TSpace Research Repository tspace.library.utoronto.ca

Monitoring tissue development in acellular matrix-based regeneration for bladder tissue engineering: Multiexponential diffusion and

T2* for improved specificity

Hai-Ling Margaret Cheng, Yasir Loai, Walid A. Farhat

Version Post-print/accepted manuscript

Citation (published version)

Cheng HL, Loai Y, Farhat WA. Monitoring tissue development in acellular matrix‐based regeneration for bladder tissue engineering: Multiexponential diffusion and T2* for improved specificity. NMR in Biomedicine. 2012 Mar;25(3):418-26.

Publisher’s Statement This is the peer reviewed version of the following article: [Cheng HL, Loai Y, Farhat WA. Monitoring tissue development in acellular matrix‐based regeneration for bladder tissue engineering: Multiexponential diffusion and T2* for improved specificity. NMR in Biomedicine. 2012 Mar;25(3):418-26.], which has been published in final form at [10.1002/nbm.1617]. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.

How to cite TSpace items

Always cite the published version, so the author(s) will receive recognition through services that track

citation counts, e.g. Scopus. If you need to cite the page number of the author manuscript from TSpace because you cannot access the published version, then cite the TSpace version in addition to the published

version using the permanent URI (handle) found on the record page.

This article was made openly accessible by U of T Faculty. Please tell us how this access benefits you. Your story matters.

MRI of acellular matrix-based bladder tissue engineering

1

Monitoring tissue development in acellular matrix-based

regeneration for bladder tissue engineering:

multiexponential diffusion and T2* for improved specificity

Hai-Ling Margaret Cheng, PhDa,b*

, Yasir Loai, BScc, Walid A. Farhat, MD

c

a Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada

b Physiology & Experimental Medicine, The Research Institute and Diagnostic

Imaging, The Hospital for Sick Children, Toronto, ON, Canada

c Developmental and Stem Cell Biology, The Research Institute and Division of

Urology, The Hospital for Sick Children, Toronto, ON, Canada

Short title: MRI of acellular matrix-based bladder tissue engineering

Grant sponsor: CIHR – Institute of Human Development Child and Youth Health /

SickKids Foundation (XG 08-017)

* Correspondence to:

Hai-Ling Margaret Cheng, PhD

Department of Diagnostic Imaging

The Hospital for Sick Children

555 University Avenue

Toronto, Ontario M5G 1X8, Canada

(tel) 416-813-5415 (fax) 416-813-7362

E-mail: hai-ling.cheng@sickkids.ca

MRI of acellular matrix-based bladder tissue engineering

2

ABSTRACT

Cell-seeded acellular matrices (ACMs) are a promising approach for tissue-

engineering soft tissues and organs such as the urinary bladder. The ACM contains site-

preferred structural and functional molecules, and degradation products derived from the

ACM play important roles in tissue remodeling. Regeneration proceeds along concurrent

trajectories of cell growth and matrix degradation, characterized by evolving biophysical

and biochemical properties. Assessment of tissue development through a non-invasive

imaging technique such as MRI must, therefore, be capable of distinguishing these

concurrent biophysical and biochemical changes. However, although MRI provides

exquisite sensitivity to tissue microstructure, composition, and function, specificity remains

limited. In this study, multiexponential diffusion and effective transverse relaxation time

T2* were investigated for their ability to assess cell growth and tissue composition,

respectively. Bladder ACMs prepared with and without hyaluronic acid, and ACMs seeded

with smooth muscle cells, were assessed on MRI. The slow diffusion fraction from

multiexponential diffusion analysis demonstrated the best correlation with cellularity, with

minimal influence from underlying matrix degradation. T2* measurements were sensitive

to macromolecular content, specifically, the presence of hyaluronic acid, without

confounding influence from tissue hydration. T2* also appeared sensitive to cell filling of

matrix pore space. Compared to these metrics, commonly used MRI parameters such as T1,

T2, and single diffusion coefficient were more limited in specificity. Use of T2 to measure

tissue structure and composition is limited by its large dependence on water content, and

single diffusion can only reflect the overall characteristics of the extra- and intracellular

MRI of acellular matrix-based bladder tissue engineering

3

environment. These findings are important for further development of more specific MRI

methods for monitoring regeneration in tissue engineered systems.

Keywords: quantitative (MRI) magnetic resonance imaging; multiexponential diffusion;

(T2*) effective transverse relaxation time; (ACM) acellular matrix; cellularity; (GAG)

glycosaminoglycan; tissue hydration; bladder tissue engineering

Abbreviations:

2D, two-dimensional

3D, three-dimensional

ACM, acellular matrix

dH2O, double distilled water

GAG, glycosaminoglycan

HA, hyaluronic acid

HBSS, Hank’s Balanced Salt Solution

H&E, hematoxylin and eosin

VEGF, vascular endothelial growth factor

MRI of acellular matrix-based bladder tissue engineering

4

GRAPHICAL ABSTRACT

Monitoring tissue development in acellular matrix-based regeneration for bladder tissue

engineering: multiexponential diffusion and T2* for improved specificity

Hai-Ling Margaret Cheng, PhD*, Yasir Loai, BSc, Walid A. Farhat, MD

Multiexponential diffusion MRI and T2* are investigated for monitoring acellular matrix-

based bladder regeneration. The slow diffusion fraction was shown to correlate with

cellularity in the presence of extracellular matrix changes, while T2* was sensitive to

macromolecules without influence from tissue hydration. These metrics provide improved

specificity over single T1, T2, and diffusion for assessing concurrent tissue engineering

processes.

I II III IV

0

0.1

0.2

T2* T2 T1

Higher GAG and water content

Slow diffusion fraction vs. cellularity

MRI of acellular matrix-based bladder tissue engineering

5

INTRODUCTION

The urinary bladder and bladder disorders

The primary function of the urinary bladder is to store urine at low pressure and

allow voluntary voiding. These functions are regulated by nerves in the spine and around

the bladder and depend on the unique composition and structure of the bladder wall. In a

normal bladder, the wall is composed primarily of three layers: the urothelium, the lamina

propria, and the detrusor muscle. The innermost urothelium consists of highly specialized

cells that provide a barrier to urine penetration into underlying tissues and bloodstream,

while the lamina propria is a connective tissue matrix that limits bladder wall distension

and overall compliance. The detrusor layer consists of smooth muscle cell bundles that are

coordinated by neural control to relax and contract during the filling-voiding cycle, thus

maintaining low pressure and protecting the kidneys from damage.

Bladder dysfunction can result from a variety of congenital and acquired conditions

and compromises quality of life for approximately 400 million people worldwide (1). In

children, the causes are often congenital and include bladder exstrophy (1 in 40,000),

urethral defects (1 in 500) leading to obstructed urine flow, and neuropathic conditions such

as myelomeningocele (1 in 800) that alter neuronal control of the bladder. Loss of proper

bladder control, with subsequent changes in function, can arise from other disruptions to

the neural pathways between the bladder and the micturition center in the brain. In adults,

for example, neurogenic bladders are often induced from trauma to the spinal cord or from

tumors. Obstructive uropathy can also present in adults for a variety of reasons, including

stone formation, enlarged prostates, or tumors. Regardless of the cause of dysfunction, the

MRI of acellular matrix-based bladder tissue engineering

6

structure, thickness, and biomechanics of the bladder wall can be significantly altered. The

result is a high-pressure, low-compliant bladder leading to incontinence and, in severe cases

where pressure extends up the ureters, swelling and impaired function of the kidneys.

The current treatment strategy for bladder dysfunction uses gastrointestinal

segments for replacing bladder tissue (2). This approach can be problematic because

gastrointestinal tissue absorbs solutes that urinary tissue excretes. Due to this difference in

tissue function, several complications can arise, such as infection, metabolic disturbances,

urolithiasis, perforation, and malignancy (2). Alternative methods and materials have been

investigated, most notably tissue expansion (3) and free grafts made from natural (e.g.

small intestinal submucosa, bladder submucosa) or synthetic (e.g. gelatin sponge, polyvinyl

sponge) materials (4,5). However, these attempts have met with limited success due to

biocompatibility, mechanical, structural, and functional problems. Thus, the use of

gastrointestinal segments has remained the gold standard despite its limitations.

Tissue engineering for bladder replacement

Alternative treatments for bladder dysfunction involving tissue engineering

techniques have been pursued in an effort to overcome limitations associated with the

conventional use of gastrointestines. Promising results have been reported in patients using

autologous cells seeded on a collagen-polymer scaffold (6) and in dogs using cell-seeded

allogeneic acellular bladder matrices (7). Prompt angiogenesis of thick tissues is critical to

maintain viability and support continued growth and long-term survival (6,8). Cell-seeding

also is required to ensure tissue regeneration beyond a distance of 0.5 cm (9) necessary for

organ engineering. Compared to the unseeded scaffold, cell seeding reduces fibrosis and

MRI of acellular matrix-based bladder tissue engineering

7

graft contracture and improves overall cellular organization (7). Another requirement is the

use of a scaffold to guide 3D tissue formation. Both synthetic materials and biological

matrices have been used. However, synthetic materials usually succumb to mechanical

failure (6), and biological matrices are believed better able to reproduce the required

biomechanics and also offer intrinsic biochemical characteristics that are difficult to

achieve with synthetic materials. A promising biological scaffold material is the acellular

matrix (ACM) derived from the extracellular matrix of the desired tissue type. The ACM

has cellular components removed to reduce immunogenicity (10) and contains site-

preferred structural and functional molecules for reconstructing tissue. Degradation of the

ACM yields products that have the required angiogenic (11), chemotactic (11,12), and

mitogenic (13) activities to facilitate tissue remodeling. Specifically, these products contain

chemotactic and mitogenic factors that increase recruitment and proliferation of

undifferentiated cells and inhibit the proliferation of differentiated cells. ACMs have been

derived from different tissues, such as skin (14), liver (15), heart valves (16), and bladder

(17), differing in their 3D structure but sharing very similar biochemical composition (13).

Despite early promise, much remains to be accomplished toward engineering a fully

functional bladder. Obstacles include lack of prompt angiogenesis, adverse host-tissue

reaction, inadequate mechanical properties, and lack of innervation. Angiogenesis must be

established promptly for cell expansion and to avoid scarring, but it is currently very

difficult to achieve functional vessels that are long-lasting (18). Another interference with

regeneration is adverse host-tissue reaction, such as that arising from immunogenic

response of cell components or nondegraded scaffold biomaterials. This reaction can be

mitigated with the use of ACM and scaffold manipulation to prevent urine penetration and

MRI of acellular matrix-based bladder tissue engineering

8

its toxic effects (19). Achieving proper biomechanics presents even greater challenges.

Evidence exists that in-vitro mechanical conditioning to simulate the in-vivo environment

can improve tissue morphogenesis, proliferation, differentiation, and function (20,21,22).

However, the type and degree of physical stimuli required and their role in the bladder

remain to be determined. Finally, innervation is an important component to bladder

function but is the least studied and understood. Strategies for nerve regeneration from

other applications perhaps may be applied, such as a recent demonstration of innervating

muscular tissue by seeding primary cultured neurons on a biological extracellular matrix

(23).

Imaging of the tissue engineered bladder

Technology for non-invasive assessment such as MRI is recognized to enable

accelerated progress in tissue engineering and eventual longitudinal monitoring of patients

with tissue engineered organs. Current regeneration approaches need to look beyond

standard histology for a more complete assessment of the function, structure, and

composition of the developing tissue throughout its entire lifespan, from in-vitro cell

seeding to long-term functioning in vivo, in order to identify more effective strategies. In

the bladder, important parameters include cell growth, distribution, viability, and host-

implant interaction in addition to ones more specific to large soft-tissue organs, particularly

those with a mechanical function. These more specific parameters include biomechanics,

functioning of an induced vascular supply, properties of new natural scaffolds for soft-

tissue regeneration, and cell-scaffold interaction.

MRI studies of the tissue engineered bladder have been few but have focused on

MRI of acellular matrix-based bladder tissue engineering

9

several different aspects of regeneration. The earliest efforts addressed the importance of

angiogenesis, investigating the potency of vascular endothelial growth factor (VEGF) and

the ability of MRI to quantify neovessel formation (24,25,26). MRI measures of vascularity

were shown to correlate with increased blood volume at higher VEGF doses (24), and

quantification on an absolute scale was feasible using a blood-pool contrast agent (25).

These studies also demonstrated that enhanced vascularization reduced fibrosis and

improved overall cell repopulation in vivo (26).

The ACM scaffold has been the other main focus because of its role in regeneration,

specifically, in providing an environment conducive to angiogenesis, tissue deposition, and

restoration of structure and function specific to the grafted site (27). MRI characterization

of the ACM has been lacking until recently but is important for establishing a baseline

against which to gauge subsequent tissue development. In a couple of recent MRI studies of

the bladder ACM (28,29), new imaging observations revealed unanticipated biological

phenomena that may be relevant in other ACM-based systems. One important finding was

matrix collagen degradation noted very early on after cell-seeding, possibly due to easier

breakdown of biological tissue compared to synthetic materials (28). Degradation

byproducts are known to modulate tissue remodeling (13), and matrix degradation

concurrent with cell growth is expected in many regeneration paradigms. Distinction of

these concurrent processes using more specific MRI techniques, including multicomponent

T2 as demonstrated in Ref (28), will be critical to accurate assessment. Another valuable

insight is that scaffold optimization from a single manipulation can produce concurrent

biophysical and biochemical changes that have unpredictable effects on MRI parameters.

Specifically, the incorporation of hyaluronic acid, a naturally occurring macromolecule that

MRI of acellular matrix-based bladder tissue engineering

10

can act as a carrier of bioactive molecules and growth factors and has shown potential in

the development of engineered tissues (30,31), was shown to increase macromolecular

content and ACM hydration (29). These concurrent changes exerted competing effects on

MRI tissue relaxation and diffusion, and single MRI parameters failed to provide adequate

specificity. These studies (28,29) highlight the ability of MRI to improve our understanding

of uncharted development of engineered tissues by detecting changes for further

investigation and validation. They also emphasize the need for more specific MRI methods

to assess tissue structure, composition, and function.

In this study, we explore MRI methods for improved specificity in monitoring tissue

development in a bladder ACM-based regeneration model. Multiexponential diffusion and

effective transverse relaxation time, T2*, are investigated for their ability to discriminate

cellularization and matrix changes without influence from concurrent or undesired sources.

EXPERIMENTAL

Acellular matrix (ACM) preparation

Acellular tissue matrices were prepared according to a published protocol (10).

Fresh urinary bladders were harvested from pigs (20-50 kg) and immediately washed in

sterile phosphate buffer saline (Sigma), longitudinally sectioned, and placed in a hypotonic

solution [5 mM EDTA , 10 mM Tris HCl, pH 8.0, 1% Triton X-100, 0.1 mg/mL Pefabloc

PlusTM

(Alexis) and Penicillin/Streptomycin] (Sigma) for 72 hours, stirring at 4˚C, to lyse

all cell structures and inhibit proteases. Bladder tissues were then immersed in hypertonic

solution [5 mM EDTA, 10 mM Tris HCl, pH 8.0, 1% Triton X-100 and 1.5 M KCl]

MRI of acellular matrix-based bladder tissue engineering

11

(Sigma), stirring for an additional 72 hours at 4˚C to denature residual proteins.

Subsequently, bladders were washed in Hank’s Balanced Salt Solution (HBBS), followed

by overnight incubation at 37˚C with Benzonase (2U/mL) diluted in HBSS to degrade

DNA and RNA components. Final extraction was performed by transferring bladders to a

0.25% CHAPS (non-denaturing detergent) based solution [50 mM Tris HCl, pH 8.0, 1%

Triton X-100 and Penicillin/Streptomycin]. The resulting ACM was washed with sterile

double distilled water (dH2O) and stored in 70% ethanol. Hematoxylin and eosin (H&E)

sections were taken to confirm acellularity. All protocols were approved by the institutional

Animal Care Committee and compliant with national policies on the humane use of

laboratory animals (CCAC guidelines).

Preparation of ACM and cell-seeded constructs

ACMs of 3 mm thickness were cut into 22 cm2 squares, dehydrated by gradual

immersion in increasing concentrations of ethanol (50%, 70%, 80%, 90%, 100%) for 1

hour each at room temperature, and lyophilized for 24 hours (VirTis-temp, -70˚C and

vacuum, 120 millitorr). Hyaluronic acid (HA) was incorporated by rehydrating ACMs in

increasing concentrations of HA (0.05, 0.1, 0.2 mg/100 mL) (Sigma) shaking for one hour,

then in 0.5 mg/100 mL HA for 24 hours, all at 37˚C. Excess HA was washed off with

dH2O. Alcian blue staining was performed to confirm HA uptake (Figure 1). ACMs with

and without HA were immersed in medium (high glucose Dulbecco’s modified essential

medium (Wisent Inc), 10% fetal bovine serum) at 4°C.

For cell-seeding, cell cultures were obtained from passage 2 smooth muscle cells

isolated from porcine bladder (32) and expanded in medium. Cells were seeded on ACMs

MRI of acellular matrix-based bladder tissue engineering

12

in a 6 well plate at 1106

cells/cm2

in 100 L of medium. The cells were allowed to attach

on the ACM for 3 hrs (37°C in a humidified atmosphere of 5% CO2) before incubating in 2

mL of medium. The medium was changed every 72 hours.

Experimental design

Cell-seeded (Ns=20) and unseeded (Nu=15) ACMs were prepared as described

above. Six of the unseeded ACMs were further processed for HA incorporation. Imaging

was performed on day 1, 3, and 7 and, for seeded ACMs, at an additional timepoint of 21

days after cell-seeding. Samples were prepared for MRI by transferring each ACM to a 5

mL, 12 mm diameter round Falcon tube containing medium, and placing the tubes in a

water-filled container. Histology was also performed, and samples were taken prior to

imaging to verify no changes occurred during MRI. Additional validation measurements

involving immunofluorescence microscropy and biochemical assays from previous

experiments are reported here to provide a benchmark.

Histology

All tissue samples were fixed in 10% formalin for 24 hours, embedded in paraffin,

and sectioned into 5 m thick slices. Sections were stained with H&E and Masson’s

Trichrome and examined under a light microscope (Nikon Eclipse E 400). Cell density was

assessed qualitatively at all time-points on H&E sections. Four different classes of

cellularity were defined: I (very few cells), II (few cells), III (patches of cells), and IV

(good cellularity). A blinded histology assessment was performed by a pathologist.

MRI of acellular matrix-based bladder tissue engineering

13

Quantitative MRI

MRI was performed on a 1.5 Tesla clinical scanner (Signa EXCITE TwinSpeed, GE

Healthcare, Milwaukee, WI) using a single 3-inch surface coil positioned under a water-

filled container in which ACM-containing tubes were placed. Pilot scans were taken to

determine placement of imaging slices perpendicular to the tubes and encompassing the

upper and lower ends of the ACMs.

Quantitative T1, T2, T2*, and diffusion imaging was performed. The longitudinal

relaxation time T1 was measured using a rapid 3D technique (33) based on a fast spoiled

gradient echo sequence with radio-frequency field correction [flip angle = 2, 3, 10, 20,

number of averages (NAVG) = 4]. The transverse relaxation time T2 was measured using a

2D spin echo sequence [9 echoes with echo times (TE) = 9 - 300 ms, repetition time (TR) =

3000 ms, NAVG = 1]. The effective transverse relaxation time T2* was measured using a 2D

multi-echo gradient-echo sequence [16 equally spaced TEs = [2.6 – 34.4] ms, TR = 118 ms,

FA = 20, NAVG = 4]. Diffusion imaging was performed using a spin echo diffusion-

weighted imaging sequence up to high b-values [b = 0 – 3000 s/mm2

in steps of 200 s/mm2,

TR = 4000 ms, NAVG = 16]. In addition to high b-value acquisition for multiexponential

diffusion analysis, single diffusion coefficients were also obtained by imaging at a b-value

of 1000 s/mm2. Slice thickness was 3 mm for all sequences; in-plane resolutions were 0.4

mm (T1, T2, T2*) and 0.8 mm for diffusion.

All image post-processing of quantitative T1, T2, and T2* data was performed on a

pixel-wise basis using in-house scripts developed in Matlab (V.7.0, Mathworks Inc.,

Natick, MA, USA). T1 parameter maps were generated as previously described, with

MRI of acellular matrix-based bladder tissue engineering

14

analytical-based flip angle correction using B1 field maps acquired separately (33). Both

T2 and T2* parameter maps were computed by fitting a mono-exponential signal decay

function to signal intensities obtained at various echo times (34). The mean T1, T2, or T2*

value for each ACM was then obtained by averaging within the ACM and through all

imaging slices through which the ACM traversed. For diffusion data, a region

encompassing the ACM was drawn and the average signal intensity at each b-value was

taken for analysis. Non-linear least squares analysis, which requires no a priori knowledge

of the number of the exponentials, was used to estimate diffusion coefficients and their

fractions. All data are expressed as mean value ± standard deviation.

Statistical analysis

Differences between groups were assessed using a two-tailed Student’s t-test.

Significance was declared at P < 0.05.

RESULTS

Histology

Cellularity varied over the time interval of 1 to 21 days post-cell seeding, with cells

aggregated mainly at the surface of the ACM. Representative H&E sections showing the

degree of cellularity associated with each of the four classifications are given in Figure 2.

No difference in cellularity was observed in samples taken before and after MRI from the

same ACM, indicating that the time spent for MRI did not affect cell viability.

Masson’s Trichome sections (Figure 3) illustrate the main constituents of the ACM:

smooth muscle bundles within a collagen extracellular matrix. Histologically, the ACM

MRI of acellular matrix-based bladder tissue engineering

15

appears consistent amongst the different preparations: cell-seeding and HA incorporation.

However, it is known from previous immunohistochemistry and biochemistry experiments

that cell-seeding results in matrix collagen degradation from cell-released collagenase (28),

and HA incorporation increases both macromolecular content and tissue hydration (29).

Table 1 summarizes these biochemical changes. These concurrent changes present

conflicting influences on single MRI parameters such as T1 and T2 and thereby limit their

specificity. For example, increased T1 and, to a much greater extent, increased T2 from

hydration can confound our assessment of cell growth or matrix glycosaminoglycan (GAG)

content. The T2 spectrum was previously shown to distinguish these concurrent changes

(Figure 4), but its assessment of biophysical boundaries (such as cells) would be limited

due to its sensitivity to biochemical composition.

Quantitative MRI

Examples MR images of the construct are shown in Figure 5. From high b-value

diffusion imaging, sensitivity to presence of cells in the ACM is shown. Figure 6 shows

that cell-seeded and unseeded ACMs exhibit biexponential and monoexponential diffusion

decay, respectively. This behaviour is consistent with a fast diffusion component associated

with the extracellular compartment and a slow diffusion component associated with the

intracellular space. Figure 7 shows that the slow diffusion fraction has the highest

correlation with cellularity (r = 0.954, P < 10-5

) compared to single diffusion coefficient (r

= -0.153, P = 0.654), T1 (r = -0.614, P < 0.05), and T2 (r = -0.309, P = 0.355).

New observations on T2*, which is predominantly used in tissue iron and oxygen

measurements, suggest its potential role as a more specific parameter for assessment of

MRI of acellular matrix-based bladder tissue engineering

16

tissue composition. Figure 8 shows that cell-seeding produces the greatest relative change

on T2* (P < 10-4

) compared to T2 (P < 0.05) or T1 (P = 0.11), although the correlation

between T2* and cellularity was not significant (r = 0.152, P = 0.676). Figure 9 shows that

the presence of the macromolecule HA in the unseeded ACM produces significant changes

on T2*, T2, and T1 (P < 0.05), but the largest relative change is seen on T2* and appears to

be due to reduced sensitivity to tissue hydration. Note that while both T1 and T2 changed in

the same direction with cell-seeding or incorporation of HA, only T2* changed in opposing

directions with these two manipulations.

DISCUSSION

Improved specificity in MRI measurements was investigated in this study for

monitoring tissue development in a bladder ACM-based paradigm for soft-tissue

regeneration. The ACM as a biological scaffold is a relatively new approach for guiding 3D

tissue formation and is especially promising for whole organ tissue engineering, especially

those with a mechanical function. Aside from reducing an immunogenic response, the

ACM possesses the required structural and mechanical properties and its degradation

releases by-products that facilitate tissue remodeling. This approach may provide truly off-

the-shelf replacement tissue if animal-derived ACMs can be used in humans. For instance,

given the similar composition and biomechanical properties between porcine and human

ACMs (35), some investigators have taken the next step and seeded human bladder cells on

porcine ACM (36). Nonetheless, experience with biological matrices is in early stages, and

there are few related imaging studies. The first MRI studies of ACM-based soft-tissue

regeneration reported early matrix degradation (28), which is not seen or probably arises

MRI of acellular matrix-based bladder tissue engineering

17

much later in conventional scaffold materials. This concurrency of matrix degradation and

cell growth can confound assessment of either process if common MRI parameters such as

T1, T2, and single diffusion coefficient are used. Also, tissue hydration can vary

substantially with matrix manipulation (29), likely much more so than in synthetic polymer

scaffolds. Because of these properties, the ACM may require more specific MRI techniques

to distinguish relevant processes amidst less important concurrent changes, such as those

related to tissue water content.

For the assessment for cell growth, this study shows that multiexponential diffusion

analysis may provide the best specificity by minimizing sensitivity to other changes, such

as matrix degradation or tissue hydration. These other changes exert an appreciable effect

on more commonly used parameters in tissue engineered systems, such as T1, T2, and

single diffusion coefficient (37), and underlie the weak correlation between these

parameters and cellularity as observed in this study. Amongst these three parameters, only

T1 exhibits a trend (decrease) consistent with cell growth. The other two parameters, T2

and diffusion, are more unpredictable probably due to their greater sensitivity to water

content and changes in the extracellular matrix structure. In contrast, the slow diffusion

component in multiexponential diffusion decay is sensitive to the intracellular compartment

and quite insensitive to changes in the extracellular environment. It also provides a more

specific technique to measure cellularity without confounders from tissue biochemistry.

New properties on the effective transverse relaxation time T2* were observed in this

study and suggest improved specificity over T2 measurements. One consistent observation

was that unlike T2, T2* was relatively insensitive to water content. This is best appreciated

by comparing ACMs with and without HA (Figure 9). In principle, macromolecules

MRI of acellular matrix-based bladder tissue engineering

18

including proteins induce a lowering in T2 and, therefore, T2* compared to free water. This

is due to more restricted water in the hydration layer on the surface of the macromolecules.

However, being hydrophilic in nature, HA also elevates tissue hydration and, thus, free

water content, therefore increasing T2. The competing T2 effects from a higher GAG

content and increased hydration resulted in a net increase, whereas T2* remained low. The

only plausible explanation is that T2* is much less sensitive to water content, likely

because an increase in free water does not introduce local centers of magnetic non-

uniformities; thus, T2* dephasing is not affected. This reduced sensitivity of T2* to

hydration was speculated in a recent study of chondrocyte transplantation in patients (38),

but histological validation was lacking. In this study, biochemistry validation confirms that

T2* is much more specific to macromolecular content than T2 or T1. A second observation

in this study was a substantially larger T2* in cell-seeded relative to unseeded ACMs

(Figure 8). This result is more difficult to interpret. A possible explanation is that cell-

seeding, which is known to fill pores of the ACM, reduces T2* dephasing associated with

pore interfaces. This result is particularly relevant in tissue engineering applications,

because scaffolds are designed to be porous for the purpose of embedding cells. The ability

to identify when cells have occupied these pores is very valuable for determining how

firmly entrenched cells are.

While this study has suggested new potential in tissue engineering applications for

multiexponential diffusion and T2* measurements, our results are consistent with MRI

studies in other applications that attempt to elucidate underlying biological mechanisms.

For example, diffusion studies of perfused breast cancer cells in vitro showed that a second

slow diffusion component associated with restricted intracellular diffusion arose when cells

MRI of acellular matrix-based bladder tissue engineering

19

were added to a gel medium (39). Other studies have shown a correlation between the slow

diffusion fraction and intracellular space in cell suspensions (40,41). However, caution

must be taken in interpreting the slow diffusion fraction in terms of absolute cellular

volume. The biexponential diffusion model is valid only when there is no exchange

between intra- and extracellular water compartments and no restriction of intracellular

water by cell membranes. When these conditions are unmet, diffusion decay deviates from

a pure biexponential, and the diffusion fractions do not correspond to the actual sizes of the

two pools (42,43,44). In order to achieve the ideal situation, water exchange must be very

slow, such as when there are short diffusion times, slower diffusion rates, large cells, or low

membrane permeability. Nonetheless, the slow diffusion fraction provides a useful metric

to estimate cellularity even if not on an absolute scale.

Our T2* findings are more challenging to interpret, as there are relatively few

literature comparisons. Nonetheless, our hypothesis that increased T2* is due to cell filling

of matrix pores, decreased T2* to higher macromolecular content, and relative insensitivity

of T2* to hydration are plausible given the known behaviour of T2* decay. The association

with pores, however, requires more explanation. Materials with pores larger than 15Ǻ

exhibit increased transverse relaxation due to the ability of large pores to take up water

(45). Scaffolds contain much larger pores, in the microns range, that are expected to contain

free water and give rise to susceptibility differences at pore interfaces. This difference leads

to T2* dephasing, the magnitude of which is likely lessened when pores are replaced with

cells.

There are a number of different avenues for future investigation, and new

challenges will present themselves as the regeneration strategy becomes increasingly

MRI of acellular matrix-based bladder tissue engineering

20

complex. For instance, matrix- and cell-related changes, which can exert opposing or

constructive influences, may need to be distinguished. Also, other essential layers of the

bladder such as the urothelium and lamina propria need to be formed, and it remains to be

determined if cells for these layers can be recruited in vivo and not require direct seeding as

does the smooth muscle layer. To identify these multiple signatures will be a challenge,

likely to require finding matrix- or cell-type-specific molecular signatures that can be

detected on MRI. Improvement can also be made in assessing cell repopulation of the

construct. First, the multiexponential diffusion method for measuring cellularity would

benefit from imaging at higher field strengths, which allows shorter diffusion times and

approach to the slow exchange limit. Measuring the number of viable cells would be even

more informative, and this may be achieved through 1H spectroscopy, where a linear

correlation between cell number and the choline peak has been previously observed

(46,47). Quantitative validation of cell number counts (e.g. DNA quantification) will be

undertaken in future studies. Further investigation is also needed to understand the T2*

changes observed in this study. Our hypothesis that T2* may identify cell filling of pore

space should be evaluated in polymer scaffolds where pore size and number can be

systematically controlled. Another finding related to T2*, namely, its insensitivity to tissue

hydration that was also observed in chondrocyte transplantation (38), is potentially valuable

in a wide variety of applications. For example, inflammation, which often results from

tissue construct-host interaction, should be investigated in the future to determine if T2* is

sensitive to infiltration of inflammatory bodies but not to accompanying edema.

In conclusion, this study has demonstrated the potential role of multiexponential

diffusion and T2* for more specific assessment of cell growth and tissue development. The

MRI of acellular matrix-based bladder tissue engineering

21

results of this study are relevant for regeneration paradigms based on biological scaffolds

such as the acellular matrix for engineering soft tissue and large organs. They also extend to

other applications where concurrent biological changes exist, for example, where cellularity

needs to determined independent of biochemical changes or where tissue hydration is a

confounder and needs to be removed.

MRI of acellular matrix-based bladder tissue engineering

22

REFERENCES

1 Korossis S, Bolland F, Ingham E, Fisher J, Kearney J, Southgate J. Review: tissue

engineering of the urinary bladder: considering structure-function relationships and

the role of mechanotransduction. Tissue Eng. 2006; 12: 635-644.

2 Atala A. Bioengineered Tissues for urogenital repair in children. Ped. Res. 2008; 63:

569-575.

3 Satar N, Yoo JJ, Atala A. Progressive dilation for bladder tissue expansion. J. Urol.

1999; 162: 829-831.

4 Vaught JD, Kropp BP, Sawyer BD, Rippy MK, Badylak SF, Shannon HE, Thor KB.

Detrusor regeneration in the rat using porcine small intestinal submucosal grafts:

functional innervation and receptor expression. J. Urol. 1996; 155: 374-378.

5 Kudish HG. The use of polyvinyl sponge for experimental cystoplasty. J. Urol. 1957;

78: 232-235.

6 Atala A, Bauer SB, Soker S, Yoo JJ, Retik AB. Tissue-engineered autologous

bladders for patients needing cystoplasty. Lancet 2006; 367: 1241-1246.

7 Oberpenning F, Meng J, Yoo JJ, Atala A. De novo reconstitution of a functional

mammalian urinary bladder by tissue engineering. Nat. Biotechnol. 1999; 17: 149-

155.

8 Mooney DJ, Mikos AG. Growing new organs. Sci. Am. 1999; 280: 60-65.

9 Dorin RP, Pohl HG, De Filippo RE, Yoo JJ, Atala A. Tubularized urethral

replacement with unseeded matrices: what is the maximum distance for normal tissue

regeneration? World J. Urol. 2008; 26: 323-326.

MRI of acellular matrix-based bladder tissue engineering

23

10 Brown AL, Farhat W, Merguerian PA, Wilson GJ, Khoury AE, Woodhouse KA. 22

week assessment of bladder acellular matrix as a bladder augmentation material in a

porcine model. Biomaterials 2002; 23: 2179-2190.

11 Li F, Li W, Johnson S, Ingram D, Yoder M, Badylak S. Low-molecular-weight

peptides derived from extracellular matrix as chemoattractants for primary endothelial

cells. Endothelium 2004; 11: 199-206.

12 Zantop T, Gilbert TW, Yoder MC, Badylak SF. Extracellular matrix scaffolds attract

bone marrow derived cells in a mouse model of Achilles tendon reconstruction. J.

Orthop. Res. 2006; 24: 1299-1309.

13 Reing JE, Zhang L, Myers-Irvin J, Cordero KE, Freytes DO, Heber-Katz E,

Bedelbaeva K, McIntosh D, Dewilde A, Braunhut SJ, Badylak SF. Degradation

products of extracellular matrix affect cell migration and proliferation. Tissue Eng.

Part A 2009; 15: 605-614.

14 Armour AD, Fish JS, Woodhouse KA, Semple JL. A comparison of human and

porcine acellularized dermis: interactions with human fibroblasts in vitro. Plast.

Reconstr. Surg. 2006; 117: 845-856.

15 Brown B, Lindberg K, Reing J, Stolz DB, Badylak SF. The basement membrane

component of biologic scaffolds derived from extracellular matrix. Tissue Eng. 2006;

12: 519-526.

16 Lichtenberg A, Tudorache I, Cebotari S, Suprunov M, Tudorache G, Goerler H, Park

JK, Hilfiker-Kleiner D, Ringes-Lichtenberg S, Karck M, Brandes G, Hilfiker A,

MRI of acellular matrix-based bladder tissue engineering

24

Haverich A. Preclinical testing of tissue-engineered heart valves re-endothelialized

under simulated physiological conditions. Circulation 2006; 114: I559-I565.

17 Chen F, Yoo JJ, Atala A. Acellular collagen matrix as a possible ‘‘off the shelf ’’

biomaterial for urethral repair. Urology 1999; 54: 407-410.

18 Jain RK. Molecular regulation of vessel maturation. Nat. Med. 2003; 9: 685-693.

19 Cartwright LM, Shou Z, Yeger H, Farhat WA. Porcine bladder acellular matrix

porosity: impact of hyaluronic acid and lyophilization. J. Biomed. Mater. Res. A

2006; 77: 180-184.

20 Ingber DE, Dike L, Hansen L, Karp S, Liley H, Maniotis A, McNamee H, Mooney D,

Plopper G, Sims J, et al. Cellular tensegrity: exploring how mechanical changes in the

cytoskeleton regulate cell growth, migration, and tissue pattern during

morphogenesis. Int. Rev. Cytol. 1994; 150: 173-224.

21 Oluwole BO, Du W, Mills I, Sumpio BE. Gene regulation by mechanical forces.

Endothelium 1997; 5: 85-93.

22 Grinnell F. Fibroblasts, myofibroblasts, and wound contraction. J. Cell Biol. 1994;

124: 401-404.

23 Agrawal V, Brown BN, Beattie AJ, Gilbert TW, Badylak SF. Evidence of innervation

following extracellular matrix scaffold-mediated remodelling of muscular tissues. J.

Tissue Eng. Regen. Med. 2009; 3: 590-600.

MRI of acellular matrix-based bladder tissue engineering

25

24 Cheng HL, Chen J, Babyn PS, Farhat WA. Dynamic Gd-DTPA enhanced MRI as a

surrogate marker of angiogenesis in tissue-engineered bladder constructs: a feasibility

study in rabbits. J. Magn. Reson. Imaging 2005; 21: 415-423.

25 Cheng HL, Wallis C, Shou Z, Farhat WA. Quantifying angiogenesis in VEGF-

enhanced tissue-engineered bladder constructs by dynamic contrast-enhanced MRI

using contrast agents of different molecular weights. J. Magn. Reson. Imaging 2007;

25: 137-145.

26 Cartwright L, Farhat WA, Sherman C, Chen J, Babyn P, Yeger H, Cheng HL.

Dynamic contrast-enhanced MRI to quantify VEGF-enhanced tissue-engineered

bladder graft neovascularization: pilot study. J. Biomed. Mater. Res. A 2006; 77: 390-

395.

27 Hodde J. Naturally occurring scaffolds for soft tissue repair and regeneration. Tissue

Eng. 2002; 8: 295-308.

28 Cheng HL, Islam SS, Loai Y, Antoon R, Beaumont M, Farhat WA. Quantitative MRI

assessment of matrix development in cell-seeded natural urinary bladder smooth

muscle tissue-engineered constructs. Tissue Eng. Part C Methods 2010 Apr 13 [Epub

ahead of print]

29 Cheng HL, Islam SS, Loai Y, Antoon R, Beaumont M, Farhat WA. The acellular

matrix (ACM) for bladder tissue-engineering: a quantitative magnetic resonance

imaging study. Magn. Reson. Med. 2010 May 25 [Epub ahead of print]

MRI of acellular matrix-based bladder tissue engineering

26

30 Chung C, Erickson IE, Mauck RL, Burdick JA. Differential behavior of auricular and

articular chondrocytes in hyaluronic acid hydrogels. Tissue Eng. Part A 2008;

14:.1121-1131.

31 Zavan B, Vindigni V, Lepidi S, Iacopetti I, Avruscio G, Abatangelo G, Cortivo R.

Neoarteries grown in vivo using a tissue-engineered hyaluronan-based scaffold.

FASEB J. 2008; 22: 2853-2861.

32 Farhat WA, Chen J, Sherman C, Cartwright L, Bahoric A, Yeger H. Impact of fibrin

glue and urinary bladder cell spraying on the in-vivo acellular matrix cellularization: a

porcine pilot study. Can. J. Urol. 2006; 13: 3000-3008.

33 Cheng HL, Wright GA. Rapid high-resolution T(1) mapping by variable flip angles:

accurate and precise measurements in the presence of radiofrequency field

inhomogeneity. Magn. Reson. Med. 2006; 55: 566-574.

34 Beaumont M, Odame I, Babyn PS, Vidarsson L, Kirby-Allen M, Cheng HL. Accurate

liver T2 measurement of iron overload: a simulations investigation and in vivo study.

J. Magn. Reson. Imaging 2009; 30: 313-320.

35 Dahms SE, Piechota HJ, Dahiya R, Lue TF, Tanagho EA. Composition and

biomechanical properties of the bladder acellular matrix graft: comparative analysis

in rat, pig and human. Br. J. Urol. 1998; 82: 411-419.

36 Ram-Liebig G, Ravens U, Balana B, Haase M, Baretton G, Wirth MP. New

approaches in the modulation of bladder smooth muscle cells on viable detrusor

constructs. World J. Urol. 2006; 24: 429-437.

MRI of acellular matrix-based bladder tissue engineering

27

37 Peptan IA, Hong L, Xu H, Magin RL. MR assessment of osteogenic differentiation in

tissue-engineered constructs. Tissue Eng. 2006; 12: 843-851.

38 Welsch GH, Trattnig S, Hughes T, Quirbach S, Olk A, Blanke M, Marlovits S,

Mamisch TC. T2 and T2* mapping in patients after matrix-associated autologous

chondrocyte transplantation: initial results on clinical use with 3.0-tesla MRI. Eur.

Radiol. 2010; 20: 1515-1523.

39 Van Zijl PC, Moonen CT, Faustino P, Pekar J, Kaplan O, Cohen JS. Complete

separation of intracellular and extracellular information in NMR spectra of perfused

cells by diffusion-weighted spectroscopy. Proc. Natl. Acad. Sci. U S A 1991; 88:

3228-3232.

40 Thelwall PE, Grant SC, Stanisz GJ, Blackband SJ. Human erythrocyte ghosts:

Exploring the origins of multiexponential water diffusion in a model biological tissue

with magnetic resonance. Magn. Reson. Med. 2002; 48: 649-657.

41 Roth Y, Ocherashvilli A, Daniels D, Ruiz-Cabello J, Maier SE, Orenstein A, Mardor

Y. Quantification of water compartmentation in cell suspensions by diffusion-

weighted and T(2)-weighted MRI. Magn. Reson. Imaging 2008; 26: 88-102.

42 Frøhlich AF, Ostergaard L, Kiselev VG. Effect of impermeable boundaries on

diffusion-attenuated MR signal. J. Magn. Reson. 2006; 179: 223-233.

43 Li JG, Stanisz GJ, Henkelman RM. Integrated analysis of diffusion and relaxation of

water in blood. Magn. Reson. Med. 1998; 40: 79-88.

MRI of acellular matrix-based bladder tissue engineering

28

44 Chin CL, Wehrli FW, Hwang SN, Takahashi M, Hackney DB. Biexponential

diffusion attenuation in the rat spinal cord: Computer simulations based on anatomic

images of axonal architecture. Magn. Reson. Med. 2002; 47: 455-460.

45 Watanabe T, Murase N, Staemmler M, Gersonde K. Multiexponential proton

relaxation processes of compartmentalized water in gels. Magn. Reson. Med. 1992;

27: 118-134.

46 Stabler CL, Long RC, Sambanis A, Constantinidis I. Noninvasive measurement of

viable cell number in tissue-engineered constructs in vitro, using 1H nuclear magnetic

resonance spectroscopy. Tissue Eng. 2005; 11: 404-414.

47 Stabler CL, Long RC Jr, Constantinidis I, Sambanis A. In vivo noninvasive

monitoring of a tissue engineered construct using 1H NMR spectroscopy. Cell

Transplant 2005; 14: 139-149.

MRI of acellular matrix-based bladder tissue engineering

29

Table 1

Biophysical and biochemical changes in ACM from different manipulations

Manipulation Tissue Changes

Incorporation of

hyaluronic acid (HA)

Total GAG concentration increases, with a 2:1 ratio of non-

sulphated (i.e. HA) versus sulphated GAG.

Water content nearly doubles.

Seeding with smooth

muscle cells

Matrix collagen I degrades

MRI of acellular matrix-based bladder tissue engineering

30

FIGURE LEGENDS

Figure 1. Alcian blue staining confirms the uptake of hyaluronic acid (HA) in HA-ACM

constructs, which have a blue hue compared with ACM constructs without HA. (magnification,

10).

Figure 2. Representative H&E stained sections for the different classifications of cellularity: I

(very few cells), II (few cells), III (patches of cells), and IV (good cellularity). (magnification,

10).

Figure 3. Masson’s Trichome staining of the ACM shows a collagen extracellular matrix and

smooth muscle bundles (upper left corner). (magnification, 10).

Figure 4. Changes in the T2 spectrum (solid to dotted curves) can distinguish a) increased tissue

water content accompanying incorporation of hyaluronic acid and b) degradation of the

extracellular matrix with cell-seeding. In a) the higher water content is evident in a shift of the

long T2 component towards higher values. In b) progressive loss of matrix collagen is evident in

a reduced fraction of the short T2 component.

Figure 5. a) Diffusion-weighted and b) T2*-weighted MR images of an ACM in a 12-mm

diameter tube. Diffusion-weighted images are illustrated for b-values of 200 s/mm2 (left) and

1600 s/mm2 (right). T2*-weighted images are illustrated for echo times of 2.6 ms (left) and 34.4

ms (right).

Figure 6. Diffusion decay characteristics are different between unseeded ACMs (filled circles)

and seeded ACMs (open circles). Unseeded ACMs exhibit monoexponential decay, whereas

biexponential behaviour emerges in seeded ACMs due to the presence of both an extra- and

intracellular compartment.

MRI of acellular matrix-based bladder tissue engineering

31

Figure 7. Quantitative MRI parameters for discriminating various levels of cellularity. The slow

diffusion fraction, which represents the proportion of intracellular space, is significantly

correlated with increasing cellularity (I, II, III, IV). The other parameters (single diffusion

coefficient D, T1, and T2) show much lower correlation or none at all.

Figure 8. Effect on MRI parameters from seeding cells on ACM. T2* exhibits a significant and

greater relative change compared to either T2 or T1. The T2* increase is postulated to arise from

cell-filling of matrix pores. Mean values and standard deviation are shown in units of ms for

unseeded (shaded) and seeded (non-shaded) ACMs. ** P < 0.01, * P < 0.05.

Figure 9. Effect on MRI parameters from incorporating hyaluronic acid in ACM. T2* exhibits a

significant and greater relative change compared to either T2 or T1. The T2* decrease is

consistent with higher macromolecular content in the ACM and suggests that unlike T2, T2* is

relatively insensitive to tissue hydration. Mean values and standard deviations are shown in units

of ms for ACM (shaded) and HA-ACM (non-shaded). * P < 0.05.

FIG. 1

HA-ACMACM

FIG. 1

I II

III IV

FIG. 2

FIG. 3

T2 (ms)

0.1

0

0.1

0100 101 102 103

ampl

itude

ampl

itude

a

b

FIG. 5

a

b

0

500

1000

1500

2000

2500

3000

0

50

100

150

200

250

300

280

300

320

340

360

380

400

420

440

280

300

320

340

360

380

400

420

440

FIG. 4

0.01

0.1

1

0 500 1000 1500 2000 2500 3000

b-value (s/mm2)

Rel

ativ

e si

gnal

(S/S

o)

3 300

2

0.2

D (1

0-3

mm

2 /s)

T1 (s

)

T2 (m

s)

1

100

0

2002

1

0

0

0.1

I II III IV I II III IV

Cellularity Cellularity

FIG. 5

Slow

diff

usio

n fr

actio

n

42.7

71.3

125

159 19921854

T2* T2 T1

a.u.

FIG. 8

** *

42.7

23.4

125

156 1992

1526

T2* T2 T1

a.u.

FIG. 9

** *