Effects of suppressing bioavailability of insulin‐like growth factor on age‐associated intervertebral disc degeneration

Abstract Suppression of the insulin‐like growth factor‐1 (IGF‐1) signaling pathway reduces age‐related disorders and increases lifespan across species, making the IGF‐1 pathway a key regulator of aging. Previous in vitro intervertebral disc cell studies have reported the pro‐anabolic effect of exogenously adding IGF‐1 on matrix production. However, the overall effects of suppressing IGF‐1 signaling on age‐related intervertebral disc degeneration (IDD) is not known. Here, the effects of suppressing IGF‐1 signaling on age‐related IDD in vivo were examined using PAPPA −/− mice. These are animals with targeted deletion of pregnancy‐associated plasma protein A (PAPPA), the major protease that cleaves inhibitory IGF binding proteins that control bioavailability of IGF‐1 for cell signaling. Compared to age‐matched wild‐type (Wt) littermates, reduced levels of matrix proteoglycan (PG) and aggrecan were seen in discs of 23‐month old PAPPA −/− mice. Decreased aggrecanolysis and expression of two key catabolic markers, matrix metalloproteinase‐3 and a disintegrin and metalloproteinase with thrombospondin motifs‐4, were also observed in discs of old PAPPA −/− mice compared to Wt littermates. Suppressing IGF‐1 signaling has been implicated to shift cellular metabolism toward maintenance rather than growth and decreasing cellular senescence. Along this line, discs of old PAPPA −/− mice also exhibited lower cellular senescence, assessed by p53 and lamin B1 markers. Collectively, the data reveal complex regulation of disc matrix homeostasis by PAPPA/IGF‐1 signaling during chronologic aging, that is, reduced IGF‐1 bioavailability confers the benefit of decreasing disc cellular senescence and matrix catabolism but also the disadvantage of decreasing disc PG matrix anabolism. This pathway requires further mechanistic elucidation before IGF‐1 could be considered as a therapeutic growth factor for treating IDD.


| INTRODUCTION
Growth factors are peptides working in an autocrine or paracrine manner to regulate cellular activity by binding to cell surface receptors and initiating downstream signaling in target cells. Insulin-like growth factors (IGF) are involved in proliferation and function of nearly every cell, tissue, and organ in the body. IGF-1 is one of two signaling proteins in the IGF family that binds to IGF-1 receptor (IGF-1R). IGF-1 signaling is a well-established modulator of aging. 1 Genetic and pharmacologic suppression of the IGF-1 signaling pathway in numerous animal models has consistently shown that decreasing IGF-1 signaling leads to healthier aging and increased lifespan. [1][2][3][4] Mutations known to impair IGF-1 receptor function, such as natural polymorphisms of the IGF1R gene, and other genetic alterations that affect IGF-1 signaling, have been found in numerous populations of centenarians around the world from Japan to New England. 1,5,6 The exact mechanisms of how the IGF-1 pathway increases longevity in mammals is unknown, highlighting the complexity of this signaling pathway. However, it has been suggested that decreasing IGF-1 signaling reduces mTOR and shifts cellular metabolism from cell growth to cell maintenance and repair activities, decreasing the accumulation of senescent cells. [6][7][8] Despite some inconsistent findings on the role of IGF-1 Among human centenarian and animal model studies, 1,9 there is overwhelming evidence to support that disruptions to the IGF-1 signaling pathway promotes healthy longevity. 1,5,6,8,10 Intervertebral disc degeneration (IDD) can be initiated by an imbalance in disc matrix homeostasis, leading to a pathologic catabolic state where more extracellular matrix is being degraded than what can be produced by disc cells. IDD can lead to biomechanical instability and a range of spine problems, including low back pain. [11][12][13][14][15] Growth factors such as IGF-1 are an attractive biologic therapeutic for treating IDD because they are able to stimulate anabolic production of extracellular matrix. 11,[16][17][18] Early studies on disc cell culture and IGF-1 stimulation showed increased proteoglycan (PG) synthesis of nucleus pulposus (NP) cells in bovine and canine models. 16,17 More recently, IGF-1 has been shown to stimulate both DNA synthesis and downstream signaling of bovine NP and annulus fibrosus (AF) cells in vitro. 19 Additionally, another in vitro study showed the beneficial effect of IGF-1 on decreasing senescent cells in human AF cells exposed to hydrogen peroxide, suggesting a protective effect of IGF-1 against oxidative stress that is typically observed in aging disc. 20 The growing research interest in the role of IGF-1 in IDD has given rise to the idea that IGF-1 can be used as a therapeutic to treat IDD.
The therapeutic potential of IGF-1 for treating IDD, however, is complex and unclear. There is evidence supporting the use of IGF-1 to treat IDD patients in vivo based on reported decreased levels of serum IGF-1 in IDD patients. These studies have correlated low circulating IGF-1 levels with increased level of IDD. 21,22 However, circulating IGF-1 level is also influenced by age, sex, and BMI, all of which influence the risk of developing IDD. [23][24][25][26] Therefore, adjustments for these factors in control vs patient groups are imperative for correctly evaluating IGF-1 levels in control subjects vs patients with IDD. Likewise, circulating IGF-1 levels do not reflect individual tissue concentrations. Free unbound IGF-1 or IGF-1 bioavailability, in the disc specifically, is a better measure of the activity of the GH/IGF-1 pathway in relation to IDD rather than total serum or total protein expression level, which are the methods used almost exclusively in previous papers studying IGF-1 and IDD.
Aging is a well-established major risk factor of IDD as discs appear to undergo age-related degenerative changes earlier in life than other tissues. 12 However, the role of IGF-1 signaling in regulating age-related IDD is unknown. On the one hand, in vitro disc studies suggest that increasing IGF-1 signaling promotes disc health through stimulating matrix production. On the other hand, decreasing IGF-1 signaling in vivo has been shown to mediate age-related disorders and increase lifespan. In light of these conflicting findings, this study investigated how IGF-1 signaling regulates disc aging in vivo using PAPPA −/ − mice. 27 These are animals with targeted deletion of pregnancy-associated plasma protein A (PAPPA), a gene encoding a protease that specifically cleaves three of the IGF binding proteins-IGFBP-2, -4, and -5-that sequester IGF-1 and reduce its bioavailability. 28 The PAPPA amino acid sequence contains five short consensus repeats in its C terminus. Two of these repeats, SCR3 and SCR4, are glycosylated and mediate the binding of PAPPA to cell surface. 28 Because of its location at the cell surface, PAPPA is responsible for the local IGF-1 bioavailability and IGF-1 signaling activity more than the circulating level of IGF-1. The mean lifespan for PAPPA −/− mice is approximately 40% greater than wild-type (Wt) mice. 1,4,28,29 In this study, the effects of reduced IGF-1 bioavailability on disc matrix homeostasis and age-

| Western blotting
Protein from caudal discs of seven mice were extracted using Tissue

| Histology
Isolated lumbar disc tissues from three PAPPA −/− and Wt mice were fixed and decalcified in Decacifier I solution (Leica 3800440) at 4 C overnight. All tissues were dehydrated through a graded alcohol series and then embedded in paraffin (Tissue Tek processor and Leica Embedder) and cut into three 4-μm thick sections in the coronal plane.
The sections were stained with safranin-O and fast green dyes by standard procedures and photographed under ×40 to ×200 magnification (Nikon Eclipse E 800). 2.6 | DMMB assay for total glycosaminoglycan PG content of disc NP from Wt and PAPPA −/− mice was measured using 1,9-dimethylmethylene blue buffer (DMMB) assay for total glycosaminoglycan (GAG). NP tissue from three lumbar discs of each mouse was isolated using the dissecting microscope and stored at

| Statistical analysis
Data are expressed as the mean ± SD of three independent samples.
For Western blot and total GAG analysis, 95% confidence intervals were calculated to determine statistical significance at P < .05 level.
The confidence intervals were calculated based on the t-distribution because of the small sample size. 33  These findings are consistent with the previous studies that reported the pro-anabolic effects of IGF-1 on matrix production in disc cell culture. 16,34,35 3.2 | Aged PAPPA −/− mice had less fragmented aggrecan in disc Since disc matrix homeostasis is determined by anabolic and catabolic matrix metabolism, decreased disc PG content in PAPPA −/− mice could be due to enhanced PG catabolism. 12 With aging, disc cells gradually lose their capacity to synthesize new PG to replace what is degraded and lost over time, resulting in a net loss of disc PG. 12 A schematic of mouse PG is pictured in Figure 2A   Specifically, the amount of ADAMTS cleaved and MMP cleaved aggrecan in aged PAPPA −/− was about 2-fold less than that seen in the Wt littermates. Therefore, decreasing IGF-1 signaling reduces age-related disc aggrecanolysis. This is consistent with findings that decreasing IGF-1 signaling leads to healthier aging and less agerelated disorders. 3.4 | Aged PAPPA −/− mice had increased disc lamin B1 and decreased p53 expression, indicating reduced disc cellular senescence Senescent disc cells have recently been reported to exhibit a catabolic phenotype, called senescence-associated secretory phenotype, and have imbalanced matrix homeostasis. 36,37 There is a strong correlation between the level of disc cellular senescence and grade of disc degeneration, 36 and growing evidence suggests senescent disc cells are a key driver of age-related IDD. 36,37 Therefore, the role of IGF-1 in disc cellular senescence was investigated in aged PAPPA −/− mice. When a healthy cell becomes senescent, it develops certain characteristic morphological changes, which include enlarged and often irregular nuclei.

| Aged
Lamin B1 is a scaffolding component of the nuclear envelope, and its expression has been shown to be inversely correlated with cellular senescence. expression. 36 Hence, greater p53 expression in a cell indicates greater likelihood of it progressing toward senescence. 36 Compared to aged Wt mice, aged PAPPA −/− mice had less p53 protein expression, but this difference was not statistically significant ( Figure 4C). Taken together, these data suggest aged PAPPA −/− mice have lower levels of disc cellular senescence compared to old Wt littermates.

| DISCUSSION
Aging is a key contributor to IDD as disc matrix homeostasis becomes PAPPA −/− mice exhibit reduced disc matrix catabolism (Figure 2A, B). In the center of the disc, the gelatinous NP loses water content with age due to proteolytic degradation of aggrecan, the major PG in the disc. 12  These findings support the idea that suppressed IGF-1 signaling in PAPPA −/− mice leads to reduced disc cellular senescence which results in an overall decrease in disc aggrecanolysis. On the contrary, Gruber and coworkers reported that supplementing IGF-1 in cell culture rescues AF cells from oxidative stress-induced premature senescence. 20 These conflicting findings in IGF-1 regulation of disc cellular senescence might be due to the differences between their in vitro condition and our in vivo model as well as the potential metabolic differences between AF and NP cells in responding to stress. Additionally, IGF-1 might also influence disc aggrecanolysis through the AKT pathway which has been implicated in regulating MMP expression, 40 although confirmatory studies are needed.
The PG content in the disc of 23-month-old PAPPA −/− mice is less than that seen in 23-month-old Wt mice, as demonstrated by safranin-O/fast green histology, DMMB assay for total GAG, and aggrecan IHC ( Figure 1). This finding is consistent with the pro-anabolic function of IGF-1 as a growth factor, that is, less IGF-1 signaling in PAPPA −/− mice leads to less matrix synthesis during development and less overall disc PG matrix. Previous work demonstrated IGF-1 as an anabolic agent for discs able to promote both PG synthesis and disc cell proliferation. 11,16,17,19,34,35,41 Concurrently, IGF-1 also stimulated matrix production in chondrocytes. Loeser and coworkers reported significant increases in PG synthesis by chondrocytes derived from knees of osteoarthritis patients that were cultured in alginate beads with exogenous IGF-1 (100-1000 ng/mL). 42 Increased PG synthesis in bovine monolayer articular chondrocytes and explant models stimulated with IGF-1 were also documented. 34,35,41,43 Given the pro-anabolic function of IGF-1 on cells, it is possible that the low disc PG in PAPPA −/− mice is due to lack of anabolic stimulation from IGF-1 since birth. Another possibility is PAPPA −/− mice have a decreased ability to synthesize new disc PG matrix as it is lost with age due to increased PG catabolism. The metabolic effects of IGF-1 on disc PG matrix homeostasis.
IGF-1 has been suggested as a therapeutic growth factor to stimulate PG synthesis to counteract the loss of disc PG with age and degeneration. However, the role of IGF-1 in regulating disc matrix homeostasis is more complex than merely up regulating matrix production as reported in vitro studies. Our study comparing Wt and PAPPA −/ − mice suggest that IGF-1 signaling is also required for disc PG production in vivo, but prolonged signaling of IGF-1 increases the risk of cellular senescence and aggrecanolysis, as seen in aged Wt mice.
Furthermore, in order to synthesize new matrix following stimulation with IGF-1, cells in the degenerated or aged disc must have the nutrients required to generate energy to perform these functions. However, the disc nutrient supply is decreased in most aged and degenerated discs due to end plate calcification. 44 ing cellular metabolism and disc matrix homeostasis, more studies are needed before this signaling pathway can be considered as a therapeutic target to treat IDD. It should be noted that IGF-1 action on the discs in our global PAPPA knockout mice could be due to its direct effects, that is, IGF-1 entering disc tissue and stimulating disc cells, or indirect effects through impacting other body tissues which then affecting disc tissue via paracrine or endocrine pathways; it is not possible to differentiate between these two mechanisms, adding to the complexity of IGF-1 action on disc biology that would require further investigation to elucidate. Nevertheless, this study suggests that decreasing global IGF-1 bioavailability is overall beneficial for slowing down age-associated IDD, but more temporal and spatial studies aimed at upregulating and downregulating IGF-1 signaling are needed to further elucidate the mechanisms by which it impacts disc matrix homeostasis and ageassociated IDD. and isolation of spines for experimental analysis. This project was