2 Immune Response, Immunogenicity, and Transplants

Perspectives on the Translational Aspects of Articular Cartilage Biology



Large mononuclear cells per group day 42

14,000



A



Normal



10,000

AB



Cells/µl



8,000



BC



6,000

4,000

CD CD



2,000



BCD

D CD



D



0

Autologous

Allogeneic

Xenogeneic



BCD



Control



Empty defect

Explant

Construct



+

-



+

-



Patella

+ +

-



+

-



+

-



+

-



+



ReacƟve



+



+



Trochlea

+ - + - +

-



+

-



+



+

-



+



+



Figure 6.5 Articular cartilage’s immunoprivilege is location dependent. As shown

here, only the xenogeneic implants elicit immune responses (e.g., as manifested

by the presence of large mononuclear cells) at day 42 when the implant is in the

trochlea. However, if implanted in the patella, both allogeneic and xenogeneic

implants continue to elicit an immune response. Images of normal and reactive large

mononuclear cells are shown on the right. (From Arzi, B. et al., Acta Biomater 23: 72-81,

2015. With permission.)



studies  have quantitatively assayed immune reaction to cartilage

implants. To understand the immunological components that may

be present in cartilage implants and transplants, it is worthwhile to

have a description of the immune system. Thus, this section begins

with a brief overview of innate and adaptive immunity. Focus will

then be placed on the immune responses during articular cartilage

transplantation and repair. The inflammation associated with osteoarthritis has already been discussed earlier (Chapter 3). It is our goal

to introduce the reader to foundations of the immunological basis of

tissue rejection, as related to assessment in articular cartilage transplantation strategies.

Allogeneic and xenogeneic cells and matrices can elicit responses from

both the innate and adaptive immune systems. Maximal activity of the

477



Articular Cartilage



innate immune system, rapidly reached following the introduction of

foreign materials, consists of nonspecific responses by both humoral

and cellular components.

The purpose of the innate immune system is to recruit immune cells and

promote healing. The complement system is the major component of the

humoral response (Figure 6.6). One of the functions of the complement

system involves the formation of the membrane attack complex, which

causes lysis of foreign cells (e.g., pathogenic bacteria). Complement can

also lead to other catabolic cascades. Positive feedback involving protease activity, peptides that recruit immune cells, and increased vascular permeability allow for the destruction of marked, foreign targets.



Humoral immunity



Cellular immunity



Extracellular microbe

(e.g., bacteria)



Intracellular microbe

(e.g., viruses)

Antigen-presenting

cell



B lymphocytes

Helper

T cell



Secreted

antibody



T-cell

receptor



Processed and

presented antigen

Cytokines



Neutralization



Proliferation

and activation

of effector cells

(macrophages,

cytotoxic T cells)



Cytokine

receptor



Lysis (complement)

Phagocytosis

(PMN, macrophage)

Lysis of

infected cell



Destruction of

phagocytosed microbes



Figure 6.6 Humoral and cellular immunity. (From Kumar, V. et al., Robbins Basic

Pathology, Philadelphia: Elsevier, Health Sciences Division, 2007. With permission.)



478



Perspectives on the Translational Aspects of Articular Cartilage Biology



Phagocytic cells that engulf and digest foreign targets include neutrophils, macrophages, and dendritic cells. During the acute phase of

immune reactions against implants, the presence, type, and number of

these cell types can be quantified, such as through the synovial fluid, in

synovial biopsies, and potentially in the implant itself. For a more thorough discussion of the innate immune system, the reader may refer to

several publications on this subject (Janeway and Medzhitov 2002; Rus

et al. 2005).

Activated by the innate immune system, the adaptive immune system

(Figure 6.7) also consists of humoral and cellular components. The

humoral response is generally directed against bacteria, though implant

rejection can also occur via this response, and is mediated by B lymphocytes. The naïve B cell recognizes the bacteria and presents it to T helper 2

cells. Cytokines then induce B cells to produce antibodies, which can work

in several ways to neutralize the antigen. With the antigen defeated, the

response is then downregulated, and memory B cells form. Antibodies

can inhibit the adhesion of bacteria by surrounding them. Antibodies

Innate immunity



Adaptive immunity



Microbe



Antibodies



B lymphocytes



Epithelial barriers



T lymphocytes



Effector

T cells



Phagocytes



Complement

0



6

Hours



NK cells

12



1

Time after infection



2



3

Days



4



5



Figure 6.7 The various components of innate and adaptive immunity. (From Kumar,

V. et al., Robbins Basic Pathology, Philadelphia: Elsevier, Health Sciences Division,

2007. With permission.)



479



Articular Cartilage



can also promote opsonization. Whereas B  lymphocytes mediate the

humoral component of the adaptive immune system, the cellular component is mediated by T lymphocytes. First, a sensitization phase occurs

whereby an antigen (e.g., a transplant) is recognized by a macrophage and

presented to T lymphocytes. Antigen-presenting cells that express appropriate antigenic ligands on their MHC receptors, in addition to required

costimulator signals, activate T lymphocytes. Cytokines then initiate the

proliferation phase, where cytotoxic T lymphocytes multiply against the

antigen. Effector immune responses then proceed to defeat the antigen.

Activated T lymphocytes secrete various cytokines (e.g., interleukin  2

[IL-2], interferon γ [IFN-γ], tumor necrosis factor beta [TNF-β]) to recruit

a variety of other host immune cells, inducing increased expression of

MHC class I and class II molecules by foreign (e.g., donor) cells. Following

the destruction of the antigen, the response is downregulated by T suppressor cells, and memory T cells mature for future recurrences (Goldsby

et al. 2003; Pietra 2003; Hale 2006; Trivedi 2007).

Both acute and chronic immune responses may be directed toward allogeneic and xenogeneic implants. At present, these responses are seldom

quantified for patients receiving chondral or osteochondral implants.

Inflammation is qualitatively assessed, but parameters such as number,

type, and location of leukocytes are not evaluated or managed. A body

of literature exists for allogeneic and xenogeneic transplants to inform

the development of future cellular therapies. Careful tracking of immunological parameters to discern and to manage immune responses

should not be neglected.

6.2.2 Allogeneic Transplants

As mentioned previously, transplant approaches using autologous

chondrocytes and osteochondral plugs are limited by the small amount

of available donor tissue. Allogeneic sources provide a larger donor

pool. However, the use of allogeneic tissues presents an elevated risk

of disease transmission, and testing for diseases such as HIV, hepatitis,

480



Perspectives on the Translational Aspects of Articular Cartilage Biology



and syphilis must first be performed. Immune responses against the

cells and tissues also present challenges.

Allogeneic chondrocyte transplantation can be difficult due to the

humoral response mounted against these cells (Langer and Gross 1974).

Experiments outside of the joint capsule show that allogeneic chondrocytes are not immunoprivileged by themselves. Allogeneic chondrocytes implanted into posterior tibial muscles formed nodules that

immediately attracted macrophages, and natural killer and cytotoxic

or suppressor T cells were also recruited to destroy the nascent cartilage over time (Romaniuk et al. 1995). This slow destruction of the

repair tissue has been shown in several other studies (Kaminski et al.

1980; Ksiazek and Moskalewski 1983; Malejczyk and Moskalewski 1988;

Malejczyk et al. 1991). In contrast, cells embedded in matrix were relatively protected. Implantation of chondrocytes alone versus chondrocytes enveloped in matrix has shown that the immune response was

greatly diminished in the latter case (Green 1977).

In animal models, when allogeneic chondrocytes were implanted

with their extracellular matrices into rat, rabbits, or dogs, neither

significant leukocyte migration nor cytotoxic humoral antibodies

were observed in several studies (Langer and Gross 1974; Aston and

Bentley 1986; Glenn et al. 2006). Another study, however, has shown

increased presence of inflammatory mononuclear cells and less

repair cartilage for antigen-mismatched transplants in canine articular defects (Stevenson et al. 1989). Due to the potential for immune

responses, and due to the lack of availability of fresh cadaveric donor

tissue, frozen or pressure-washed osteochondral allografts have

been examined. In these cases, the cells are likely dead, reducing

the immunogenic response (Elves 1974; Friedlaender 1983), but the

grafts are biochemically and histologically inferior to fresh grafts.

However, cryopreservation allows for tissues to be banked and

greater time in screening the tissues for diseases (Flynn et al. 1994;

Bakay et al. 1998).

481



Articular Cartilage



Allografts have shown considerable success and are often used for

defects larger than 2-3 cm2. Recent consensus statements indicate that

allografts have sufficient evidence for considering their use in lesions

greater than 4 cm2 (Biant et al. 2015). In long-term follow-ups of up to

15  years of patients receiving fresh osteochondral allografts, allograft

survival rates of 75-95% at 5 years, 64-80% at 10 years, and greater than

60% during years 14 and 15 were observed (Beaver et al. 1992; Gross et al.

2005). However, compared to unipolar repairs, clinical trials have demonstrated that allograft implantation is unsuitable for bipolar (replacement of both tibial and femoral condyle in one compartment) lesion

repairs, with 50% of grafts surviving after 6 years, compared to 84% in

unipolar repairs (Zukor et al. 1989; Chu et al. 1999). Cryopreserved and

frozen allografts have yielded good to excellent scores following transplantation in roughly 70% of patients for up to 4 years. It is worth noting

that while success has been demonstrated in the treatment of condylar

lesions using allografts, the procedure is still considered to be a salvage

operation and is currently only suited for young, active patients with

isolated patellofemoral articular cartilage disease, for whom previous

procedures have failed. Also, a more vigilant effort in identifying the

number, type, and location of leukocytes present should be made to

observe the time course of immune reactions against the implants. Such

studies would determine if allogeneic implants elicit low but chronic

immune reactivity and whether this contributes to implant failure.

Live allograft cartilage (De NoVo graft pieces) is currently used as cartilage filler for cartilage defects (Figure 6.8). Alhough not yet available

as therapies, emerging technologies employing in vitro tissue engineering have shown success when allogeneic cells are combined with scaffolds. Implantation of allogeneic chondrocytes embedded in collagen

(Wakitani et al. 1989, 1998; Masuoka et al. 2005), agarose (Rahfoth et

al. 1998), and poly-glycolic acid (PGA) (Freed et al. 1994; Schreiber et

al. 1999), among other materials, has been examined in various animal

models. In general, hyaline histological appearances were found with

little to no sign of immunologic reactions. Similarly, the implantation

of allogeneic mesenchymal stem cells in a hyaluronic acid-based gel in

482



Perspectives on the Translational Aspects of Articular Cartilage Biology



Figure 6.8 Treatment of a patellar chondral defect using juvenile articular cartilage

allografts. A second-look arthroscopy at 2 years showed persistent filling of the defect

with cartilage. (From Griffin, J. W. et al., Arthrosc Tech 2(4): e351-e354, 2013. With

permission.)



a goat model has shown only mild immunologic rejection (ButnariuEphrat et al. 1996). As seen with the contrast between chondrocytes

alone and chondrocytes with associated matrix, in vitro seeding and

culture of these tissue-engineered constructs allow for the formation of

a protective, hyaline-like matrix around the cells prior to implantation.

6.2.3 Xenogeneic Transplants

While allogeneic tissues are more easily procured than autologous tissues, xenogeneic tissues are of even greater abundance. In this case, the

source is of a different species, and immunological concerns are heightened. No articular cartilage product using live xenogeneic cells currently

483



Articular Cartilage



exists, although the methodology has been examined in several animal

models. Rat chondrocytes implanted in rabbit muscle resulted in the

complete destruction of the implant by macrophages and giant foreign

body cells (Osiecka-Iwan et al. 2003). However, as articular cartilage

is thought to exhibit some immunoprivilege, xeno-implantation into

articular defects has been studied, with mixed results. Using fibrin glue

as a matrix, rabbit chondrocytes transplanted into the medial femoral

condyle of goats resulted in mild synovitis and the formation of fibrous

repair tissue (van Susante et al. 1999). Implantation of pig chondrocytes

into osteochondral defects of adult rabbits resulted in the production of

hyaline-like tissue with the absence of inflammatory cells (Ramallal et

al. 2004). The immunogenicity of chondrocytes from transgenic pigs has

also been examined in vitro. Chondrocytes isolated from H-transferase

transgenic pigs have been shown to have lower expression of the Gal

alpha(1,3)Gal antigen (alpha-gal) that humans reject (Costa et al. 2002)

and, as a result, experience lowered complement deposition and monoblast adhesion (Costa et al. 2008). Knockout animals have been produced

that lack α-1, 3-galactosyltransferase and do not produce alpha-gal.

These animals can produce tissue with lower immunogenicity (Figure

6.9). Aside from chondrocytes, no immune reaction was found when

human mesenchymal stem cells were implanted into a swine model to

restore the articular surface (Li et al. 2009). This may be in part because

mesenchymal stem cells have been shown to display immune suppressive properties (Ren et al. 2008; Bassi et al. 2011; Kuci et al. 2011; Choi et

al. 2012; Han et al. 2012; Yang et al. 2012; Yi et al. 2012).

In addition to cells, xenogeneic tissues have also been examined. In general, the implantation of xenograft tissue results in hyperacute rejection

of the implant (Auchincloss 1988; Platt et al. 1990). Typically, joint tissues

such as articular cartilage are not subject to hyperacute rejection due to

the avascularity of the tissue; thus, cartilage is considered to be relatively

nonimmunogenic (Jackson et al. 1992; Revell and Athanasiou 2009). It

has been believed that decellularizing xenogeneic tissue will be a viable

option for the generation of replacement tissue. Decellularization is a

process by which the antigenic intracellular proteins and nucleic acids

484



Perspectives on the Translational Aspects of Articular Cartilage Biology



Figure 6.9 Joy is one of five cloned piglets produced with the α-1,3-galactosyl

transferase gene knocked out and, therefore, does not produce alpha-gal. (From

cover of Nature Biotechnology, March 2002, Volume 20, No. 3. Dai, Y. et al., Nat

Biotechnol 20(3): 251-255, 2002. With permission.)



are removed. It is typically evaluated histologically by the absence of

nuclei, and the intent is to preserve the functional properties of the tissue’s extracellular matrix. Ideally, the biomechanical properties of the

tissue will also be preserved.

Examples of decellularized tissues include an acellular dermal matrix

(Chen et al. 2004) that has seen successful use clinically as the FDAapproved Alloderm product or the porcine-derived Strattice product.

Experimental investigations into acellular xenogeneic tissues have been

conducted for many musculoskeletal applications, including replacements for the knee meniscus (Stapleton et al. 2008), temporomandibular

485



Articular Cartilage



joint disc (Lumpkins et al. 2008), tendon (Cartmell and Dunn 2000), and

ACL (Woods and Gratzer 2005). Other tissues, including heart valves

(Kasimir et al. 2003; Grauss et al. 2005; Meyer et al. 2005, 2006; Tudorache

et al. 2007; Liao et al. 2008; Seebacher et al. 2008), bladder (Rosario et al.

2008), artery (Dahl et al. 2003), and small intestinal submucosa (Hodde

and Hiles 2002; Hodde et al. 2007), have also been decellularized. Despite

this large body of literature, the absence of cells is not tantamount to the

absence of antigens.

The primary assessment method for many decellularization studies

is the qualitative or semiquantitative assessment for the presence of

cells by histology and nuclear staining. However, it has been shown

that (1) residual cells and (2) immunoreactive antigens do not correlate

well with the lack of cells (Wong et al. 2011). Decellularization is not

tantamount to antigen removal, and residue antigens elicit an immune

response. Methods that rupture the cells and degrade the DNA do not

necessarily remove the resulting cellular and DNA fragments or debris

from the decellularized tissues, especially if the tissues are dense.

Nonetheless, short- to mid-term positive effects can be seen with using

decellularized tissues. For example, while a photooxidation approach

for decellularization of bovine xenografts implanted into sheep femoral

condyles reduced monocyte and plasma cell infiltration in the implant

after 6 months, the method did not remove DNA or reduce antigens

(von Rechenberg et al. 2003). For these implants, however, it is unclear

if low-level, chronic immune responses eventually lead to rejection or

tissue failure.

Various chemical treatments have been developed for decellularization,

such as 1% sodium dodecyl sulfate (SDS), 2% SDS, 2% tributyl phosphate (TnBP), 2% Triton X-100, and hypotonic followed by hypertonic

solutions (Cartmell and Dunn 2000; Hodde and Hiles 2002; Kasimir et

al. 2003; Grauss et al. 2005; Meyer et al. 2005, 2006; Woods and Gratzer

2005; Hodde et al. 2007; Tudorache et al. 2007; Liao et al. 2008; Lumpkins

et al. 2008; Seebacher et al. 2008; Stapleton et al. 2008). Decellularization

methods have also been applied to self-assembled tissue-engineered

486



Perspectives on the Translational Aspects of Articular Cartilage Biology



cartilage constructs and cartilage explants (Elder et al. 2009, 2010). All

SDS treatments resulted in cell removal histologically, but 2% SDS for

1 hour decreased DNA content by 33%, while maintaining biochemical

and biomechanical properties. Additionally, 2% SDS for 8 hours resulted

in complete histological decellularization and a 46% reduction in DNA

content, although compressive stiffness and glycosaminoglycan content

were significantly compromised. There is a dearth of studies demonstrating the effects of tissue decellularization on native as well as engineered

articular cartilage constructs. Future work in this area should focus on

antigen removal while preserving cartilage biomechanical properties.

Antigen removal should be used as the primary assessment criterion

in decellularization processes, and the absence of cell and nuclear

materials histologically should serve as an adjunctive but secondary

assessment (Figure 6.10). This is because, aside from cell debris that

can remain in the tissue after decellularization, the extracellular matrix

can also contain antigens that would not appear with nuclear staining,



Harsh

decellularization

(e.g., SDS)



Possible mechanical

failure

Hydrophilic protein

Lipophilic protein



Gentle

decellularization

(e.g., hypotonic saline)



Collagen



Possible immune

rejection



Glycosaminoglycan

Hyaluronate

Chondrocyte



Targeted

antigen

removal



Macrophage



Successful graft



Figure 6.10 For articular cartilage, strategies to remove cells need to balance

immunogenicity and mechanical properties because the chemicals used to remove cells

can also remove cartilage glycosaminoglycans. Thus, it may be more strategic to remove

only the antigens that are responsible for rejection, but no other extracellular component.

(From Cissell, D. D. et al., J Biomech 47(9): 1987-1996, 2014. With permission.)



487