Biomaterials 47 (2015) 29e40

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Reengineering autologous bone grafts with the stem cell activator WNT3A Wei Jing a, 1, 2, Andrew A. Smith a, 1, Bo Liu a, Jingtao Li a, 2, Daniel J. Hunter a, Girija Dhamdhere a, Benjamin Salmon b, c, Jie Jiang d, Du Cheng a, 3, Chelsey A. Johnson a, 4, Serafine Chen a, Katherine Lee a, Gurpreet Singh a, Jill A. Helms a, * a

Division of Plastic and Reconstructive Surgery, Department of Surgery, Stanford School of Medicine, Stanford, CA 94305, USA Dental School University Paris Descartes PRES Sorbonne Paris Cit e, EA 2496, Montrouge, France ^pitaux Universitaires Paris Nord Val de Seine, Paris, France AP-HP Odontology Department Bretonneau, Ho d UCLA Clinical and Translational Science Institute, Los Angeles, CA, USA b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 21 July 2014 Accepted 16 December 2014 Available online 2 February 2015

Autologous bone grafting represents the standard of care for treating bone defects but this biomaterial is unreliable in older patients. The efficacy of an autograft can be traced back to multipotent stem cells residing within the bone graft. Aging attenuates the viability and function of these stem cells, leading to inconsistent rates of bony union. We show that age-related changes in autograft efficacy are caused by a loss in endogenous Wnt signaling. Blocking this endogenous Wnt signal using Dkk1 abrogates autograft efficacy whereas providing a Wnt signal in the form of liposome-reconstituted WNT3A protein (LWNT3A) restores bone forming potential to autografts from aged animals. The bioengineered autograft exhibits significantly better survival in the hosting site. Mesenchymal and skeletal stem cell populations in the autograft are activated by L-WNT3A and mitotic activity and osteogenic differentiation are significantly enhanced. In a spinal fusion model, aged autografts treated with L-WNT3A demonstrate superior bone forming capacity compared to the standard of care. Thus, a brief incubation in L-WNT3A reliably improves autologous bone grafting efficacy, which has the potential to significantly improve patient care in the elderly. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Age/ageing Biomimetic material Bone regeneration Liposome Osteogenesis Spinal surgery

1. Introduction It is generally accepted that as we age, healing potential diminishes. This is especially obvious in the skeleton: compared to adolescent or adult skeletons, the geriatric skeleton is usually osteoporotic [1e3] and co-morbidities such as decreased vascularization, poor metabolism, and accumulated DNA damage contribute to slow bone healing in the elderly. Consequently, there is an increasing demand for biomaterials that take age-related skeletal changes into consideration. The most common treatment for bony non-unions and delayed unions is autologous bone grafting, or autografting. Autografts are a

* Corresponding author. 1 Contributed equally to the manuscript. 2 Current address: State Key Laboratory of Oral Diseases, West China Hospital of Stomatology, Sichuan University, Chengdu 610041, China. 3 Current address: Weill School of Medicine, Cornell University, New York, NY 10065, USA. 4 Current address: University of Arizona College of Medicine, Tucson, AZ, USA. http://dx.doi.org/10.1016/j.biomaterials.2014.12.014 0142-9612/© 2014 Elsevier Ltd. All rights reserved.

heterogenous collection of marrow blood products, connective tissue stroma, bony extracellular matrix, and a variety of hematopoietic, vascular, and osteogenic stem cell populations [4e7]. The physical, biological, and chemical composition of autografts makes them an ideal bone-regenerating biomaterial in young patients [8,9]; in older individuals, however, autografts are unpredictable [10e12]. A number of bone graft substitutes have been developed to address this need [13]. For example, synthetic scaffolds such as ceramics (e.g., tricalcium phosphate, hydroxyapatite) and bioactive glass (silica and calcium oxide) have been fabricated to resemble the micro-porosity and compressive strength of bone [14,15]. While these synthetic materials are generally considered biocompatible, they exhibit no inherent osteogenic activity [16] and cannot adapt to changing physiologic conditions [17]. Cadaveric demineralized bone matrix (DBM) can replace the mineralized component of an autograft [18] and while DBM appears to support osteogenesis [19,20] the material is devoid of viable cells and disease transmission remains a concern [21]. Other engineered bone substitutes include allogeneic stem cell products [22] but whether they are

30

W. Jing et al. / Biomaterials 47 (2015) 29e40

indicate that elevated Wnt signaling induces bone formation [32e34] whereas reduced Wnt signaling induces bone loss [35,36]. We posited that a reduction in Wnt signaling might be responsible for the loss in osteogenic potential of autografts. To counteract the age-related decline in Wnt signaling we supplied a chemical stimulus in the form of a Wnt protein to autografts from aged animals. In previous work we cataloged the distribution of canonical and non-canonical Wnt ligands in the intact and injured skeleton, and this analysis revealed that Wnt3a was most broadly expressed [37]. Further, the expression level of Wnt3a was the most severely affected by aging [38]. Consequently, our study here focused on delivery of WNT3A to autografts from aged animals.

osteogenic still remains a matter of considerable debate [23,24]. None of these bone graft substitutes take into account the changing skeletal properties of the aging patient. If we understood why autografts fail in older patients we might be in a position to improve this standard of care for bone regeneration. We know that the physical properties of autografts change with age: the marrow undergoes fatty degeneration [25] and the mineralized extracellular matrix component of an autograft is significantly reduced because of osteoporotic changes [22]. Aging also impacts the chemical properties of autografts: aged stem cells are less responsive to the growth factor stimuli in their environments [26,27] and accumulating evidence indicates that both local and systemic levels of growth factor stimuli decline in the elderly (reviewed in Ref. [28]). Here, we tested the hypotheses that the osteogenic potential of an autograft is attributable to stem cells in the graft material, and that aging impacts the Wnt responsive status of these stem/progenitor populations. We focused on the role of Wnt signaling in this regard because the pathway is widely recognized as a key regulator of bone mass [29e31]. Experimental and clinical evidence both

femur

illiac crest

A

femur

3

The Stanford Committee on Animal Research approved all procedures. Betaactin-enhanced green fluorescent protein (ACTB-eGFP), and CD1 syngeneic hosts, as well as Axin2CreERT2/þ and R26RmTmG/þ mice were used; the latter were purchased (The Jackson Laboratory, CA). Mice 10 months were considered aged. Aged Lewis rats (“retired breeders”, Charles Rivers, MA), were used for spinal fusion surgeries.

E

C

illiac crest

tibia

BrdU

F

* *

* *

2.5

2.1. Animal care

tibia

B

2 1.5

Runx2

ALP

p=0.4732

0

p=0.00035

0.5

p=0.8086

1 p=0.0190

fold mRNA expression

D

2. Methods

Osteocalcin

G

Runx2

H

Sox9

J

Aniline Blue

K

SafraninO

* * *

I

PPARγ

L

Gomori

*

Fig. 1. Bone graft material contains stem and progenitor cell populations. (A) Gomori staining of BGM harvested from the rat femur, (B) the iliac crest (where trabecular bone chips are indicated with a dotted line), and (C) the tibia. (D) Quantitative RT-PCR analyses of endogenous osteogenic gene expression in BGM from the indicated sources. (E) Schematic of the experimental design, where autologous BGM is harvested from the iliac crest of a rat and transplanted into the SRC. On post-transplant day 7 (F) representative tissue sections were stained to detect BrdU, (G) Runx2, (H) Sox9, and (I) PPARg expression. (J) Representative tissue sections were also stained with Aniline blue to detect osteoid matrix; asterisks indicate new bone matrix as opposed to old bone chips (dotted lines). The surface of the renal capsule is indicated with a dashed white line. (K) Representative tissue sections were stained with Safranin O/Fast green histology to detect proteoglycan-rich cartilage (red), and (L) Gomori trichrome staining to detect adipocytes. Abbreviations: BrdU, bromodeoxyuridine. Scale bars: 50 mm, asterisks: p < 0.05.

Periosteum

Endosteum

B

number of cells (log scale)

A

Bone Marrow Cavity

C

D

0x10

10

1x1011

0x10

Protein expression 1 0.8

* *

0.6 0.4 0.2 0

p=0.0468

young

fold protein expression

F BGM

2x1011

III IV V VI VII VIII sample

**

Wnt3a beta catenin Axin2

p=0.9032

I

p=0.0004

0

p=0.0108

1

BGM aged

p=0.5820

5x108

3x1011

R =0.9593

0x10

1x109

mRNA copies/ng RNA

2x10 7

p=0.1156

4x10 7

5x109

p=0.1146

6x10 7

100

GAPDH

2x109

R =0.8393

8x10

mRNA copies/ng RNA

Lef1

7

R =0.9952

mRNA copies/ng RNA

Axin2 1x10 8

1000

W. Jing et al. / Biomaterials 47 (2015) 29e40

Axin2 CreERT2; R26 mTmG

10000

E Endogenous Wnt signal

BGM DAPI +VE cells GFP +VE cells

beta actin

Fig. 2. Bone graft material is Wnt responsive. (A) GFP immunostaining of tissue sections from Axin2CreERT2/þR26RmTmG mice; GFPþve cells are present in the periosteum and (B) endosteum. (C) Quantification of GFPþve cells/total cells within specified microscopic fields of view. (D) GFPþve cells in the BGM visualized by fluorescence. (E) Quantitative absolute RT-PCR for endogenous Axin2, Lef1, and GAPDH expression in BGMyoung (green bars) and BGMaged (grey bars). (F) Western blot analyses for Wnt3a, total beta catenin, Axin2, and beta actin in BGMyoung (green bars) and BGMaged (grey bars). Scale bars ¼ 50 mm. Asterisks: p < 0.05.

31

32

W. Jing et al. / Biomaterials 47 (2015) 29e40

B

BGM young BGM aged

1.2 1.0 0.8

*

0.2 0

* ALP Osterix Oc

GFP fluor

young

aged

BGM

D

E

kidney

F

5

kidney

4 3 2 1 0

H

I

5

ALP

%ALP +ve pixels

G

K

**

4 3 2 1 0

J

**

p=0.0002

0.4

*

p=0.0013

0.6

BGM

Aniline Blue

C

%AB+ve pixels

fold mRNA expression

A

L

10

5

0

p=0.6075

%GFP+ve pixels

GFP

15

Fig. 3. The osteogenic differentiation potential of BGM declines with age. (A) Quantitative RT-PCR analyses for expression of ALP, Osterix, and Osteocalcin in BGMyoung (green bars) and BGMaged (grey bars). (B) BGM harvested from ACTB-eGFP mice, transplanted into the SRC and visualized under brightfield and (C) fluorescent light to detect the GFP signal in BGM. (D) Representative tissue sections stained with Aniline blue from BGMyoung (N ¼ 5) and (E) BGMaged (N ¼ 5). Dotted line indicates the surface of the renal capsule. (F) Histomorphometric analyses of Aniline blueþve pixels within the total area occupied by the BGM on post-transplant day 7. (G) Representative tissue sections stained to detect ALP activity from BGMyoung (N ¼ 5) and (H) BGMaged (N ¼ 5). (I) Quantification of ALPþve pixels within the total area occupied by the BGM on post-transplant day 7. (J) Representative tissue sections immunostained for GFP from BGMyoung (N ¼ 5) and (K) BGMaged (N ¼ 5). (L) Quantification of GFPþve pixels within the total area occupied by the BGM on posttransplant day 7. Abbreviations: ALP, alkaline phosphatase; Oc, Osteocalcin. Scale bars: 100 mm. Asterisks: p < 0.05; double asterisks: p < 0.01.

W. Jing et al. / Biomaterials 47 (2015) 29e40 2.2. Collection of bone graft material The use of mice allows for a broad spectrum of molecular analyses, however, because autografts are highly invasive for these small animals we used rats when performing autografts (e.g., Figs. 1 and 6), and syngeneic mice when employing advanced molecular techniques (Figs. 2e5). In all cases, bone graft material (BGM) was harvested from femurs, tibiae and iliac crest by splitting the bones lengthwise, gently scraping the endosteal surface with a sharp instrument, and irrigating the marrow contents into a collection dish. This method mimicked the reamer/irrigator/ aspirator (RIA) technique [39,40]. To induce recombination in Axin2CreERT2/þR26RmTmG/þ mice (Fig. 2), animals were given 160 mg/g body weight tamoxifen via intraperitoneal (IP) injection for 5 days. Tissues were harvested 7 days after the first treatment for analyses. To ensure BGM aliquots were equivalent in terms of cellular content, BGM was harvested and divided into 20 mL aliquots. DNA was extracted (DNeasy Tissue Kit,

BGM young + Ad-Fc

33

QIAGEN) and DNA concentration was measured (Quant-iT PicoGreen dsDNA Kit, Invitrogen) and microplate fluorescence reader (BERTHOLD, Bad Wildbad, Germany). The percent variation in DNA content was

Reengineering autologous bone grafts with the stem cell activator WNT3A.

Autologous bone grafting represents the standard of care for treating bone defects but this biomaterial is unreliable in older patients. The efficacy ...
4MB Sizes 0 Downloads 11 Views