Human Gene Therapy
Attenuation of Skeletal Muscle Wasting with Recombinant Human
Growth Hormone Secreted from a Tissue-Engineered Bioartificial
Muscle
Nov 1998, Vol. 9, No. 17 : 2555 -2564
Herman Vandenburgh, Michael Del Tatto, Janet Shansky, Lisa Goldstein, Kristina Russell,
Nicholas Genes, Joseph Chromiak, Shigeru Yamada
Skeletal muscle wasting is a significant problem in elderly and debilitated patients. Growth
hormone (GH) is an anabolic growth factor for skeletal muscle but is difficult to deliver in a
therapeutic manner by injection owing to its in vivo instability. A novel method is presented for the
sustained secretion of recombinant human GH (rhGH) from genetically modified skeletal muscle
implants, which reduces host muscle wasting. Proliferating murine C2C12 skeletal myoblasts
stably transduced with the rhGH gene were tissue engineered in vitro into bioartificial muscles
(C2-BAMs) containing organized postmitotic myofibers secreting 3-5 mu g of rhGH/day in vitro.
When implanted subcutaneously into syngeneic mice, C2-BAMs delivered a sustained
physiologic dose of 2.5 to 11.3 ng of rhGH per milliliter of serum. rhGH synthesized and secreted
by the myofibers was in the 22-kDa monomeric form and was biologically active, based on
downregulation of a GH-sensitive protein synthesized in the liver. Skeletal muscle disuse atrophy
was induced in mice by hindlimb unloading, causing the fast plantaris and slow soleus muscles to
atrophy by 21 to 35% (p < 0.02). This atrophy was significantly attenuated 41 to 55% (p < 0.02) in
animals that received C2-BAM implants, but not in animals receiving daily injections of purified
rhGH (1 mg/kg/day). These data support the concept that delivery of rhGH from BAMs may be
efficacious in treating muscle-wasting disorders.
Tissue-engineered skeletal muscle organoids for reversible
gene therapy.

Vandenburgh H, Del Tatto M, Shansky J, Lemaire J,
Chang A, Payumo F, Lee P, Goodyear A, Raven L.
Department of Pathology, Brown University School of Medicine,
Providence, RI 02906, USA.
Genetically modified murine skeletal myoblasts were tissue
engineered in vitro into organ-like structures (organoids)
containing only postmitotic myofibers secreting
pharmacological levels of recombinant human growth
hormone (rhGH). Subcutaneous organoid implantation under
tension led to the rapid and stable appearance of
physiological sera levels of rhGH for up to 12 weeks, whereas
surgical removal led to its rapid disappearance. Reversible
delivery of bioactive compounds from postmitotic cells in
tissue engineered organs has several advantages over other
forms of muscle gene therapy.
PMID: 8934233 [PubMed - indexed for MEDLINE]
Mechanical stimulation improves tissue-engineered
human skeletal muscle
Courtney A. Powell1, Beth L. Smiley2, John Mills2, and Herman H. Vandenburgh1,2,3
1
Department of Molecular Pharmacology, Physiology, and Biotechnology, Brown
University, Providence, Rhode Island 02912; 2 Cell Based Delivery, Incorporated,
Providence, Rhode Island 02906; and 3 Department of Pathology, Brown University
Medical School/The Miriam Hospital, Providence, Rhode Island 02906
ABSTRACT
Human bioartificial muscles (HBAMs) are tissue engineered by suspending muscle cells
in collagen/MATRIGEL, casting in a silicone mold containing end attachment sites, and
allowing the cells to differentiate for 8 to 16 days. The resulting HBAMs are
representative of skeletal muscle in that they contain parallel arrays of postmitotic
myofibers; however, they differ in many other morphological characteristics. To engineer
improved HBAMs, i.e., more in vivo-like, we developed Mechanical Cell Stimulator
(MCS) hardware to apply in vivo-like forces directly to the engineered tissue. A sensitive
force transducer attached to the HBAM measured real-time, internally generated, as well
as externally applied, forces. The muscle cells generated increasing internal forces during
formation which were inhibitable with a cytoskeleton depolymerizer. Repetitive
stretch/relaxation for 8 days increased the HBAM elasticity two- to threefold, mean
myofiber diameter 12%, and myofiber area percent 40%. This system allows engineering
of improved skeletal muscle analogs as well as a nondestructive method to determine
passive force and viscoelastic properties of the resulting tissue.
muscle hypertrophy; viscoelasticity; tension; collagen gel; force transducer;
organogenesis
Bioengineered muscle constructs and tissuebased therapy for cardiac disease
Herman H. Vandenburgh
,
Department of Pathology, Brown Medical School/Miriam Hospital, RISE Research Bldg., 164
Summit Ave., Providence, R.I. 02906, USA
Available online 8 February 2006.
Abstract
Bioengineering of contractile muscle tissues and organs for disease treatment is currently in the
preclinical experimental stage of development. Cell biology over the last several decades has
primarily focused on breaking down tissues and cells to smaller and smaller components to
understand their functioning at the molecular level. We are just beginning to understand how to
reassemble these components back into larger functional units. The merging of the fields of
traditional engineering, biomaterials, cell biology, and computer science will lead to the ex vivo
engineering tissues and organs for many cardiovascular applications in the future. For pediatric
cardiology, bioengineered contractile tissues may serve as force generating cardiac patches,
heart valve papillary muscle substitutes, or as living therapeutic protein delivery ‘devices’ when
bioengineered from genetically engineered muscle cells.
Keywords: Tissue engineering; Force generation; Gene therapy; Therapeutic protein delivery
Article Outline
1. Introduction
2. In vitro bioengineering
3. In vivo bioengineering
4. Bioengineering striated muscle for therapeutic protein delivery
Acknowledgements
References
1. Introduction
Bioengineering is a new discipline for the in vitro construction of implantable tissues such as
pancreatic islets, liver, skin, cartilage, bone, muscle, and blood vessels [1] and [2]. The primary
goal of bioengineering is to replace defective organs, but with current technology, bioengineering
human skeletal and cardiac muscle for structural repair of the damaged heart is not possible.
Organized contractile forces of sufficient magnitude required for supplementing the failing forces
generated by diseased or aged muscle is not currently possible, and will require the development
of new, more advanced bioengineering processes. In this short review, the progress that has
been made to date in engineering of contractile muscle will be outlined and areas requiring further
study indicated. A more complete review of skeletal muscle tissue engineering has been recently
published [3]. The potential use of current technologies for non-force generating clinical
applications of striated muscle such as therapeutic protein delivery will be summarized.
2. In vitro bioengineering
The isolation and growth of skeletal and cardiac muscle cells in monolayer tissue cultures is well
established for rodent and human skeletal muscle cells [4] and for rodent neonatal [5] and adult
[6] cardiomyocytes. Primary human cardiomyocytes are much more difficult to isolate and grow in
vitro than skeletal muscle cells, making the current interest in embryonic and adult stem cell
transformation into functioning cardiomyocytes [7] and [8] an attractive alternative cell source for
future bioengineering applications.
Striated muscle cells can be reorganized in tissue culture into in vivo like two-dimensional
structures by several methods. Neonatal rodent cardiomyocytes grown on an elastic substratum
and mechanically stretched can be aligned into thin sheets of organized rod-shaped contractile
cells [9] and [10]. Adult rodent cardiomyocytes are also aligned parallel to each other using
mechanical forces, but they are difficult to maintain in vitro in their rod-shaped differentiated
contractile state for more than several days [11]. Electrical field forces can also be used to
organize rodent cardiomyocytes [12]. Skeletal muscle cells grown in monolayers can also be
aligned into parallel arrays of contractile multi-nucleated muscle fibers (myofibers) by mechanical
forces [13]. In all cases, many variables determine the response of the cells to these organizing
mechanical forces which are similar to those which occur during embryonic development. The
pattern of mechanical loading, percent stretch, rate of stretch and rest periods are all important
factors, and each cell type responds differently to these parameters. Future bioengineering
studies will be required to understand the mechanisms underlying the cell-to-cell and cell-toextracellular matrix interactions involved in tissue organization, and to utilize this knowledge in the
bioengineering of functional organs [14].
Contractile muscle cells have also been bioengineered into three-dimensional cardiac [15], [16],
[17] and [18] and skeletal muscle tissues [19], [20] and [21] with some of the functional and
structural properties associated with cardiac and skeletal muscle. A standard method used for
bioengineering these tissues is to suspend several million cells in a small volume of ice cold
extracellular matrix collagen solution and cast the cell matrix mixture into a mold of the desired
shape (Fig. 1). When warmed to 37 °C, the collagen matrix gels and over several days undergo
cell cytoskeletal-induced gel contracture. If ‘attachment’ sites are provided in the mold, the
contracted cell:gel mixture will remain under passive tension, and this tension will align the
proliferating cells parallel to each other in the direction of the matrix tension vector (Fig. 1, lower
left insert). As the differentiating cells interact with their neighboring cells and the matrix, they
form parallel arrays of postmitotic skeletal muscle fibers or cardiomyocytes linked by intercalated
disks, allowing directed force generation when stimulated. Repetitive mechanical forces can be
applied in vitro to these ‘bioartificial muscle’ (BAM) tissues using various electromechanical
stimulation hardware, and chronic repetitive loading improves their functionality [22] and [23]. But
we still do not have the capability to generate adult striated muscle cells in vitro of the appropriate
size and cell packing density to generate the directed forces necessary to perform functional work
in vivo. The best active forces generated to date are only 2–5% of those generated by in vivo
adult muscle [18] and [20] (Vandenburgh et al, unpublished data). The limited force generating
capacity of bioengineered muscle results partly from the current ability to provide nutrients to the
multilayered in vitro tissues only through diffusion since they lack a vasculature. Because of
striated muscles' high metabolic needs when performing work, skeletal and cardiac muscles are
very highly vascularized in vivo, with approximately one capillary associated with each cell.
Manipulation of the tissue culture medium, growth factor supplements, perfusion conditions,
electrical stimulation, mechanical loading/conditioning, and co-culturing with blood vessel forming
cells [24], [25] and [26] or neurons [27] will eventually generate more functional muscle tissues for
transplantation. Currently, these parameters have to be individually determined by trial and error.
In the future, computer-controlled feed back systems will be developed to optimize the
bioengineering process, based on maximizing force generating capacity of the tissue.
(82K)
Fig. 1. Bioengineering a bioartificial muscle (BAM). Passive internal forces generated by
proliferating muscle cells cast in a collagen extracellular matrix gel causes the cell:gel mixture to
contract away from the mold walls and remain attached to end support structures (stainless steel
screening, lower right photo). The cell:gel mixture is under passive tension, as seen when one
end of the construct is detached from its end support (photo, lower left). The passive tension
aligns the fusing muscle cells with the matrix tension vector, causing them to form aligned
contractile muscle fibers which express sarcomeric tropomyosin (whole mount immunostain,
middle left photo). In cross section (photo, upper left), a high packing density of muscle fibers
(myosin immunostained) occurs if the BAM diameter is 200 μm or less.
Enhancing tissues through genetic engineering may also help circumvent some of the current
roadblocks to bioengineering tissues in vitro for their successful transplantation and functionality
in vivo. For example, genetic engineering of BAMs to locally secrete growth factors such as
insulin-like growth factor-1 significantly increases the number of BAM skeletal muscle fibers and
their force generation ability (Shansky et al., submitted for publication; Lee et al., in preparation).
In another example, genetic engineering of bioengineered muscle tissues to locally secrete the
angiogenic protein vascular endothelial growth factor (VEGF) stimulates their rapid in vivo
vascularization [28] and muscle cell survival [29]. These results indicate an important research
area for future bioengineering with genetically modified cells.
3. In vivo bioengineering
We are currently able to isolate adult human skeletal muscle stem cells (satellite cells), expand
them to billions of myogenic cells (myoblasts) in tissue culture, and reconstitute them in vitro into
organized tissue containing partially differentiated contractile myofibers as outlined above. A
similar readily available source of adult cardiac stem cell source is not available, although there is
evidence of their in vivo existence [30] and [31]. Proliferating skeletal myoblasts have been
shown to successfully generate functional contractile tissue (in vivo bioengineering) when injected
into damaged heart tissue in animal studies [32] and [33], and there are a number of current
clinical trials underway using adult human skeletal muscle myoblasts injected into the damaged
heart for functional repair [34], [35] and [36]. Only a small percent of the injected cells survives
and partially differentiates, while the large majority of cells (> 90%) undergoes rapid apoptosis
and die. As with in vitro bioengineering, genetically enhanced muscle cells offer an interesting
option to improve muscle cell transplantation protocols for this “in vivo bioengineering” approach.
For example, skeletal muscle myoblasts genetically engineered to express the intercalated disc
protein connexin43 form better intercellular gap junctions in vitro [37], and cells engineered to
express either the angiogenic protein VEGF [38], or a fibrotic inhibitor protein IL-1 inhibitor [39]
performed better when injected into an ischemic heart model. The later IL-1 inhibitor also
improved muscle cell survival when implanted into the heart [40]. This combination of genetic
engineering of cells for transplantation survival and functionality holds much promise for future in
vivo bioengineering for clinical cardiovascular applications.
4. Bioengineering striated muscle for therapeutic
protein delivery
While the structural use of the bioengineered contractile tissue is a distant goal, several near term
goals are more feasible. These include their use for therapeutic drug testing for muscle wasting
diseases [19] and [41], high content drug screening in vitro [23] and [42], and as implantable
protein delivery ‘devices’ following genetic engineering of the cells to secrete a therapeutic protein
(Fig. 2). We have successfully engineered skeletal muscle BAMs to secrete systemically
therapeutic levels of recombinant proteins such as growth hormone (GH), vascular endothelial
growth factor (VEGF), and the blood clotting Factor 9 (F9) for up to 6 months in mice. The use of
genetically modified myoblasts to bioengineer implantable BAMs for the chronic delivery of
therapeutic proteins would have numerous potential advantages over other forms of muscle cell
and gene protein therapy. First, myoblast fusion efficiency into myofibers in vitro is several orders
of magnitude greater than that which occurs in vivo from injected myoblasts. Thus, while in vivo
fusion efficiencies in human myoblast cell therapy are on the order of 0.5% to 5% [43], in vitro
fusion efficiencies of myoblasts are 60% to 80%. The majority of the myofiber nuclei in BAMs
formed from these myoblasts therefore contain the foreign transgene, making a more
reproducible, high-secreting delivery implant from non-dividing cells [44]. Second, myoblast viral
transduction efficiencies are more targeted and generally greater in vitro (90–100%) than in vivo
[29]. Third, formation of BAMs in vitro allows the preimplantation monitoring of protein secretory
levels, leading to predictive protein dosage delivery in vivo [28] and [44]. Fourth, implantation of
non-migratory, non-proliferating postmitotic cells containing the foreign transgene reduces the
possibility of proliferating cell transformation, migration, and tumor formation [45]. Finally,
implanting a defined organ-like structure allows surgical removal of the implant, if necessary [44].
The recent instance of in vitro retroviral gene therapy-related leukemia in several SCID infants
(which resulted from the specific nature of the transgene inserted and stem cell type utilized [46]),
warrants a cautious approach for any new gene therapy protocols. Thus, potential retrievability of
implanted genetically engineered cells is critically important for any new viral vector transgene
insertional approach along with the potential future use of nonviral gene transfer protocols [47].
(66K)
Fig. 2. Autologous bioengineered tissue delivery of a therapeutic protein.
The feasibility of long term recombinant protein delivery from BAMs was initially established using
murine myoblasts stably transduced with the GH transgene [44]. These cells were tissue
engineered in vitro into BAMs containing organized postmitotic skeletal myofibers secreting high
levels of GH. Subcutaneous (subQ) implantation of the BAMs under tension into syngeneic mice
led to the rapid and stable appearance of physiological levels of GH in the serum for up to six
months. The implanted BAMs retained their preimplantation structure, allowing surgical removal,
and rapid disappearance of GH from the blood [44]. Maintenance of tension on the myofibers in
the BAMs was critical for preventing myofiber atrophy and a > 70% reduction in recombinant
protein output [44]. When implanted subQ under tension into hindlimb-unloaded mice, the atrophy
of the host soleus and plantaris muscles was significantly attenuated compared to animals
implanted with non-GH secreting BAMs [48]. The BAMs secreting GH were more effective in
attenuating muscle atrophy than daily injections of GH. Similar studies have been performed with
murine BAMs delivering VEGF to an ischemic limb [28], and autologous sheep BAMs delivering
VEGF to the sheep heart [49]. Bioengineered tissue delivery of therapeutic proteins is therefore
conceptually proven, but scale-up to a clinically useful product in humans is still a major
challenge.
Recently, human BAMs (HBAMs) bioengineered with FDA approved materials were shown to
deliver F9 systemically for greater than eighty days in a predictive manner in an immunodeficient
mouse [50], a preclinical animal model accepted by the FDA for hemophilia gene therapy [51].
This predictable and reproducible delivery of rF9 from in vitro engineered and implanted HBAMs
is in marked contrast to the poor predictability and tumorgenicity of implanted embryonic stem
cells genetically engineered in vitro to express F9 [52]. Coupling the idea of genetically enhancing
implanted tissue survival with systemic therapeutic protein delivery, we showed that localized
VEGF from implanted HBAMs elevated several-fold the systemic long term delivery of F9 [29].
Major challenges in this application of engineered tissues include improved cell survival following
transplantation and increased therapeutic protein secretion levels that would be useful in humans.
Future technological developments in HBAMs as protein delivery devices will provide useful
information for eventual bioengineering of skeletal muscle implants as structural contractile tissue
substitutes. For pediatric cardiology, bioengineered contractile tissues may serve as force
generating cardiac patches, heart valve papillary muscle substitutes, or as living therapeutic
protein delivery ‘devices’ when bioengineered from genetically engineered muscle cells. We are
just beginning to learn to use the tools of cell and molecular biology for bioengineering tissues
which will eventually revolutionize medical treatment for many cardiovascular disorders.