MMP-9 mutant impairs homing of chronic lymphocytic leukemia cells by altering migration regulatory pathways

Biochemical and Biophysical Research Communications xxx (2017) 1e7

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic leukemia cells by altering migration regulatory pathways
n a, Noemí Aguilera-Montilla a, Alejandra Gutie
rrez-Gonza
lez a, Elvira Bailo b b Estefanía Ugarte-Berzal , Philippe E. Van den Steen , Ghislain Opdenakker b,
A. García-Marco c, Angeles García-Pardo a, * Jose
gicas, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain Cellular and Molecular Medicine Department, Centro de Investigaciones Biolo Rega Institute for Medical Research, Department of Microbiology and Immunology, University of Leuven, KU Leuven, Belgium c Hematology Department, Hospital Universitario Puerta de Hierro, Madrid, Spain a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 October 2017 Accepted 25 October 2017 Available online xxx

We previously showed that MMP-9 overexpression impairs migration of primary CLL cells and MEC1 cells transfected with MMP-9. To determine the contribution of non-proteolytic activities to this effect we generated MEC-1 transfectants stably expressing catalytically inactive MMP-9MutE (MMP-9MutEcells). In xenograft models in mice, MMP-9MutE-cells showed impaired homing to spleen and bone marrow, compared to cells transfected with empty vector (Mock-cells). In vitro transendothelial and random migration of MMP-9MutE-cells were also reduced. Biochemical analyses indicated that RhoAGTPase and p-Akt were not downregulated by MMP-9MutE, at difference with MMP-9. However, MMP-9MutE-cells or primary cells incubated with MMP-9MutE had significantly reduced p-ERK and increased PTEN, accounting for the impaired migration. Our results emphasize the role of non-proteolytic MMP-9 functions contributing to CLL progression. © 2017 Elsevier Inc. All rights reserved.

Keywords: CLL Non-proteolytic MMP-9 Signaling pathways Cell migration Homing

1. Introduction Chronic lymphocytic leukemia (CLL), the most common leukemia in Western countries, is characterized by the accumulation of CD5þ B lymphocytes in peripheral blood and lymphoid organs [1]. Localization in these niches allows malignant cells to receive proliferative and survival signals, thus contributing to CLL progression [2]. Several molecules regulate the migration and organ localization of CLL cells, including chemokines, integrins and gelatinase-B/ matrix metalloproteinase-9 (MMP-9) [3]. MMP-9 is synthesized by CLL cells and is abundant in the CLL microenvironment [4,5]. We previously showed that binding of MMP-9 to primary CLL cells or stably transfecting MMP-9 into MEC1 cells impairs in vivo and in vitro cell migration [6]. The exact mechanism accounting for this effect is not known, but it includes regulation of relevant signalling molecules, such as RhoAGTPase,

* Corresponding author. Cellular and Molecular Medicine Department, Centro de
gicas, CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain. Investigaciones Biolo E-mail address: [email protected] (A. García-Pardo).

Akt, ERK and PTEN [6]. We have also shown that interaction of MMP-9 with a4b1 integrin in CLL cells induces a signalling pathway that leads to cell survival [7]. Importantly, the MMP-9 hemopexin domain or an MMP-9 mutant devoid of catalytic activity (MMP9MutE) also induced cell survival, indicating that MMP-9 may exert functions not involving its enzymatic activity. Non-enzymatic functions have also been reported for other MMPs [8]. For example, MMP-1 and MMP-2 induce intracellular signaling upon binding to a2b1 or aVb3 integrins, respectively [9,10]. MMP-3 mediates epithelial cell growth by non-proteolytic mechanisms [11]. The cytoplasmic tail of MMP-14, but not the catalytic domain, was involved in macrophage invasion and myeloid cell fusion [12,13]. These evidences, together with our previous findings, highlight the multiple roles of MMPs and the need to continue studying these additional MMP properties. To further establish the non-enzymatic functions of MMP-9 that contribute to CLL pathology, in the present report we have generated MEC-1 cells stably expressing the catalytically inactive MMP9MutE protein and have studied their behaviour in cell migration. We demonstrate that MMP-9MutE-cells have altered migration

https://doi.org/10.1016/j.bbrc.2017.10.129 0006-291X/© 2017 Elsevier Inc. All rights reserved.


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo

2

regulatory pathways, resulting in impaired transendothelial migration and homing to lymphoid organs. MMP-9MutE-cells are therefore an excellent model to identify non-catalytic functions of MMP-9 contributing to CLL progression. 2. Materials and methods 2.1. Patients, cells and cell cultures Approval was obtained from the CSIC Bioethics Review Board for these studies. Peripheral blood samples from 6 CLL patients (Table 1) were obtained after informed consent and B-lymphocytes were purified as described [5e7]. The MEC-1 cell line, established from a CLL patient [14], was purchased from the German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany) and maintained in IMDM medium (Lonza, Basel, Switzerland), 10% FBS. HUVEC were purchased from Lonza and cultured as reported [5,6]. HEK293T cells (Invitrogen, Prat del Llobregat, Barcelona, Spain) were cultured in DMEM (Lonza), 10% FBS, 1% Glutamax (Invitrogen), 10 mg/ml penicillin/streptomycin.

CGGCGCATGCGTTCGGCCA-30 (forward) and 50 -TGGCCGAACGCATGCGCCG-30 (reverse). The mutated sequence was re-inserted into the pRRLsin18.CMV.IRES.eGFP lentivirus and confirmed by DNA sequencing (Secugen, Madrid, Spain). HEK293T cells were transiently transfected with this construct and viral supernatants were used to transfect MEC-1 cells as reported [6]. GFP-expressing cells were selected by several cell sorting steps until more than 95% of the cells were clearly positive for expression. These MEC-1 cell transfectants will be referred to as MMP-9MutE-cells.

2.4. Flow cytometry 1.5  105 primary CLL cells or MEC-1 transfectants were incubated (30 min, 4  C) in 100 ml PBS/1%BSA with appropriate primary antibodies, washed with cold PBS and incubated (30 min, 4  C) with Alexa 488-labeles (CLL) or Alexa 647-labeled (MEC-1) secondary Abs. Samples were analyzed on a Coulter Epics XL or FC 500 flow cytometer (Beckman Coulter, Fullerton, CA).

2.2. Antibodies and reagents

2.5. Immunofluorescence analyses

Rabbit IgG (sc-3888), rabbit polyclonal antibodies (RpAb) to MMP-9 (sc-6841R), monoclonal antibodies (mAbs) against total Akt (sc-5298), PTEN (sc-7974), RhoA (sc-418), and CD44 (sc-7297) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA Diego, CA, USA). mAbs to CD38 (16BDH), a4 (HP1/7) and b1 (Alex1/4) integrin subunits were from Dr. F. S
anchez-Madrid (Hospital de la Princesa, Madrid, Spain). Rat mAb anti-CCR7 (3D12) was from BD Pharmigen (Franklin Lakes, NJ, USA). RpAbs against phospho-Akt (Ser473, #9271), phospho-myosin light chain (MLC, Thr18/Ser19, #3674), phospho-ERK (Thr202/Tyr204, #9101) and total ERK (#9102) were from Cell Signalling Technology, Inc. (Beverly, MA, USA). mAb to vinculin (#V9131) was from Sigma-Aldrich (St. Louis, MO, USA). HRP-labeled Abs to rabbit or mouse Ig were from Dako (Glostrup, Denmark). Alexa 488- or 647-labeled Abs, Alexa 568-Phalloidin (#A12380) and Alexa 647-Phalloidin (#A22287) were from Molecular Probes (Eugene, OR, USA). VCAM-1 (vascular cell adhesion molecule-1) and TNFa were from R&D Systems (Minneapolis, MN, USA). CCL21 was from Peprotech EC (London, UK).

5  105 cell transfectants were added to glass coverslips previously coated with 5 mg/ml poly-lysine-1% BSA and incubated at 37  C for 2 h. Cells were fixed with 4% paraformaldehyde, permeabilized with PBS/0.1% Triton X-100 (15 min, RT), and incubated with Alexa 568-Phalloidin for 1 h. Cells were washed with PBS/1% BSA and images acquired on a Leica TCS-SP2-AOBS-UV confocal microscope with a 63 oil immersion objective. The LAS-AS Leica software was used for cell area determination.

2.6. DQ-gelatin degradation assay This assay was performed as described [15,16]. Briefly, 2  106 MEC-1 cells were cultured in 2 ml IMDM/0.1%FBS in 6-well plates. After 24 h, the conditioned media were concentrated and added to a solution of 2.5 mg/ml DQ-gelatin (Invitrogen) in a 96 well plate. The plates were immediately placed in a microplate fluorescence reader (FL600, Biotek, Highland Park, IL, USA) and fluorescence was measured every 10 min for 2 h at 37  C (ex 485 nm/em 530 nm).

2.3. Plasmid construction and lentiviral production and infection The preparation and characterization of MEC-1 cells stably expressing proMMP-9 (MMP-9-cells) or empty vector (Mock-cells) was previously described [6]. MEC-1 cells stably expressing the catalytically inactive proMMP-9MutE mutant were generated from the pRRLsin18.CMV.IRES.eGFP-proMMP-9 lentivirus construct by mutating the glutamate residue at position 402 in the proMMP-9 sequence to alanine, using the primers 50 -

Table 1 Clinical characteristics of CLL patients. Patients Sex/Agea Stageb Ig Statusc CD38 a4 subunit (%) b1 subunit (%) 1 2 3 4 5 6 a b c

M/79 M/86 M/78 M/66 M/75 M/75

B/II A/I B/II B/II ND B/II

UM ND UM Mut Mut UM

þ e þ þ þ þ

30.1 39.8 20.0 90.9 45.6 75.7

M, male. Stage according to references [26,27]. Mut, mutated Ig; UM, unmutated Ig; ND, not determined.

95.7 16.8 37.5 99.4 59.9 73.9

2.7. Cell migration assays In vivo experiments with mice were performed with the approval of the Ethics Committee of the CSIC. 5  106 Mock-cells, MMP-9-cells or MMP-9MutE-cells were labeled with 10 mM CFSE (Invitrogen) and injected into the tail vein of 6- to 10-week-old NOD/SCID mice. After 3 h, mice were sacrified and organs extracted and disaggregated. The number of migrated cells was determined by flow cytometry. Numbers of homed cells were normalized to the number of injected, viable CLL cells and to the total number of mouse cells in that organ. Transendothelial migration assays were performed exactly as described [5,6]. For time-lapse microscopy analyses, 5  104 MEC-1 cells were added to chemotaxis chambers (Ibidi, Martinsried, Germany) coated with 2 mg/ml VCAM-1. Images were acquired every 30 s for 3 h using a Leica AF6000 LX microscope (20 objective). Mean accumulated distance, representing the movement of 40 cells, was determined using ImageJ software (U.S. National Institutes of Health, Bethesda, MD, USA) and chemotaxis tool (Ibidi). Additional methods are available as Supplemental Material.


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo

3. Results 3.1. Generation and characterization of cell transfectants stably expressing the catalytically inactive MMP-9MutE mutant To establish MMP-9 functions not attributable to its proteolytic activity, we prepared cell transfectants stably expressing the catalytically inactive MMP-9MutE mutant. MEC-1 cells were chosen for these experiments because of their very low constitutive expression of MMP-9 [6]. Cells were transfected with a lentivirus containing the mutated MMP-9 DNA, following the same protocol previously used to generate MMP-9-cells (containing wild-type MMP-9) and Mock-cells (containing the empty vector) [6]. Transfection efficiency on selected clones (named 1H3 and 2H11) was 100%, measured by flow cytometry. Expression of MMP-9MutE mRNA in the two clones was confirmed by qPCR and was similar to the MMP-9 mRNA expression in MMP-9-cells (Fig. 1A). MMP-9 mRNA was present at very low levels in Mock-cells (Fig. 1A), as we previously reported [6]. To assess MMP-9 expression at the protein level, MEC-1 cell transfectants were cultured in the absence of serum for 24 h and cell lysates and conditioned media were analyzed by Western blotting. Fig. 1B shows that MMP-9 was undetectable in Mock-cells, but clearly present in MMP-9-cells and in the two clones of MMP9MutE-cells, both as intracellular (cell lysate) and secreted form (conditioned medium). Because MMP-9 is also generally present at the surface of CLL cells [17,18], we analyzed cell-associated MMP-9 in MEC-1 transfectants by flow cytometry. Mock-cells showed little surface MMP-9 (Fig. S1A). Both 1H3 and 2H11 MMP-9MutE-cell clones displayed membrane-bound MMP-9, with similar

expression (range 10.9%e20.6%) than MMP-9-cells (range 15.9%e 20.2%) (Fig. S1A). Additionally, transfection of MEC-1 cells with MMP-9MutE did not affect the surface expression of relevant surface molecules, such as a4b1 integrin, the CCL21 receptor CCR7, or CD44 (Fig. S1B). To confirm the lack of catalytic activity of MMP-9 in MMP9MutE-cells, we performed gelatin zymography analyses, using identical aliquots from the conditioned medium of Mock-, MMP-9-, and MMP-9MutE-cells. Fig. 1C shows that MMP-9 gelatinolysis was only visible in the medium of MMP-9-cells. We also studied the degradation of the fluorogenic substrate DQ-gelatin. In these assays, MMP-9-cells conditioned medium efficiently degraded DQgelatin in a time-dependent manner, while the medium of 1H3 or 2H11 MMP-9MutE-cells did not (Fig. 1D). Altogether these results confirmed the lack of catalytic activity of the MMP-9MutE protein and established that MMP-9MutE-cells constitute a good cellular model to study MMP-9 functions not attributable to its catalytic activity. Because both MMP-9MutE clones behaved similarly all subsequent experiments were performed with MutE-1H3 clone and referred to as MMP-9MutE-cells.

3.2. MMP-9MutE-cells display impaired in vivo and in vitro migration We previously showed that MMP-9-cells display impaired in vivo and in vitro migration, compared to Mock-cells [6]. To determine the role of the catalytic activity of MMP-9 in this effect we studied the migratory behaviour of MMP-9MutE-cells. CFSElabeled Mock-, MMP-9-, or MMP-9MutE-cells were injected into the tail vein of NOD-SCID mice, using six mice for each cell type.

B

MMP-9 mRNA expression

A

3

MMP-9MutE

2500 n=3

***

2000 1500

Mock MMP-9 1H3 2H11

***

MMP-9

***

1000

MMP-9

500 0

Cell lysate

Vinculin

Cond. medium

Vinculin Mock MMP-9 1H3

2H11

MMP-9MutE

D MMP-9MutE MMP-9MutE Mock Mock MMP-9 MMP-9 1H3 1H3 2H11 2H11 MMP-9 MMP-9 Vinculin Vinculin

20 n=3 Fluorecent units

C

15 Mock MMP-9 1H3 2H11

10 5 0

0

20

40

60

80 100 120 Time (min)

Fig. 1. Characterization of MMP-9MutE-cells. (A) qPCR analysis of MMP-9 mRNA expression in Mock-cells, MMP-9-cells and two different clones (1H3 and 2H11) of MMP-9MutEcells. (B) Expression of MMP-9 protein in cell lysates and conditioned media from clones 1H3 and 2H11, analyzed by Western blotting. Mock-cells and MMP-9-cells were also analyzed for comparison. (C) Gelatin zymography analyses of the conditioned media of the indicated MEC-1 cell transfectants (2  106 cells). (D) The conditioned media of MEC1 cell transfectants (2  106 cells) were concentrated and added to plates coated with 2.5 mg/ml DQ-gelatin. Fluorescence was determined at the indicated times and average values from three experiments are shown. ***P < 0.001.


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo

4

42193 cells

21197 cells

Mock-cells

PBS

298 cells

MMP-9MutE-cells

MMP-9-cells

10612 cells

35629 cells

MMP-9MutE-cells

8363 cells

6162 cells

% Cells homed to Spleen

594 cells

MMP-9-cells

% Cells homed to Bone marrow

Mock-cells

PBS

FL1 LOG

Bone marrow CFSE MEC-1 cells

Spleen CFSE MEC-1 cells

A

FL2 LOG

C ***

40 20 0

- CCL21

MMP-9-cells

Mock-cells

+ CCL21

Y axis (μm)

120 n=6 Mock 100 MMP-9 80 MutE 60

***

% Transendothelial migration

B

120 n=6 100 * 80 *** 60 40 20 0 Ctrl Mock MMP-9 MutE 120 n=6 100 80 *** 60 40 20 0 Ctrl Mock MMP-9 MutE

MMP-9MutE-cells

500

500

500

0

0

0

-500

0

500

X axis (μm)

-500

0

500

X axis (μm)

-500

0

500

X axis (μm)

Fig. 2. Effect of MMP-9MutE expression on MEC-1 cell migration and homing. (A) Cells were labeled with 10 mM CFSE and injected into the tail vein of NOD/SCID mice (5  106 cells/ mouse). Six mice were used for each cell type and control mice (Ctrl) received only PBS. After 3 h, labeled cells in the spleen and bone marrow were determined by flow cytometry. Numbers inside the charts represent CFSE cells in one femur or spleen, normalized according to the total number of cells/ml in that organ and the number of injected, viable cells. Average values ± SE after normalizing the migration of Mock-cells to 100 are shown. (B) MEC-1 cell transfectants (3  105) were added to the upper chamber of Transwell filters coated with HUVEC and allowed to migrate in response to CCL21 (100 ng/mL) for 24 h. Migration was determined by flow cytometry, and average values represent the percentage of the total number of cells added, after normalizing the number of migrated Mock-cells to 100. (C) 5  104 cells were added to chemotaxis chambers coated with 2 mg/ml VCAM-1, and cell movement was monitored every 30 s for 3 h using a Live Cell Imaging microscope. *P < 0.05; ***P < 0.001.

Control mice were injected with PBS. Mice were sacrificed after 3 h and the presence of MEC-1 cells in bone marrow and spleen was assessed by organ disaggregation and flow cytometry. Fig. 2A shows that the homing of MMP-9-cells to spleen and bone marrow was reduced by 46% and 34%, respectively, compared with Mockcells, confirming our previous results [6]. Notably, the migration of MMP-9MutE-cells to spleen was also significantly reduced by 26%, compared to Mock-cells (Fig. 2A). MMP-9MutE cell homing to bone marrow was reduced by 15% (Fig. 2A). We also analyzed the behavior of MMP-9MutE-cells in transendothelial migration assays in vitro. 41% of Mock-cells migrated through HUVEC in response to CCL21 and this was normalized to 100. The migration of MMP-9MutE-cells was partially (26%) and significantly reduced compared to Mock-cells (Fig. 2B). In these experiments, the migration of MMP-9-cells was reduced by 67% (Fig. 2B). Lastly, we measured random migration of Mock-, MMP-9and MMP-9MutE-cells by time-lapse microscopy. Fig. 2C shows that the motility of MMP-9MutE-cells was reduced compared to Mock-cells, with a calculated mean accumulated distance of 293.3 mm (MMP-9MutE-cells) and 439.7 mm (Mock-cells). The mean accumulated distance for MMP-9-cells was lower (255 mm) than for MMP-9MutE-cells, in agreement with the partial effect of MMP-9MutE on migration shown above. Altogether, these results established that the catalytically inactive MMP-9MutE reproduced the effects of MMP-9 on CLL cell migration, although it was less

effective than wild-type MMP-9. 3.3. MMP-9MutE affects migration regulatory pathways in MEC-1 transfectants and primary CLL cells Impairment of MEC-1 and primary CLL cell migration by MMP-9 involves regulation of several molecules, including RhoA, Akt, ERK, and PTEN (6). We thus determined whether MMP-9MutE also regulated these molecules. At difference with the effect of MMP-9, RhoA activation or MLC phosphorylation at Thr18/Ser19 (Rho kinase target) were not downregulated in MMP-9MutE-cells, compared to Mock-cells (Figs. S2A and B). In agreement with a functional RhoA, Mock-cells and MMP-9MutE-cells displayed filopodia-like structures with similar areas (402.1 mm2 vs 417.5 mm2, respectively, average of 80 cells), while the area of MMP-9-cells was significantly reduced (218.8 mm2) (Fig. S2C). Likewise, Akt phosphorylation was not significantly altered in MMP-9MutE-cells, (Fig. 3A). In contrast, ERK phosphorylation was significantly downregulated and PTEN significantly upregulated in both types of transfectants (Fig. 3A). As observed for cell migration, the effect of MMP-9MutE on these molecules was partial, compared to the effect of wild-type MMP-9 (Fig. 3A). To further confirm these results and to enhance their physiological significance, we studied whether regulation of these molecules by MMP-9MutE was also observed in primary CLL cells. CLL


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo

A

5

MEC-1 cell transfectants

0.8

***

0.4

p-ERK/t-ERK

p-Akt/t-Akt

1.2

Mock MMP-9 MutE

B

*

0.8

***

0.4 0

Mock MMP-9 MutE PTEN Vinc.

1.2 n=8

1.6 n=5

0

Mock MMP-9 MutE p-ERK t-ERK

PTEN/Vinculin

Mock MMP-9 MutE p-Akt t-Akt

Mock MMP-9 MutE

5 n=7 *** 4 * 3 2 1 0 Mock MMP-9 MutE

Primary CLL cells Ctrl +MMP-9 +MutE

p-Akt t-Akt

0.8

1.2 n=6

**

0.4 0

Ctrl +MMP-9 +MutE

pERK/tERK

p-Akt/t-Akt

1.2 n=5

0.8

* ***

0.4 0

Ctrl +MMP-9 +MutE PTEN Vinc.

p-ERK t-ERK

Ctrl +MMP-9 +MutE

1.6 PTEN/Vinculin

Ctrl +MMP-9 +MutE

1.2

n=5

**

*

0.8 0.4 0

Ctrl +MMP-9 +MutE

Fig. 3. MMP-9MutE expression affects the activation of molecules involved in cell migration. (A) 2  106 MEC-1 cell transfectants were cultured for 2 h in IMDM/0.1% FBS, lysed and analyzed by Western blotting, using vinculin (Vinc) or total protein as loading controls. Mock-cell values were normalized to one. (B) 5-10  106 primary CLL cells were serumstarved for 1 h and incubated with or without (control, Ctrl) with 110 nM MMP-9 or MMP-9MutE protein, in RPMI/0.1% FBS. After 1 h (Akt, ERK) or 3 h (PTEN) cells were lysed and analyzed by Western blotting. Control cell values were normalized to 1.*P < 0.05; **P < 0.01; ***P < 0.001.

cells were incubated with MMP-9MutE for 1 h (Akt, ERK), or 3 h (PTEN), lysed and analyzed by Western blotting. Cells were also incubated with MMP-9 for comparison. Fig. 3B shows that incubation of primary CLL cells with MMP-9MutE did not affect Akt phosphorylation, but significantly downregulated p-ERK and upregulated PTEN, confirming the behaviour of MMP-9MutE-cells. Altogether, these results indicated that the catalytically inactive MMP-9MutE mutant was able to regulate signaling pathways affecting migration of MEC-1 cell transfectants and primary CLL cells. 4. Discussion To further understand the functions of MMP-9 contributing to CLL progression, in this study we have focused on possible nonproteolytic roles of MMP-9 on cell migration. Using MEC-1 cell transfectants stably expressing the catalytically inactive MMP9MutE or incubating primary CLL cells with this mutant, we demonstrate that MMP-9 also regulates CLL cell migration by nonproteolytic mechanisms. The MEC-1 MMP-9MutE-cells generated in this study expressed and secreted similar levels of mutant MMP-9 than the transfectants carrying wild-type MMP-9, but were clearly devoid of MMP-9 proteolytic activity. When compared in migration experiments, MMP-9MutE-cells displayed reduced migration in vivo and in vitro, as previously observed for MMP-9-cells ([6] and this report). However, MMP-9MutE was less effective than MMP-9 in regulating signaling pathways and impairing cell migration. This may suggest that the MMP-9 catalytic and non-catalytic mechanisms of regulating CLL cell migration are different or, perhaps more likely, that

proper regulation of cell migration involves two (or more) MMP-9 domains, one being the catalytic region. In the absence of an active catalytic site the inactive mutant can only achieve a partial effect. In a previous study, we showed that dysregulation of RhoA, Akt, ERK and PTEN was involved in the impaired migration and homing of MEC-1 MMP-9-cells and of primary CLL cells incubated with MMP-9 [6]. In the present report we have addressed whether similar mechanisms accounted for the reduced migration of MMP9-MutE-cells. We have found some differences between the two cell systems. First, the constitutive RhoA activity in MMP-9MutEcells was not reduced with respect to Mock-cells, as was the case for MMP-9-cells. This suggested that 1) downregulation of this GTPase likely requires the MMP-9 proteolytic activity, and 2) RhoA activity is not involved in the impaired migration of MMP-9MutEcells. In agreement with a functional RhoA in MMP-9MutE-cells, our present data shows that phosphorylation of MLC was similar in MMP-9MutE-cells and in Mock-cells. Another signaling pathway not affected by MMP-9MutE expression was the phosphorylation of Akt. This also differs from the effect of wild-type MMP-9, which significantly downregulates p-Akt ([6] and this report). MMP-9MutE, however, behaved like wild-type MMP-9 in the ability to downregulate the constitutive phosphorylation of ERK and to upregulate PTEN. Notably, our study shows a similar response in primary CLL cells incubated with the recombinant MMP-9MutE protein, strongly supporting the results obtained with MMP-9MutE-cells and highlighting the role of ERK and PTEN in CLL cell migration. Indeed, ERK is a well-established regulator of cell migration [19] and we previously demonstrated that blocking ERK (or Akt) activity reduces CLL cell transendothelial migration [5,6]. The role of PTEN as inhibitor of cell migration, via


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo

6

its phosphatase activity, is also well defined [20]. In fact, PTEN can prevent ERK phosphorylation via inhibition of the Shc adaptor protein [21], thus establishing a link between the two molecules regulated by MMP-9MutE. Modulation of these proteins thus provides a mechanism for the reduced in vivo and in vitro migration of MMP-9MutE-cells reported here, and for the previously observed impaired homing of primary CLL cells incubated with MMP-9MutE [6]. An increasing number of studies are reporting non-enzymatic activities of MMPs in many cell systems. For example, the cytoplasmic tail of MT1-MMP, but not the catalytic domain, enhances HIF and the expression of HIF target genes, such as VEGF, in several tumor models [22]. Likewise, MT1-MMP binding to tissue inhibitor of metalloproteinase-2 or to b1 integrin activates ERK1/2 independently of its catalytic activity [23,24]. In the case of MMP9, its interaction with CD44 activates EGFR and its downstream effectors pERK and pAkt, resulting in increased cell migration [25]. MMP-9 also induces CLL cell survival by non-proteolytic mechanisms, as we previously reported [7]. Our present study is the first to report that a catalytically inactive MMP-9 mutant modulates relevant pathways impacting in CLL cell migration. MMP-9MutE exerts this function when stably expressed in MEC-1 cells and when externally added to primary CLL cells, and may therefore involve common receptor(s) in both cell systems. A possible candidate receptor is a4b1 integrin, which is expressed in many CLL cases, including all samples studied here, and binds equally well to MMP-9 and MMP-9MutE [18]. In support of this, induction of CLL cell survival by MMP-9 or MMP-9MutE was mediated by a4b1 integrin [7]. The involvement of other receptors, however, cannot be disregarded. The ability of MMP-9 to inhibit migration may be very relevant for CLL cells in lymphoid organs, where MMP-9 levels, both intrinsic and/or extrinsic, are increased. Retention of CLL cells in lymphoid niches would favor cell survival and drug resistance, leading to CLL progression. The fact that these functions may or may not involve the MMP-9 catalytic activity implies that they can be exerted even in the presence of proteolytic inhibitors. Our study therefore expands the pathological roles of MMP-9 in CLL, further highlighting that it may constitute a therapeutic target in this disease. Authorship E.B. performed most of the research, designed some experiments, and analyzed data. N.A-M, A.G-G and E.U-B. performed some experiments and analyzed data. J.A.G.-M. contributed patient samples and clinical data. G.O. and P.E.V.d.S. prepared and characterized the recombinant MMP-9 variants and critically reviewed the manuscript. A.G-P. designed and supervised research and wrote the paper. All authors approved the final version of the manuscript. Acknowledgements We thank Drs. Pedro Lastres and Gema Elvira for help with the flow cytometry and confocal analyses, respectively. This work was
tica de Invessupported by Grant SAF2015-69180R and Red Tema
n Cooperativa en Ca
ncer Grant RD12/0036/0061 from the tigacio Ministry of Economy and Competitivity (Spain) (to A.G-P.); S2010/ BMD-2314 (to A.G-P.) from the Comunidad de Madrid/European Union; and by the Concerted Research Actions (Grant GOA 2013/ 015 and grant C1/2017 project C16/17/010), and the Foundation for Scientific Research of Flanders (FWO-Vlaanderen, Grants G0A7516N and G0A5716N) to G.O. EUB is a postdoctoral researcher of the FWO-Vlaanderenc.

Transparency document Transparency document related to this article can be found online at https://doi.org/10.1016/j.bbrc.2017.10.129. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.bbrc.2017.10.129. References [1] M. Hallek, Chronic lymphocytic leukemia: 2017 update on diagnosis, risk stratification, and treatment, Am. J. Hematol. 92 (2017) 946e965. [2] E. Ten Hacken, J.A. Burger, Microenvironment interactions and B-cell receptor signaling in Chronic Lymphocytic Leukemia: implications for disease pathogenesis and treatment, Biochim. Biophys. Acta 1863 (2016) 401e413. [3] M.S. Davids, J.A. Burger, Cell trafficking in chronic lymphocytic leukemia, Open J. Hematol. 3 (2012). [4] B. Bauvois, J. Dumont, C. Mathiot, et al., Production of matrix metalloproteinase-9 in early stage B-CLL: suppression by interferons, Leukemia 16 (2002) 791e798. ~ oz, E. Escobar-Díaz, R. Samaniego, et al., MMP-9 in B-cell [5] J. Redondo-Mun chronic lymphocytic leukemia is up-regulated by alpha4beta1 integrin or CXCR4 engagement via distinct signaling pathways, localizes to podosomes, and is involved in cell invasion and migration, Blood 108 (2006) 3143e3151.
n, E. Ugarte-Berzal, I. Amigo-Jime
nez, et al., Overexpression of proge[6] E. Bailo latinase B/proMMP-9 affects migration regulatory pathways and impairs chronic lymphocytic leukemia cell homing to bone marrow and spleen, J. Leuk. Biol. 96 (2014) 185e199. ~ oz, E. Ugarte-Berzal, M.J. Terol, et al., Matrix [7] J. Redondo-Mun metalloproteinase-9 (MMP-9) promotes chronic lymphocytic leukemia B-cell survival through its hemopexin domain, Cancer Cell 17 (2010) 160e172. [8] A. Garcia-Pardo, G. Opdenakker, Nonproteolytic functions of matrix metalloproteinases in pathology and insights for the development of novel therapeutic inhibitors, Metalloproteinases Med. 2 (2015) 19e28. [9] K. Conant, C. St Hillaire, H. Nagase, et al., Matrix metalloproteinase 1 interacts with neuronal integrins and stimulates dephosphorylation of Akt, J. Biol. Chem. 279 (2004) 8056e8062. [10] C. Chetty, S.S. Lakka, P. Bhoopathi, et al., MMP-2 alters VEGF expression via alphaVbeta3 integrin-mediated PI3K/AKT signaling in A549 lung cancer cells, Int. J. Cancer 127 (2010) 1081e1095. [11] K. Kessenbrock, G.J. Dijkgraaf, D.A. Lawson, et al., A role for matrix metalloproteinases in regulating mammary stem cell function via the Wnt signaling pathway, Cell Stem Cell 13 (2013) 300e313. [12] T. Sakamoto, M. Seiki, Cytoplasmic tail of MT1-MMP regulates macrophage motility independently from its protease activity, Genes Cells 14 (2009) 617e626.
ndez-Riquer, et al., MT1-MMP is [13] P. Gonzalo, M.C. Guadamillas, M.V. Herna required for myeloid cell fusion via regulation of Rac1 signaling, Dev. Cell 18 (2010) 77e89. [14] A. Stacchini, M. Aragno, A. Vallario, et al., MEC1 and MEC2: two new cell lines derived from B-chronic lymphocytic leukaemia in prolymphocytoid transformation, Leuk. Res. 23 (1999) 127e136. [15] J. Vandooren, N. Geurts, E. Martens, et al., Gelatin degradation assay reveals MMP-9 inhibitors and function of O-glycosylated domain, World J. Biol. Chem. 2 (2011) 14e24.
n, et al., Inhibition of MMP-9[16] E. Ugarte-Berzal, J. Vandooren, E. Bailo dependent degradation of gelatin, but not other MMP-9 substrates, by the MMP-9 hemopexin domain blades 1 and 4, J. Biol. Chem. 291 (2016) 11751e11760. [17] A.S. Kamiguti, E.S. Lee, K.J. Till, et al., The role of matrix metalloproteinase 9 in the pathogenesis of chronic lymphocytic leukaemia, Br. J. Haematol. 125 (2004) 128e140. ~ oz, E. Ugarte-Berzal, J.A. García-Marco, et al., a4b1 integrin [18] J. Redondo-Mun and 190 kDa CD44v constitute a cell surface docking complex for gelatinase B/ MMP-9 in chronic leukemic but not in normal B cells, Blood 112 (2008) 169e178. [19] C. Huang, K. Jacobson, M.D. Schaller, MAP kinases and cell migration, J. Cell. Sci. 117 (2004) 4619e4628. [20] M. Milella, I. Falcone, F. Conciatori, et al., PTEN: multiple functions in human malignant tumors, Front. Oncol. 5 (2015) 24. [21] J. Gu, M. Tamura, K.M. Yamada, Tumor suppressor PTEN inhibits integrin- and growth factor-mediated mitogen-activated protein (MAP) kinase signaling pathways, J. Cell. Biol. 143 (1998) 1375e1383. [22] T. Sakamoto, M. Seiki, Integrated functions of membrane-type 1 matrix metalloproteinase in regulating cancer malignancy: beyond a proteinase, Cancer Sci. 108 (2017) 1095e1100. [23] S. D'Alessio, G. Ferrari, K. Cinnante, et al., Tissue inhibitor of metalloproteinases-2 binding to membrane-type 1 matrix metalloproteinase induces MAPK activation and cell growth by a non-proteolytic mechanism,


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129


n et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e7 E. Bailo J. Biol. Chem. 283 (2008) 87e99. [24] H. Mori, A.T. Lo, J.L. Inman, et al., Transmembrane/cytoplasmic, rather than catalytic, domains of Mmp14 signal to MAPK activation and mammary branching morphogenesis via binding to integrin b1, Development 140 (2013) 343e352. [25] A. Dufour, S. Zucker, N.S. Sampson, et al., Role of matrix metalloproteinase-9 dimers in cell migration: design of inhibitory peptides, J. Biol. Chem. 285

7

(2010) 35944e35956. [26] J.L. Binet, A. Auquier, G. Dighiero, et al., A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate analysis, Cancer 48 (1981) 198e206. [27] K.R. Rai, A. Sawitsky, E.P. Cronkite, et al., Clinical staging of chronic lymphocytic leukemia, Blood 46 (1975) 219e234.


n, et al., A catalytically inactive gelatinase B/MMP-9 mutant impairs homing of chronic lymphocytic Please cite this article in press as: E. Bailo leukemia cells by altering migration regulatory pathways, Biochemical and Biophysical Research Communications (2017), https://doi.org/ 10.1016/j.bbrc.2017.10.129