How a growth plate works

Where a bone gets longer, the three stages a cartilage cell passes through on its way to becoming bone, what sets the pace, and why it stops.

Where length is added

A bone gets longer at its growth plates: thin discs of cartilage near each end of a long bone, such as the thigh bone or the shin. Cartilage is made there and then replaced by bone, so the bone lengthens from near its ends rather than stretching along the middle. Almost the whole skeleton is built this way.

The growth plate is a disc of cartilage between the epiphysis at the end of a long bone and the metaphysis of its shaft. For most bones, elongation is driven primarily by chondrogenesis there, through three processes: chondrocyte proliferation, hypertrophy, and secretion of extracellular matrix (J Mol Endocrinol, 2014). The cartilage is then replaced by bone. That sequence, endochondral ossification, is how almost the entire skeleton is generated (Nature, 2003).

The three zones

The plate is arranged in three layers, and each cartilage cell passes through them in order: a reserve layer, a layer where it divides, and a layer where it stops dividing and swells.

The cartilage is organised into a resting zone, a proliferative zone and a hypertrophic zone, and each is a stage a chondrocyte passes through (Endocrinology, 2002).

Resting zone

This is the reserve layer. Its cells are the source of everything below them: in an animal experiment, when the rest of the plate was removed, this layer often grew it back within a week. It also seems to tell the dividing cells which way to line up.

• Contains stem-like cells that give rise to clones of proliferative chondrocytes
• Strip a rabbit growth plate down to its resting zone in vivo and a complete proliferative and hypertrophic zone often regenerates within a week
• It appears to produce a morphogen that aligns those clones into columns parallel to the long axis of the bone, and possibly a second that holds back terminal differentiation nearby (Endocrinology, 2002)

Proliferative zone

Here the cells divide again and again, stacking up in columns that run along the length of the bone. A feedback loop between the top and bottom of the plate decides when each one stops.

• Chondrocytes divide and stack into columns parallel to the long axis of the bone (Endocrinology, 2002)
• When a chondrocyte stops proliferating and commits to hypertrophy is set by the IHH–PTHrP loop

Hypertrophic zone

Here the cells stop dividing and swell to many times their size, and that swelling adds more length than anything else in the plate. At the bottom, blood vessels grow in and the cartilage is replaced with bone. These cells were long assumed to die there, but in mice some survive and become bone-building cells themselves.

• Chondrocytes stop dividing and enlarge. That increase in volume is the largest single contribution to a bone’s lengthening, and it comes in three distinct phases, one of them a swelling that dilutes the cell’s dry mass (Nature, 2013). What drives it is in mTORC1 and chondrocyte volume
• At the base of the plate, blood vessels invade and the osteoblasts arriving with them lay down bone to replace the cartilage (PNAS, 2014). See HIF-1α and VEGF
• Terminally differentiated hypertrophic chondrocytes were long held to die there. Lineage tracing in mice shows they can survive the cartilage-to-bone transition and become osteoblasts and osteocytes that persist into adulthood (PNAS, 2014)

What sets the pace

Growth hormone gets the attention, but it is one of many things controlling a growth plate. Some signals reach the plate through the blood, like growth hormone and thyroid hormone. Some pass between the plate's own cells. Some work inside each cell, and some are part of the cartilage itself. Most of the genes behind ordinary differences in height turn out to act through the plate, not through growth hormone. The library is grouped the same way, with puberty and timing on their own.

The GH–IGF-1 axis was long treated as the system that sets childhood growth, and so short and tall stature. It is one of many. Chondrogenesis at the growth plate, the process that produces height gain, is regulated by hormones, paracrine factors, extracellular matrix molecules and intracellular proteins, and genome-wide association studies attribute most normal variation in height to genes outside the GH–IGF-1 axis that act at the plate (Nat Rev Endocrinol, 2015).

• Hormones reach the plate from elsewhere: growth hormone and IGF-1, thyroid hormone, glucocorticoids
• Paracrine signals pass between its own cells: the IHH–PTHrP loop, and CNP acting through NPR2 against FGFR3
• Intracellular proteins turn those signals into a decision: SOX9 keeps a cell a chondrocyte, RUNX2 commits it towards bone
• Matrix molecules are what the cartilage is made of: aggrecan

The library is grouped the same way, with timing on its own.

Why it stops

A growth plate ages. Its cells divide more slowly over the years and the plate thins, and the cause seems to sit inside the plate itself; one explanation is that the reserve layer has only so many divisions in it. When they run out, the last of the cartilage is replaced by bone in one go and the plate closes. Oestrogen brings this on in both sexes; in animal studies it speeds the ageing up, so the plate runs out sooner. Once a plate has closed, that bone cannot get any longer.

• Growth slows with age mainly because chondrocyte proliferation slows, with structural changes in the plate alongside. This growth plate senescence does not appear to be hormonal or systemic. It is intrinsic to the plate, possibly because the resting zone’s stem-like cells have a finite proliferative capacity that is gradually exhausted (Trends Endocrinol Metab, 2004)
• In mice, resting-zone progenitors are used up during fetal and newborn growth, but once the secondary ossification centre forms they start renewing themselves, and a stem cell niche supplies the plate for a long period (Nature, 2019). How that fits with exhaustion is not settled
• Oestrogen is critical for epiphyseal fusion in both sexes. In juvenile rabbits whose ovaries had been removed, the plate declined steadily in growth rate, proliferation rate, height, number of proliferative and hypertrophic cells, size of terminal hypertrophic cells and column density. Oestradiol accelerated every one of those declines (PNAS, 2001)
• Fusion came when proliferation approached zero, as an abrupt event in which the remaining chondrocytes were replaced by bone. Oestrogen did not appear to trigger the ossification directly. It brought the exhaustion, and so the fusion, forward (PNAS, 2001)
• In humans too, proliferative exhaustion is followed by epiphyseal fusion, in which the growth plate cartilage is replaced completely by bone (Trends Endocrinol Metab, 2004). That ends lengthening of that bone. The human evidence on oestrogen is in Oestrogen

Educational content, not medical advice. Nothing here is a recommendation to take or stop taking anything — talk to a doctor about your own situation.