Cytomorphology of notochordal and chondrocytic cells from the nucleus pulposus: a species comparison


  • Christopher J. Hunter,

    1. McCaig Centre for Joint Injury and Arthritis Research,
    2. Department of Civil Engineering,
    3. Department of Cell Biology & Anatomy, and
    Search for more papers by this author
  • John R. Matyas,

    1. McCaig Centre for Joint Injury and Arthritis Research,
    2. Department of Cell Biology & Anatomy, and
    3. Department of Pathology & Laboratory Medicine, University of Calgary, Canada
    Search for more papers by this author
  • Neil A. Duncan

    Corresponding author
    1. McCaig Centre for Joint Injury and Arthritis Research,
    2. Department of Civil Engineering,
    3. Department of Surgery,
    Search for more papers by this author

Dr N.A. Duncan, Department of Civil Engineering, University of Calgary, 2500 University Drive NW, Calgary, Alberta, Canada T2N 1N4. E:


The nuclei pulposi of the intervertebral discs (IVDs) contain a mixed population of cell types at various stages of maturation. This tissue is formed either by or with the help of cells from the embryonic notochord, which appear to be replaced during development by a population of chondrocyte-like cells of uncertain origin. However, this transition occurs at widely varying times, depending upon the species – or even breed – of the animal being examined. There is considerable debate among spine researchers as to whether the presence of these residual notochordal cells has a significant impact upon IVD degeneration models, and thus which models may best represent the human condition. The present study examines several different species commonly used in lumbar spine investigations to explore the variability of notochordal cells in the IVD.


The intervertebral discs (IVDs) develop from both the embryonic mesenchyme and the notochord (Walmsley, 1953). During embryogenesis, the notochord is surrounded by mesenchymal cells, which synthesize the fibrocartilaginous annulus fibrosus and the bony vertebral bodies. The notochord becomes discontinuous, persisting only inside the primitive annulus fibrosus, where it is believed that the entrapped notochord cells are involved in formation of the primitive nucleus pulposus (Horwitz, 1977; Bell, 1996; Hayes et al. 2001). In some species the notochordal cells (cells presumptively derived from the embryonic notochord) persist through most of adult life, whereas in other species they gradually disappear during aging (Butler, 1989). The cells of the adult nuclei pulposi more closely resemble articular chondrocytes (Walmsley, 1953; Trout et al. 1982a; Maldonado & Oegema, 1992; Errington et al. 1998). It is presently unclear whether this change in cell populations is due to the continued (possibly terminal) differentiation of the notochordal cells into the chondrocytic phenotype, or due to the programmed death (apoptosis) of the resident cells with the subsequent invasion of the nucleus by cells derived from the cartilaginous endplates or annulus fibrosus (Butler, 1989; Walmsley, 1953; Urban, 1996).

Disc disease is typified by changes in the mechanical integrity and biochemical composition of the nucleus (DePalma & Rothman, 1970; Holm, 1996). In humans, the disappearance of notochordal cells precedes the onset of disc degeneration, but it remains unclear if the disappearance of these cells might have a role in initiating disc disease. This uncertainty is due largely to a poor understanding of the function of notochordal cells in normal, healthy discs during development and in young adults.

A number of animal models of IVD degeneration have been developed (Frick et al. 1994; Lotz et al. 1998; Iatridis et al. 1999; Hutton et al. 2000; Kadoya et al. 2001; Kawchuk et al. 2001). However, these studies vary considerably in the status of the notochord and the reported age of loss of notochordal cells in these animals (Table 1). The comparative cell biology of the nucleus pulposus in various species is poorly defined, which seems the prerequisite for testing whether notochordal cells are part of the pathogenesis of disc degeneration, or whether their presence might affect interspecies comparisons of disc degeneration models. We have previously reported observations on the complex cytomorphology and intercellular connections in large clusters of notochordal cells from non-chondrodystrophic canine nuclei pulposi, including the presence of functional gap junctions and large vacuoles with cortical actin filaments (Hunter et al. 2003b, 2004). In the present report, we extend these results to other species by comparing the cytomorphology of the nucleus pulposus from a wide variety of mammals commonly used in laboratory research.

Table 1.  Summary of animal ages in the literature and the current study
SpeciesNo. of animals examined in present studyAge of skeletal maturityAge at loss of notochordal cells (according to literature)Age used in present studyTypical age in previous studiesReferences*
  1. c, chondrodystrophoid (beagles); n/c, non-chondrodystrophoid (mongrels); n/d: no data available.

  2. References: 1, (Hansen, 1952; Hutton et al. 2000; Ganey et al. 2003). 2, (Frick et al. 1994; Hansen, 1952; Maldonado & Oegema, 1992; Katsuura & Hukuda, 1994; Matsuzaki et al. 1996). 3, (Anderson et al. 2002; Smith & Serafini-Fracassini, 1968; Scott et al. 1980; Nomura et al. 2001). 4, (Holm et al. 2004; Kawchuk et al. 2001). 5, (Hansen, 1959; Butler, 1989; Kathmann et al. 2000). 6, (Kadoya et al. 2001). 7, (Adler et al. 1983; Moskowitz et al. 1990; Nishimura & Mochida, 1998; Iatridis et al. 1999; Mente et al. 1999; MacLean et al. 2003). 8, (Ariga et al. 2001; Lotz et al. 1998; Lotz & Chin, 2000; Walsh & Lotz, 2004). 9, (Horwitz, 1977).

Dog (c)212 months12 months2 years2 years1
Dog (n/c)612 months60 months2 & 5 years‘adult’2
Rabbit310 months6 months12 months5–6 weeks3
Pig212 monthsUnknown3 months3–8 months4
Cat124 monthsNever2 monthsn/d5
Ferret1n/dNever2 monthsn/d 
Sheep212 monthsUnknown4 year2–3 years6
Rat62 months12 months2, 16 months5 weeks–16 months7
Mouse24 monthsn/d2 months5–17 weeks8
Human 20 years6–10 yearsn/dn/d9

Materials and methods

Lumbar and caudal IVDs were collected (as secondary use specimens) within 2–5 h from pigs, sheep, dogs, rabbits, cats, ferrets, rats and mice killed for other experiments (Table 1). All animals were pure-bred, with the exception of the mongrel dogs, which were a purpose-bred mix of husky and German shepherd lines (and were therefore presumed to be non-chondrodystrophoid (Hansen, 1952). Lumbar spines (L1–L4) were removed en bloc, the IVDs were opened with a transverse cut, and the nuclei pulposi were removed, placed into Petri dishes and covered with a thin layer of 1% low-melting-point agarose in phosphate-buffered saline, pH 7.2, to immobilize the samples. Tissue-agarose samples were fixed in 10% neutral-buffered formalin. In cases where the boundary between the nucleus pulposus and annulus fibrosus was indistinct – grade III discs – a 6-mm-diameter biopsy punch was used to remove only the central fibrotic region of the disc (Thompson et al. 1990; Bray & Burbidge, 1998).

Agarose-embedded nucleus samples were stained and imaged by confocal microscopy. Samples were incubated in 1.5% Triton X-100 for 5 min, followed by 1.5% normal sheep serum for 30 min, and then stained with 5 µm Oregon Green-labelled phalloidin (Molecular Probes) and counter-stained with 2.5 µm propidium iodide for 10 min to visualize the cell nuclei (except beagle tissue, which was stained with phalloidin only). For the purposes of the present report, we chose to compare the cytomorphology of these cells using actin staining, which can be used to distinguish notochordal and chondrocytic cell types (Hunter et al. 2003b).

Three-dimensional fluorescent image stacks were collected on a Zeiss LSM 510 confocal laser scanning microscope using either a 10× 0.30NA lens or a 63× 1.40 NA oil-immersion objective lens (Carl Zeiss Inc.) immediately following staining. The laser power setting (488-nm excitation line) was adjusted as required for the particular specimen (typically 20–30%). Images were collected as stacks of optical sections (the z-spacing was 10 µm for the 10× objective and 1 µm for the 63× objective) at 512 × 512 pixel resolution with two-line averaging. The images are presented both as individual optical slices or as two-dimensional projections of a three-dimensional stack, created using the Zeiss LSM 5 Image Browser software (the specific stack depth is indicated in the image captions). No further image processing was performed.

Notochordal cells appeared as interconnected clusters of large (> 30 µm in diameter) cells with a prominent actin cytoskeleton and large vacuoles that occupy greater than 25% of the cell volume (Hunter et al. 2003b).


With the exception of the 5-year-old mongrel dogs, 2-year-old beagles and 4-year-old sheep, notochordal cells were present in all animals examined (Figs 1 and 4). Notochordal cells had an actin-filled cytoplasm with a dense cortical meshwork around large vacuoles. The old mongrels, beagles and sheep all exhibited small chondrocyte-like cells that lacked interconnections and large vacuoles (Figs 1 and 4).

Figure 1.

Comparison of cell cytomorphology from 2- and 5-year-old non-chondrodystrophoid mongrel dogs, 2-month-old mice, 2-month-old cats, 2-month-old ferrets, 3-month-old pigs, 2-month-old rats and 1-year-old rabbits. One-micrometre-thick optical slices; scale bar: 20 µm. Red: cell nuclei (propidium iodide), green: actin.

Figure 4.

Comparison of 2-year-old non-chondrodystrophoid mongrel dogs (left) and 2-year-old chondrodystrophoid beagles (right). Three-dimensional reconstruction of 30 1-µm-thick optical slices; scale bar: 20 µm. Red: cell nuclei, green: actin.

The appearance of notochordal cells varied with age; whereas mongrel dogs and sheep appeared to lose their notochordal cells at some point after skeletal maturity (or at least prior to the age of examination), rats retained a small number of notochordal cells well after mature skeletal length was achieved (Fig. 2). It is noteworthy that rats had at least some notochordal cells up until 16 months of age, although the predominant cell type at all ages was a small, chondrocyte-like cell (Fig. 3).

Figure 2.

Comparison of aged animals; 5-year-old non-chondrodystrophoid mongrel dog, 4-year-old sheep and 16-month-old rats. One-micrometre-thick optical slices; scale bar: 20 µm. Red: cell nuclei, green: actin.

Figure 3.

Comparison of chondrocyte-like (left) and notochordal (right) cells from 2-month-old rats; both cell types were found in the rat discs at all ages, but with a substantially reduced prevalence of notochordal cells in the older animal. One-micrometre-thick optical slices; scale bar: 20 µm. Red: cell nuclei, green: actin.

The most striking finding of this study is the contrast between the cells observed in 2-year-old mongrels and pure-bred beagles: the mongrels exhibited large, interconnected clusters of notochordal cells; beagles exhibited small clusters of unconnected chondrocyte-like cells (Fig. 4).


There is much debate among the scientific community regarding the functional significance of notochordal cells of the postnatal IVD. Human notochordal cells are believed to disappear by 10 years of age (Trout et al. 1982b), but many experimental species appear to retain their notochordal cells for an extended period after maturity. Previous studies of IVD cells in the literature (Hansen, 1959; Butler, 1989) have compared tissues that were prepared and examined with different techniques, which has made comparisons among various species difficult. The present study uses a single, reliable method to visualize the cytomorphology of cells in the nucleus pulposus of eight different species for their rigorous comparison. Based upon the results of this small but uniform study, it seems that mature sheep, chondrodystrophoid dogs and rats are among the few species that retain cell populations mimicking those found in adult humans.

Previous reports of notochordal cells in postnatal animals have been inconsistent, and the present findings suggest that in at least in the case of rabbits, the age-related loss of notochordal cells does not occur as suggested in the literature (Table 1). A critical evaluation of the role of notochordal cells in IVD degeneration requires a clear definition of their rate of disappearance in both humans and any species used to model disc degeneration. At a minimum, the careful characterization of the notochordal cell population in animal models is needed both before and after degeneration is initiated. The presumption that notochordal cells are either present or absent should not be assumed based upon the literature, or ignored when reporting results. The use of morphology (cell size, presence of vacuoles, actin cytoskeleton) to define cell populations in the nucleus is very useful (Hunter et al. 2003b), although the development of specific protein and molecular markers will certainly aid the further definition of these cell types and populations (Hunter et al. 2003a).

It is important to note that the functional significance of notochordal cells in an experimental model of degenerative disc disease has never been studied directly. As notochordal cells appear to influence the metabolic behaviour of other cell types (Butler, 1989; Aguiar et al. 1999), it is unclear exactly how the presence of notochordal cells may influence the biology or longevity of chondrocytic (or fibrocytic) cells in animal models of disc degeneration. Regardless, studying the discs of animals that retain their notochordal cells will help to understand the cell biology of nucleus pulposus in humans during disc maturation. Indeed, these studies support the further study of notochordal cells and their role in development, maintenance and degeneration of the IVD.


C.J.H. is supported by the McCaig Centre for Joint Injury and Arthritis Research at the University of Calgary, J.R.M. is a senior scholar of The Arthritis Society, and N.A.D. is supported by a Canada Research Chair in Orthopaedic Bioengineering.