Cellular stroke therapy: from cell replacement to trophic support

The past decade has marked a significant paradigm shift in the postulated recovery mechanisms behind cell-based therapies for stroke. Although initially considered as primarily a cell-replacement strategy, researchers today attribute a combination of both cellular repair and trophic support from transplanted cells as mediators of recovery. There is an increasing body of evidence in support of the stem cell as a ‘micropump’ of proangiogenic factors, immunomodulators and enhancers of neural plasticity, as well as endogenous neurogenesis. It is crucial to design preclinical studies aimed at understanding the influence of stem cells on the microenvironment of the brain after stroke. These findings will significantly impact how we will design future clinical trials. Key variables including cell type, mode of delivery and use of imaging modalities, as a measure of both safety and efficacy will all be paramount in the translation of novel therapies from bench to bedside.

Formidable challenge of brain repair after stroke

Unlike organs with repeating microstructures, such as the liver and pancreas, the brain contains discrete subunits of multiple cell types with layered organization, interconnected in a complex manner and supported by elaborate microvasculature. While pathologies such as Parkinson’s disease or amyotrophic lateral sclerosis are characterized by the loss of a discrete subpopulation of cells that can be potentially replaced, ischemic injury gives rise to destruction of the entire cerebro-architecture. Moreover, in addition to local damage, pre- and post-synaptic degeneration leads to functional loss in distant but related areas of the brain. As a result, the reconstruction of this complex network by local cell injection may be an unrealistic goal. Although engraftment of rodent and human neural stem cells has been demonstrated repeatedly, showing neuronal, astroglial and oligodendroglial differentiation [1–3], the role of secreted factors derived from transplanted cells that support the induction of post-injured brain remodeling appears to carry increasing importance.

In this article, I will consider four main areas of trophic support: increase in angiogenesis, enhancement of neural plasticity, influence on endogenous neurogenesis and immunomodulation.

Increase in angiogenesis

In the post-stroke human brain, endogenous angiogenesis has been shown to be mediated by factors including TGF-β [4], PDGF [5], VEGF [6] and FGF-2 [7]. We have recently demonstrated endogenous upregulation of the VEGF receptor and angiogenesis in post-stroke rats as early as 2 days after stroke, with a peak expression at 10 days, using in vivo⁶⁴Cu-DOTA-VEGF₁₂₁ PET imaging [8]. Both rodent and human studies have suggested that the majority of post-stroke angiogenesis occurs in the tissue adjacent to the core of the stroke. This tissue, which remains at risk, can persist for several days [9] and therefore represents a viable target for therapeutic intervention. There are now an increasing number of studies showing that transplanted neural stem cells [10], bone marrow stromal cells [11], cells from human cord blood [12] and peripheral blood enhance endogenous angiogenesis via the secretion of proangiogenic factors leading to behavioral recovery.

Another important aspect of post-stroke vascular recovery is the coupling of angiogenesis and neurogenesis. A major regulator of the vascular niche is VEGF and its receptor VEGFR-2 [13]. Several types of stem cells have been shown to secrete VEGF and could therefore play an important role in restoring the neurovascular unit [14]. This neurovascular niche has been shown to not only play a major role in hippocampal function but also in stroke-induced endogenous neurogenesis. Activated endothelial cells secrete VEGF and chemoattractant factors such as stromal-derived factor-1, which attract CXCR4-expressing neuroblasts from neurogenic areas to the site of injury [15]. This interplay between the vasculature and brain parenchyma demonstrates the importance of fortifying the neurovascular unit for post-stroke recovery.

Enhancement of neural plasticity

As with angiogenesis, the ability of cell-based therapies to enhance injury-induced plasticity has been demonstrated with several different stem cell types. Mechanisms illustrated in these studies include an increase in neurite length and outgrowth through VEGFR-2, MAPK and PI3K/Akt pathways. Our group has observed a similar phenomenon with fetal-derived neural stem cells and human embryonic stem cell-derived neural progenitors leading to enhanced dendritic branching in both the ipsi- and contra-lesional hemispheres [16,17]. In addition to neurite outgrowth and dendritic branching, we have shown cell-mediated enhancement of synaptogenesis through secretion of thrombospondins, a family of large extracellular matrix proteins [18]. Finally, studies using novel MRI modalities, such as fractional anisotropy and diffusion tensor imaging, have enabled the observation of post-transplantation reorganization of white matter tracts surrounding an infarction [19].

Influence on endogenous neurogenesis

Endogenous neurogenesis increases after a stroke [20,21]. The function of this process is not fully understood, but may represent a natural repair mechanism that could be enhanced by transplanted cells. There is precedence for this as demonstrated with cord blood- and bone marrow-derived cells [11,22]. Mesenchymal stem cell-treated rats demonstrated higher numbers of oligodendrocyte precursors, which increased in concert with enhanced white matter areas [23]. Whether there is cell-mediated enhancement of endogenous neurogenesis or cell-mediated neuroprotection of newly born endogenous stem cells is unclear. We have recently shown promotion of endogenous neurogenesis and reduction in apoptotic cell death mediated through the canonical Wnt pathway [24], a signaling process that could potentially be modulated by cell transplantation.

Imunomodulation

In addition to angiogenesis, neuroprotection and enhancement of injury-induced plasticity, transplanted neural progenitor cells can also decrease post-ischemic inflammatory damage. Recently, the ability of transplanted neural progenitor cells to promote neuroprotection through an immunomodulatory strategy has been described [25]. Once within inflamed CNS areas, systemically injected neural progenitor cells persist around the perivascular space where inflamed endothelial cells, reactive astrocytes and blood-borne infiltrating T lymphocytes co-reside. In this perivascular space, the neural progenitor cells can then promote neuroprotection by releasing anti-inflammatory chemokines and by expressing immunomodulatory molecules (e.g., FasL, Apo3L and TNF-α-related apoptosis-inducing ligand) [25], influencing the invasion of inflammatory cells such as T cells. Regulatory T lymphocytes, for instance, have been shown to be key cerebroprotective immunmodulators in acute experimental stroke [26]. Modulation of inflammation after stroke can promote neuroprotection and endogenous neurogenesis [21]. Thus, stem cell-mediated manipulation of the intricate cross-talk between neural and immune cells reveals additional therapeutic opportunities.

To have a maximal impact on the injured host tissue via the aforementioned repair mechanisms, stem cells would need to be distributed widely in the affected brain areas. Numerous studies have reported extensive migration of stereotactically transplanted stem cells after stroke in the rodent brain [2,27–29]. There is, however, no evidence suggesting that this extensive migration is present in the human brain [30]. It is, therefore, unclear whether transplanted cells could reach all of the appropriate sites of action on a microscopic scale after stereotactic injection. An interesting alternative is intravascular cell transplantation. This approach would allow for widespread distribution of stem cells in the injured brain and would rely less on cellular migration. Furthermore, this technique carries the added clinical benefit of being less invasive than stereotactic transplantation. Whether an intravenous or an intra-arterial approach should be chosen is yet another matter of debate. Two potential problems exist with intravenous injection: the lack of efficacious cellular integration in the injured brain, and the pulmonary first-pass effect resulting in the trapping of a majority of injected cells in the lungs [31,32]. Despite minimal cellular integration in the injured brain after intravenous injection, studies have demonstrated beneficial effects after intravenous stem cell injection, questioning the necessity of cellular homing to the injured brain area [33]. However, it remains unknown whether long-term trophic support fortified by integration of stem cells is superior to a short-term release of modulating factors in the systemic circulation, as observed after bone marrow-derived stem cell injection. We have previously described a robust CD49d-dependent (CD49d is the ligand to VCAM-1) and CCR2-dependent (chemokine receptor for monocyte chemoattractant protein) migration of neural stem cells to the ischemic hemisphere following intra-arterial injection [10,34]. We found a good cell survival at 2 weeks with an increase in angiogenesis and an improved functional recovery in cell-treated animals. Intra-arterial cell injection seems an attractive and efficacious option, and would be conceivable in patients via a selective endovascular catheter injection. One recently described caveat is the danger of microembolic strokes induced by intra-arterial cell delivery, a potential problem that needs to be addressed prior to clinical application [35]. The number of cells, the cell concentration and the maintenance of cerebral blood flow during the injection process are critical factors that must be controlled.

Another area of intense research is the further development of molecular imaging modalities in conjunction with the discovery of biomarkers to determine the optimal time of cellular injection. Treatment scenarios in which we guide our time of therapy based on the expression of adhesion molecules [10] and chemokine secretion or try to achieve maximal effect of stem cell-secreted VEGF during VEGF receptor upregulation [8] could rely on the availability of clinical molecular imaging and biomarkers. Furthermore, clinical studies will also benefit from noninvasive imaging methods to monitor the transplanted cells and quantify their treatment effects. Imaging techniques such as MRI [2,36], bioluminescence imaging [37], PET [37] and in vivo fluorescence microscopy [36] have been used to track transplanted stem cells in vivo. Most likely, only MRI, PET and possibly CT will find clinical application. Both MRI and PET can be used to monitor the cellular treatment effect [8,38–40].

To date, the three human trials assessing the utility of stem cell therapies for stroke [30,41–44] have utilized stereotactic transplantation into the border zone of the stroke and treated patients at least 6 months from their stroke (between 6 months and 4.5 years [44], between 1 and 6 years [42], and between 1.5 and 10 years [41], respectively). The three Phase I/II studies were designed as cell-replacement therapies and thus offer only limited information on the question of delivery method, timing and microenvironment modulation. A consortium paper on ‘stem cell therapies as an emerging paradigm in stroke’ has discussed several points of future clinical trial design [45]. Several intravenous and two intra-arterial stem cell-treatment trials for stroke are currently underway and will potentially answer some of the questions that have been discussed [101].

Conclusion

Functional recovery observed after transplantation of neural progenitor cells cannot be therapeutically explained by cell replacement alone. Stem cell transplantation promotes CNS repair through alternative mechanisms, including the secretion of neuroprotective growth factors, the induction of synaptic plasticity and angiogenic remodeling, and modulation of deleterious components of the post-ischemic inflammatory response. Knowledge of nonconventional stem cell-mediated therapeutic mechanisms might allow us to achieve tighter control and regulation in vivo, of the multifaceted and potentially divergent ways by which stem cells exhibit their therapeutic effects. It may in fact be the case that the most optimal treatments will require multiple transplantations, of a mosaic of differentiated neural progenitor cells, combined with other somatic stem cell types and/or immune-based therapies. In the future, guiding cell-based therapies utilizing specific receptor expression and molecular secretion patterns may rely on clinical molecular imaging and biomarker availability. To benefit from the full therapeutic potential of stem cells we will have to determine the host microenvironment that is most conducive to cell-induced repair.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.