] Hu Y, Conway TW. 2-Aminopurine inhibits the double-stranded RNA-dependent protein kinase both in vitro and in vivo. J Interferon Res 1993;13:323e8. [41] Weeks BS, Iuorno J. Phosphatase inhibitors promote ICAM-1 and LFA-1 mediated homotypic aggregation of Jurkat and U937 cells. Biochem Biophys Res Commun 1996;226:82e7. [42] Weeks BS, Goldman S, Touma S, Payne M, Cadet P, Stefano GB. Morphine inhibits indolactam V-induced U937 cell adhesion and gelatinase secretion. J Cell Physiol 2001;189:179e88. [43] Offermann MK, Zimring J, Mellits KH, Hagan MK, Shaw R, Medford RM, et al. Activation of the double-stranded-RNA-activated protein kinase and induction of vascular cell adhesion molecule-1 by poly(I)poly(C) in endothelial cells. Eur J Biochem 1995;232:28e36. [44] Yang J, Xu Y, Zhu C, Hagen MK, Lawley T, Offermann MK. Regulation of adhesion molecule expression in Kaposi’s sarcoma cells. J Immunol 1994;152:361e73. [45] Toru H, Kinashi T, Ra C, Nonoyama S, Yata JI, Nakahata T. Interleukin-4 induces homotypic aggregation of human mast cells by promoting LFA-1/ICAM-1 adhesion molecules. Blood 1997;89:3296e302. [46] Kazai F, Chang J, Lopez A, Vadas M, Cunningham AL. Interleukin 4 and human immunodeficiency virus stimulate LFA-1-ICAM-1-mediated aggregation of monocytes and subsequent giant cell formation. J Gen Virol 1994;75:2795e802. [47] Laing KJ, Bols NC, Secombes CJ. A CXC chemokine sequence isolated from the rainbow trout Oncorhynchus mykiss resembles the closely related interferon-gamma-inducible chemokines CXCL9, CXCL10 and CXCL11. Eur Cytokine Netw 2002;13:462e73. [48] DeWitte-Orr SJ, Bols NC. Expression of antiviral genes in different rainbow trout cell lines. Integr Comp Biol 2002;42:1220. [49] Smits HH, de Jong EC, Schuitemaker JSN, Geijtenbeek TBH, van Kooyk Y, Kapsenberg ML, et al. Intercellular adhesion molecule-1/LFA-1 ligation favors human Th1 development. J Immunol 2002;168:1710e6. [50] Simmons D, Makgoba MW, Seed B. ICAM, an adhesion ligand of LFA-1, is homologous to the neural cell adhesion molecule NCAM. Nature 1988;331:624e7. [51] Gadek TR, Burdick DJ, McDowell RS, Stanley MS, Marsters JC, Paris KJ, et al. Generation of an LFA-1 antagonist by the transfer of the ICAM-1 immunoregulatory epitope to a small molecule. Science 2002;295:1086e9. [52] Hirashima M, Kashio Y, Nishi N, Yamauchi A, Imaizumi T, Kageshita T, et al. Galectin-9 in physiological and pathological conditions. Glycoconj J 2004;19:593e600. [53] Ishikawa A, Imaizumi T, Yoshida H, Nishi N, Nakamura T, Hirashima M, et al. Double-stranded RNA enhances the expression of galectin-9 in vascular endothelial cells. Immunol Cell Biol 2004;82:410e4. [54] O’Farrell C, Vaghefi N, Cantonnet M, Buteau B, Boudinot P, Benmansour A. Survey of transcript expression in rainbow trout leukocytes reveals a major contribution of interferon-responsive genes in the early response to a rhabdovirus infection. J Virol 2002;76:8040e9.