The I-type cell was initially identified in cultured cell lines because it appeared "intermediate" in morphology between N and S cells. It exhibits morphological features of both N-type cells (short neurite-like cell processes and growth to high saturation densities) and S-type cells (strong adhesion to and extensive migration over the substrate) (Biedler et al. 1997). These cells also express proteins of both differentiation pathways - noradrenergic biosynthetic enzymes, granins (CgA and SgII), and neurofilament proteins of neuroblasts as well as S cell proteins vimentin, EGF receptor, and CD44. Examples of I-type cells include the cell lines GOTO, NUB-7, BE(2)-C, SH-IN, and LA-N-2 (Biedler et al. 1988,1997; Ross et al. 1995,2002).
Continuing research indicates that this cell represents a unique cell type within the NB repertoire. Its ability to generate daughter cells with the same phe-notype (self-renewal) and to differentiate bidirection-
ally along either neuroblastic or Schwann/glial pathways suggests that it is a neural crest cancer stem cell.
First demonstrated for BE(2)-C cell clones and subsequently for other I-type cell lines, I-type cells become neuroblastic when treated for 7-14 days with retinoic acid (RA), but differentiate into S-type cells following treatment with BUdR (Ross et al. 1995). Unlike N (or S) cells, I cells retain the ability to convert to two distinctly different cell lineages.
The most provocative finding regarding the NB I-type stem cell is its malignant potential. As a group, these stem cells are more malignant than neuroblas-tic variants; they have four- to fivefold higher colony-forming efficiencies in soft agar than N cells and have an over sixfold greater capacity to form tumors in nude mice (Ross et al. 2003; Spengler et al. 1986; Walton et al. 2004). Moreover, phenotype rather than MYCN amplification/overexpression determines malignancy; e.g., NB I-type stem cells lacking MYCN amplification are more tumorigenic than N-type cells which contain >150-fold amplified genes. Thus, research to date on cell lines suggests that the I-like stem cell could be the truly tumorigenic cell component of NB tumors.
Malignant stem cells in tumors could exert a significant negative effect on prognosis and long-term survival; however, distinguishing putative stem cells from those with a neuroblastic phenotype in tumor sections by routine hematoxylin-eosin analysis is difficult, if not impossible. To specifically search for I-like cells, tumor sections immunostained conjointly with antibodies specific for N or S cells were examined for the presence and frequency of doublelabeled cells (Ross et al. 2003). In preliminary analyses, doubly labeled I-like cells were present in all tumors (Fig. 6.2). When the tumors were grouped as either good risk (typically local regional or stage 4s) or poor risk (stages 3 or 4), the frequency of I-like cells was significantly higher (~ fivefold) in the latter group (B.A. Spengler, personal communication). The characterization of this previously unnoticed NB cell type in cell lines and its potential role in refractory high-risk tumors may have identified an important new target for experimental therapeutics.
Neuroblastoma tumor section stained conjointly for expression of S100A6 (an S cell marker; red) and neurofilament 160 (an N-cell marker; gray). Examples of I-like cells expressing both proteins are indicated by filled arrows, whereas N cells expressing only neurofilament protein are denoted by open arrows
Was this article helpful?