Cellular changes in the enteric nervous system during ageing

Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 Q1 15 16 17 18 19 20 21Q5 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Contents lists available at SciVerse ScienceDirect

Developmental Biology journal homepage: www.elsevier.com/locate/developmentalbiology

Cellular changes in the enteric nervous system during ageing M. Jill Saffrey n Department Life, Health & Chemical Sciences, Open University, Walton Hall, Milton Keynes MK7 6AA, United Kingdom

art ic l e i nf o

Keywords: Aging Neurodegeneration Caloric restriction Microbiota Inflammation Gut–brain axis Oxidative stress Cell senescence Neurotrophic factors

a b s t r a c t The intrinsic neurons of the gut, enteric neurons, have an essential role in gastrointestinal functions. The enteric nervous system is plastic and continues to undergo changes throughout life, as the gut grows and responds to dietary and other environmental changes. Detailed analysis of changes in the ENS during ageing suggests that enteric neurons are more vulnerable to age-related degeneration and cell death than neurons in other parts of the nervous system, although there is considerable variation in the extent and time course of age-related enteric neuronal loss reported in different studies. Specific neuronal subpopulations, particularly cholinergic myenteric neurons, may be more vulnerable than others to age-associated loss or damage. Enteric degeneration and other age-related neuronal changes may contribute to gastrointestinal dysfunction that is common in the elderly population. Evidence suggests that caloric restriction protects against age-associated loss of enteric neurons, but recent advances in the understanding of the effects of the microbiota and the complex interactions between enteric ganglion cells, mucosal immune system and intestinal epithelium indicate that other factors may well influence ageing of enteric neurons. Much remains to be understood about the mechanisms of neuronal loss and damage in the gut, although there is evidence that reactive oxygen species, neurotrophic factor dysregulation and/or activation of a senescence associated phenotype may be involved. To date, there is no evidence for ongoing neurogenesis that might replace dying neurons in the ageing gut, although small local sites of neurogenesis would be difficult to detect. Finally, despite the considerable evidence for enteric neurodegeneration during ageing, and evidence for some physiological changes in animal models, the ageing gut appears to maintain its function remarkably well in animals that exhibit major neuronal loss, indicating that the ENS has considerable functional reserve. & 2013 Published by Elsevier Inc.

Introduction It is established that the mammalian enteric nervous system (ENS) undergoes developmental changes postnatally (e.g. de Vries et al., 2010; Foong et al., 2012). It is also clear that the cells of the adult ENS exhibit plasticity and continue to undergo changes later in life, due both to the highly dynamic nature of the gastrointestinal (GI) system, and to cellular changes that take place as part of the process of ageing. Throughout life, particularly that of longer-lived animals, the gastrointestinal (GI) system is exposed to dietary variation and also to changes in the microbiota, to disease, inflammation and, in humans, medication. The properties of intestinal cells may be affected as a result of these changing conditions in the gastrointestinal environment. In addition, in long-lived animals, and in animals living in protected laboratory environments, cellular changes that occur simply as a result of ageing will affect the cells of the gut, although different cell types, such as muscle, epithelial cells and neurons, are likely to be affected in different ways. These

n

Fax: þ44 1908 654267. E-mail address: [email protected]

age-associated cellular changes may impact upon specialised functions, and hence affect GI physiology. There is also evidence from animal models that external factors, including diet, can alter ‘normal’ (i.e. non-pathological) ageing of the ENS. Understanding the contribution made by normal cellular ageing to the age-associated physiological changes that take place in the gut is important, because gastrointestinal disorders increase in prevalence in older people, and some conditions, such as constipation (see Gallagher and O’Mahony, 2009 Rao and Go, 2010), faecal incontinence (see Chatoor et al., 2007) and gastric reflux (see Franceschi et al., 2009) are very common among the elderly population. These conditions have a serious impact on the quality of life of the elderly, affecting their independence, and also have a significant impact on healthcare costs. Age-associated enteric neurodegeneration (see Bitar et al., 2011; Camilleri et al., 2008; Phillips and Powley, 2007; Saffrey, 2004; Wade and Cowen, 2004; Wiskur and Greenwood-Van Meerveld, 2010) and ageing of the other cells of the gut, such as smooth muscle cells (see Bitar et al., 2011; Bitar and Patil, 2004) are believed to be major contributory factors to the these age-associated GI disorders. Here, the changes that occur in the cells of the ENS during ageing and the implications of these changes for GI function are reviewed. Ageing of intestinal smooth muscle (see Bitar et al.,

0012-1606/$ - see front matter & 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.ydbio.2013.03.015

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

2011; Bitar and Patil, 2004), endocrine cells (see Sandstrom and El-Salhy, 2001) and intestinal epithelial stem cells (Kirkwood, 2004) are reviewed elsewhere. Very few studies have investigated changes in the properties or numbers of Interstitial cells of Cajal and Fibroblast-like cells during ageing. Changes in GI physiology in the ageing human population are well-known, but identifying the contribution made by cellular ageing to these changes is difficult, because of the many confounding factors that may also influence GI function in humans. For example, co-morbidity (e.g. vascular disease), medications (e.g. anti-cholinergic medications), change in diet and immobility may all influence GI physiology (e.g. see Rao and Go, 2010). The complications associated with discrimination between changes that are genuinely due to ageing, and those due to extraneous conditions, such as those listed above however, make carefullymaintained laboratory animals a valuable resource for studies of GI ageing; confounding factors such as variations in diet and medication are eliminated or reduced to a minimum. Most work on ENS ageing has been carried out on the myenteric plexus, and analysis of age-related neuronal changes has focused predominantly on changes in the numbers of neuronal cell bodies, although some studies of the density of the myenteric plexus network and nerve fibre density in smooth muscle have also been performed. Cellular and biochemical properties that have been reported to be changed during ageing of other cell types and markers of cellular senescence have also been investigated in enteric neurons and are discussed here. The changes that occur in GI neurophysiology of animals during ageing are being increasingly studied, although the GI changes that are common in ageing humans, such as constipation and incontinence, have not been widely looked for in the animals that have been used for the study of ENS ageing. Nevertheless recent evidence indicates that ageing mice do exhibit changes in the size and frequency of stools (Patel et al., 2012), indicating that they undergo changes that are at least similar to some of the exhibited by ageing humans. Physiological and pharmacological studies of the ageing ENS are not discussed here.

Plasticity of the ENS during ageing The ENS is at all times a highly plastic system in terms of its physical characteristics, because the shape of enteric ganglia changes during and between the distensions that occur during peristalsis and other gut movements (Gabella, 1990). However, in addition to these continual dynamic changes, the size of the GI tract changes during the lifespan. The intestine undergoes considerable growth during adult life. For example, Gabella (1989) reported that the length of the guinea-pig small intestine increases by 27%, and its circumference by 30% during ageing. This growth results in an increase of 70% of serosal area under the conditions studied. Interestingly, the maximum dimensions were measured in middle-aged animals (Gabella, 1989). A number of others have also measured an increase in length and/or circumference with age (e.g. Choi et al., 2008 Gamage et al., in press Phillips and Powley, 2001; Peck et al., 2009). An increase in circumference may be due to changes in the thickness of the mucosa or muscularis externa, or both. Increases in the thickness of the muscle layers during ageing have been reported in several studies (e.g. Peck et al., 2009; Southwell et al., 2010), but not in others (e.g. Marese et al., 2007). With respect to the mucosa, it has been reported that there are changes in villus width and depth (Drozdowski and Thomson, 2006). Increases in the length and circumference of the intestine and changes in the volume of the different layers of the gut would be expected to have an impact on the arrangement of the enteric plexuses, and potentially on the density of nerve fibres

in different layers. It is therefore likely that continued growth, and possibly some rearrangement, of enteric nerve fibres occurs throughout periods of gut growth, which take place during a considerable period of the lifespan. During such periods of gut growth, some types of myenteric neuron, such as intrinsic sensory neurons, and those interneurons that project longer distances in the plexus, would need to increase the length of their processes to maintain contact with their target cells. Such a response may not be needed by all myenteric neuronal subpopulations however, since an increase in smooth muscle volume for example, could result from an increase in the size or number of muscle cells, and since they are well-coupled (Hoyle and Burnstock, 1989), effective transmission could perhaps continue even if the density of fibres decreased. Nevertheless, it would seem that the ENS, and enteric neurons in particular, must remain plastic during adulthood.

Changes in the gross morphology of the myenteric plexus during ageing Changes in the general morphology of the enteric ganglia during ageing have been described in some studies. Gabella (1989) has reported that there is greater separation between myenteric ganglia in old guinea-pigs, and that myenteric neurons appear to be less densely packed within the ganglia. Abalo et al. (2005) described a reduction in size of myenteric ganglia of guinea-pig ileum with age, and that myenteric ganglia occupied a reduced area. No change in the number of ganglia per unit area during ageing was measured in the same study. A change in the size and the appearance of myenteric ganglia with age has also been described in humans (Hanani et al., 2004); in samples from older individuals the overall ganglionic area was found to be larger, and gaps or spaces were observed within the ganglia. Ganglia from older individuals had increased numbers of spaces, and also an increased proportion of ganglia with spaces, in both the ileum and colon. The extent of these changes correlated with advancing age, and was greater in the colon than the ileum. No difference between males and females was measured. The authors speculated that the spaces could occur due to a stretch effect resulting from gut growth, but pointed out that this was unlikely, because normal ganglia, without spaces, were seen adjacent to abnormal ganglia.

Changes in the general morphology of enteric neurons during ageing The size, general shape and distribution of myenteric neurons have also been reported to change during ageing. Gabella (1989) reported that myenteric neurons were smaller in the ageing guinea-pig small intestine and Gomes et al. (1997) also measured a reduction in myenteric neuronal perikaryon area in ageing human colon. Santer and Baker (1988); however, found no change in the size of myenteric neurons in the rat small or large intestine, while others have measured an increase in neuronal cell body size in older rats (Phillips et al., 2003). Some of the discrepancies between these results may be due to the methods used (see section on challenges of analysis of changes of neuronal numbers in the ageing gut). The shape of neurons in the ageing guinea-pig gut has been reported to be different from that in younger animals; neurons in ageing animals were found to have a ‘horny’ profile (Gabella, 1989). Some single neurons, which were larger and with a smooth profile, were found to be present at the edges of myenteric ganglia in old animals in the same study. Such peripherally-located

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

3

Table 1 Changes in myenteric neuronal numbers during ageing. Species & strain

Regions studied

Ages studied

Populations studied

Diet

Loss (notes)

Ref.

RAT Wistar ♂

Jej, il, prox col, rect

6, 24m

Total (NADHd)/area

N/S

Santer and Baker (1988)

SI Ileum

4, 24, 30m 4, 6, 24m

Nitrergic (NADPHd) Nitrergic (NADPHd) as proportion of total (PGP 9.5) /ganglion

AL

Total (NADHd and PGP CR 9.5), Sub popns: NADPHd, ChAT-, SP-, ENK-, VIP-, NPY-, SOM-IR / area Total (PGP 9.5, NADHd), AL, CR DCR Sub popns: ChAT-IR, NADPHd /area nNOS/ganglion, NOS protein & mRNA, PGP 9.5 protein levels

∼40% loss in SI & rectum, ∼60% loss in colon [Changes in intestinal dimensions not corrected for] 15% loss at 24m, also at 30m Increased proportion of nitrergic neurons in colon, but not in ileum. [analysis independent of intestinal dimensions] 15% loss NADHd neurons but no change in PGP 9.5-IR neurons, or NADPHd positive or neurons expressing any other markers

Wistar ♂

Colon

Santer (1994) Belai et al. (1995a)

Johnson et al. (1998)

Sprague-Dawley

Ileum

4, 24m

Sprague-Dawley ♂

Ileum

4–6, 16, 20, 24, 30m

F1

Mid-colon

4–8, 22–28m

Fischer 344 ♂

Stomach SI regions

3,12, 21,24, 27m

Total (Cuprolinic Blue)/ area

AL

3, 24m

Total (Cuprolinic Blue) Nitrergic (NADPHd) /area Total (PGP 9.5)

AL

Total 3–7% loss in stomach, 10–27% Phillips et al. (2003) loss in SI regions, 37–41% loss in LI.

AL

Wu et al. (2003)

Nitrergic (NADPHd & nNOS) /ganglion Total neurons (HuC/D)

Total neuronal loss ∼27%; Strain difference in nitrergic changes, reduction in Sprague-Dawley, no changein Wistar [analysis independent of intestinal dimensions]

AL

7–37% neuronal loss / gang area; 18–48% loss /ganglion, dependent on GI region (glial losses also observed, see text)

Phillips et al. (2004)

Al, CR

PGP9.5 (24m): AL 51% loss, CR 22% loss Hu C/D (17m): AL 50% loss, CR no data Calretinin (17m) AL& CR show similar non-significant loss (∼30%) Calbindin (17m) AL; 62% loss, CR no loss Reduction in neuronal density in younger animals because of gut growth, in older animals because of neuronal loss (and see text) [Changes in intestinal dimensions not corrected for] AL 29% total loss, CR 17% loss Increase (20%) in NADPHd neuronal numbers in AL-fed ageing animals, smaller increase in CR

Thrasivoulou et al. (2006)

Fischer 344 ♂

Sprague-Dawley &Wistar

LI regions Stomach SI regions LI regions Oesophagus

3–4.5,

18–20m

Fischer 344 ♂

SI regions

5–6, 26m

LI regions

Sprague-Dawley ♂

Ileum

Total glia (S100)/gang &/gang area 6, 12, 13, 17, 24m

Total (PGP9.5, Hu C/D) n

Sub pop s: Calretinin, calbindin /area

Wistar ♂

Duodenum

21, 60, 90, 210, Total (Geimsa & Myosin AL 345 & 428d V-IR) /area

Wistar ♂

Duodenum

6, 18m

Total (Geimsa) Nitrergic (NADPHd)/ area

Small intestine

3–4, 26–30m

Ileum

1–3, 8–10, 22–26m

Total (NADH diaphorase) Total (HuC/D) calretinin/ganglionic area & /area

AL, CR*

∼50% loss of total, 64% loss cholinergic neurons in AL-fed only. Nitrergic neurons spared. Significant reduction in number of nNOS-IR neurons per ganglion. Reduction in NOS protein & mRNA. No change in PGP 9.5 levels. [analysis independent of intestinal dimensions] Differential loss in regions. None in antrum & corpus at 24m, late 38% loss in forestomach. 17–31% losses in SI, 38–39% in LI

Cowen et al. (2000)

Takahashi et al. (2000)

Phillips and Powley (2001)

Marese et al. (2007)

da Silva Porto et al. (2012)

GUINEA-PIG

Dunkin-Hartley ♀

2–3, 8–10, 22–26m

Total HuC/D

40–60% loss total N/S

Gabella (1989)

∼30% loss total, ∼50% loss calretinin Abalo et al. (2005) [Changes in intestinal dimensions not corrected for] Total – decreased in 22–26m cf 8–10m

Abalo et al. (2007)

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

Q3

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Table 1 (continued ) Species & strain

Regions studied

Ages studied

Populations studied

Diet

Sub popns: nNOS, calbindin, calretinin, neurofilament (double labelling) /ganglionic area

Mid -colon

2w, 6–8m, 24–32m

Total (HuC/D)/area & gang area

MOUSE NMRI:

Stomach (antrum)

1, 3, 12, 24m

Total (PGP 9.5)

Bom

Duodenum

/ganglion

Colon

Sections

C57/BL6

Distal colon

GERBIL

Duodenum

3–4, 12–13, 18–19, 24– 25m 1.5, 3, 6, 12, 24m

Total (HuC/D) Calbindin-, nNOS-IR /area Total (Geimsa)

AL

AL

AL

HUMAN

Oesophagus

20–40 , 4 70y Total (Geimsa stain) / area

N/A

♀♂

Small intestine

20–40, 69– 76 yr

Total (Geimsa stain)/ area

N/A

♀♂

Colon

20–35 , 4 65 yr

Total (Geimsa stain) /area

N/A

Ileum

42–71, 78– 86 yr

N/A

♂♀

Ileum Colon

10d – 92 yr (n¼ 168)

Nitrergic (NADPHd)& calretinin proportion of total PGP9.5-IR Vital dye:4-Di-2-ASP* Nitrergic (NADPHd) /ganglion

♂♀

Colon

33–99 yr

Sections Total (HuC/D, PGP9.5),

N/A

/ganglion and /unit length

Ref.

Calbindin, Calretinin and NFT alone, no change at 22–26m,

Calbindin

ChAT, NOS

Loss (notes)

Calretinin þNTF(co-localised) and NOS alone, loss in 22–26m, [Changes in intestinal dimensions not corrected for] 19% (non-significant) loss

Peck et al. (2009)

∼40% loss between 3 and 12m, no change 12–24m [Changes in intestinal dimensions not corrected for]

El-Salhy et al. (1999)

No change in any population after individual correction for change in LI dimensions, and stretch 17% loss total neurons, 5% loss calbindin [Changes in intestinal dimensions not corrected for] 22–62% loss [Changes in intestinal dimensions not corrected for] 34% loss [corrected for changes in gut diameter] 37% loss [Changes in intestinal dimensions not corrected for] Nitrergic and calretinin-IR increase as proportion of total in aged samples. [analysis independent of intestinal dimensions] Total not counted (ganglion morphology studied).Nonsignificant increase in proportion NADPHd/ganglion with age HuC/D & ChaT neuronal number negatively correlated with increasing age. Non-significant NOS-IR decrease neurons ∼38% total loss between ages of 30 and 60 yr [Changes in intestinal dimensions not corrected for]

Gamage et al. (in press) Choi et al. (2008)

Meciano Filho et al. (1995) de Souza et al. (1993)

Gomes et al. (1997)

Belai and Burnstock (1999)

Hanani et al. (2004)

Bernard et al. (2009)

Key: AL¼ Ad libitum; ChAT ¼choline acetyltransferase; CR ¼ calorically restricted; ENK ¼enkephalin; IR ¼ immunoreactive; PGP¼ protein gene product; N/A¼ not applicable (human study), NADH ¼nicotinamide adenine dinucleotide; NADPH ¼ nicotinamide adenine dinucleotide phosphate; N/S¼ not stated; NOS¼nitric oxide synthase; NPY¼ neuropeptide Y; SOM¼ somatostatin; SP¼ substance P; VIP ¼ vasoactive intestinal polypeptide.

neurons have also been described in ageing human myenteric ganglia (Hanani et al., 2004) and Abalo et al. (2005) have also reported an increased occurrence of extraganglionic neurons in the myenteric plexus of ageing guinea-pig ileum.

Changes in the number of enteric neurons during ageing Myenteric neurons The first work to quantify possible changes in myenteric neuronal numbers during ageing was by Santer and Baker (1988), Santer and Baker (1988) on rats, and Gabella (1989), on guinea-pigs. Since then, analysis of changes in myenteric neuronal numbers has been performed in several additional species. Most studies have been performed on rats, but guinea-pigs, mice, gerbils and humans have also been investigated. The majority of research

has been carried out on the small and large intestines, but some groups have investigated age-related changes in myenteric neuronal numbers in the oesophagus, stomach and rectum. The results of studies of changes in myenteric neuronal numbers are summarised in Table 1. A reduction in the number of myenteric neurons during ageing has been described in most, but not in all studies. Moreover, the proportion of the total population of neurons that are reported to be lost varies considerably, for example in ad libitum (AL) fed rat ileum, losses of between 17% (Phillips and Powley, 2001) and 50% (Cowen et al., 2000) have been described. Similar ranges of neuronal losses are reported in three different strains of rats (Wistar, Sprague Dawley and Fischer 344, see Table 1). Myenteric neuronal loss has been found to vary in different regions of the rat gut, being greater in the large intestine than in the small intestine (Phillips et al., 2003; Phillips and Powley, 2001, 2007; Santer and Baker, 1988). Losses in rat stomach have also been found to vary

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

according to region; in the antrum and corpus of 24 month old animals losses were either very low (3–7%, Phillips et al., 2003) or absent (Phillips and Powley, 2001) but in the forestomach, although no losses were detected at 24 months, a 38% loss was reached by 27 months, but no such late loss was seen in the antrum or corpus (Phillips and Powley, 2001). The time course of myenteric neuronal loss during ageing has also been investigated, most studies showing that it begins at ‘middle-age’, but continues for variable periods thereafter. For example, Phillips and Powley (2001) found only a relatively small reduction in AL-fed 12 month-old Fischer 344 rats compared to 3 month-old controls (10–15%, depending upon the GI region studied), but that losses continued, in a linear manner in most gut regions, at least until 21 months of age. In contrast others, working on AL-fed Sprague-Dawley rats, found that neuronal losses began at about 13 months, but were complete by 16 months of age (Cowen et al., 2000; Thrasivoulou et al., 2006). Thus the onset and rate of neuronal loss may be variable in different cohorts of ageing animals, and may, of course be affected by differences in animal husbandry in the different studies (see Section Challenges for the analysis of changes of neuronal numbers in the ageing gut, below). In this context, it is important to note that caloric restriction has been found to have a marked effect on age-associated neuronal losses in the ENS. The loss of myenteric neurons observed in AL-fed rats has been reported to be eliminated (Cowen et al., 2000; Johnson et al., 1998) or significantly reduced (da Silva Porto et al., 2012; Thrasivoulou et al., 2006) by caloric restriction introduced at 6 months of age. Subpopulations of myenteric neurons The subpopulations of myenteric neurons that are lost during ageing have been investigated by a number of groups. Several studies have presented evidence that age-associated myenteric neuronal loss is specific to the cholinergic subpopulation in rats (e.g. Cowen et al., 2000; Phillips et al., 2003) and in humans (Bernard et al., 2009), and that nitrergic neurons are maintained (e.g. Belai et al., 1995; Cowen et al., 2000; Gamage et al., in press). In some cases the loss of cholinergic neurons has been indirectly deduced because the proportion of NADPH diaphorase (NADPHd) neurons have been found to increase in older animals (e.g. Phillips et al., 2003), although a direct reduction in ChAT-IR neurons has also been reported (Bernard et al., 2009; Cowen et al., 2000). Evidence about the fate of nitrergic neurons during ageing, however, is conflicting, because a loss of nitrergic neurons has been measured during ageing in some studies (Abalo et al., 2007; Santer, 1994; Takahashi et al., 2000). In one study of myenteric neurons in the oesophagus, numbers of nitrergic neurons were found to be reduced in Sprague Dawley rats, but maintained in Wistar rats (Wu et al., 2003). Swollen and dystrophic processes of both NADPHd (Phillips et al., 2003) and nNOS-IR (Gamage et al., in press) nerve fibres however, have been observed in ageing rat and mouse myenteric plexus wholemounts respectively, indicating that nitrergic neurons may undergo degenerative changes during ageing, even if they do not die in detectable numbers at the ages studied. Few studies have looked in detail at other neuronal subpopulations, although no loss of any subpopulation of a range studied (including SP, VIP and somatostatin-IR neurons) was measured in ageing rats fed a calorically-restricted diet (Johnson et al., 1998, see Table 1). The possibility that a subpopulation of cholinergic myenteric neurons, such as intrinsic sensory, motoneurons or interneurons, is selectively vulnerable to age-related cell death has also not been studied in detail. Calbindin-IR myenteric neurons were found to be reduced in number in AL-fed ageing animals (Thrasivoulou et al., 2006), but no age-related change in

5

calbindin-IR neurons was detected in guinea-pig ileum (Abalo et al., 2007) or mouse distal colon (Gamage et al., in press). Calretinin-IR neurons were reported to be lost in Guinea-pig ileum (Abalo et al., 2007), but not in mouse distal colon (Gamage et al., in press), or rat ileum (despite a non-significant 30% loss in this rat tissue Thrasivoulou et al., 2006). The possible reasons for discrepancy between results from different studies, and for possible selective vulnerability of particular neuronal populations to degeneration or are considered in later sections of this review. Submucosal neurons Changes in the numbers of submucous plexus neurons during ageing have been little studied in comparison to those of myenteric neurons. Changes in the total population (identified by Cuprolinic blue staining) of submucosal neurons during ageing has been studied in the rat proximal and distal colon (Phillips et al., 2007). In this study, which analysed total neurons in samples from 6, 12, 18, 24 and 27 month old animals, a linear loss of submucosal neurons was seen from 12 months of age. The loss was greater in the distal colon than the proximal colon (38% compared to 27% at 27 months), and the maximal loss was also reached later in the distal colon (27 months compared to 24 months). The mean number of neurons per ganglion also decreased during ageing. El-Salhy et al. (1999) also described a reduction in the number of neurons per submucosal ganglion in the antrum, duodenum and colon of 12 and 24 months compared to 3 month-old mice. Bernard et al. (2009); however, did not detect an age-associated reduction in the number of HuC/D-IR neurons per submucosal ganglion in human colon samples. Age-related changes in enteric glial cells Enteric glial cells have been little studied during ageing. Phillips et al. (2004b); however, compared glial cell numbers in different regions of the small and large intestines in 5–6 and 26 month-old rats. These authors reported a significant loss of enteric glia (expressed as counts per ganglion and per ganglionic area) in all areas except the rectum, where loss was observed, but was not significant. The glial cell loss was also found to be proportional to the neuronal loss in this study.

Challenges in the analysis of changes of neuronal numbers in the ageing gut The previous section (and Table 1) demonstrates that although age-related neuronal loss, particularly of cholinergic neurons, has been widely described, not all evidence supports the hypothesis that extensive neuronal death takes place in the ENS during ageing. Furthermore, there are large variations in the proportions of neurons that have reported to be lost in AL-fed animals. Since caloric restriction has been shown to reduce or eliminate neuronal loss, this variation raises an important question about possible differences in food intake by AL-fed animals, which could result in a range of body weights and variable levels of neuronal loss. It is also notable that, for example, reported losses of as much as 20% (total neurons, Peck et al., 2009) and 30% (calretinin-IR, Thrasivoulou et al., 2006) have been found to be non-significant, indicating that variation in the data may be large, even for major neuronal subpopulations; this is also evident in some other studies. Analysis of changes in smaller subpopulations of enteric neurons, distributed over a wide area is therefore inevitably more difficult. There are a number of other possible reasons for the variation in the published data on neuronal losses in the ENS, and why the

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

evidence is conflicting. Careful consideration of these reasons may not only inform future studies in this area, but provide valuable information about the nature, and possible causes, of neuronal ageing in the gut. The sources of variation and issues that contribute to complication of analysis of ageing ENS analysis, and which make comparison of data from different studies problematic, are numerous and complex. They are summarised in Table 2 and discussed here (also see Phillips et al., 2007). The anatomy of the enteric plexuses poses difficulties for accurate neuronal quantification As is well-known among enteric neurobiologists, the arrangement and structure of the enteric ganglionic networks lying within a non-uniform tube make quantification of enteric neurons challenging. In contrast to other peripheral nervous system ganglia, which are discrete structures in which it is possible to perform accurate quantification on an entire ganglion using unbiased stereology, myenteric ganglia have an irregular size and arrangement in most parts of the gut. Methods for the quantification of enteric neurons and comparison of the abundance of different neuronal populations using whole mount preparations are now well-established. However in ageing studies it is not always possible to obtain sufficient material for whole mount preparations, or the time needed for preparations of samples for whole mounts may not be available, so some studies have analysed changes in neuronal numbers using tissue sections.

studied, and then applied a correction factor to compensate for the effect. However, not all studies have performed this correction, so some reports of neuronal loss may in fact be due to a dilution effect. This effect is a particular problem for analysis of human specimens, which may be biopsies, or post-mortem samples taken under conditions in which measurements of the gut cannot be taken. There are four further complicating factors that affect the use even of this correction factor. First is that a change in the overall length of the small (or large) intestine may not be the result of a uniform increase along the length of the organ – growth may be different in different regions. Second is that the degree of change in dimensions may vary even within a cohort of animals of the same age, so application of a standard correction factor may result in inaccuracies (Gamage et al., in press). Third is that the degree of distension of the gut at the time of sampling, but this is usually controlled for by the use of maximal distension or pharmacological relaxants. Finally, the degree of stretch during preparation of whole mounts may vary; this has not been controlled for in many studies (Gamage et al., in press). In a different approach to avoiding the problems of gut growth, some workers have expressed neuronal density per ganglion or per ganglionic area. While individual myenteric ganglia are regular in shape and easily distinguished in the ileum, in other areas, particularly in the large intestine, it is not possible to ascertain the borders of individual ganglia with certainty, so in this case expression of numbers per ganglionic area is more reliable and so preferable, but this measure has not been used in all studies.

Changes in the dimensions of the gut and arrangement of the plexuses during the lifespan

Issues of sampling

The issue of gut growth is very important to consider during analysis of changes in the density of enteric neuronal cell bodies during ageing, since the same number of neuronal cell bodies in a larger sample will appear smaller when expressed per unit area (a ‘dilution’ effect, see Cowen et al., 2000; Peck et al., 2009; Phillips and Powley, 2007). Thus a false loss of neurons would be recorded. To avoid this dilution effect due to gut growth, many authors have measured the size of the intestine in the animals that they have

A wide variation in the sampling methods has been used in the analysis of changes in neuronal number during ageing. This variation applies not only to the preparation of samples (e.g. wholemounts versus tissue sections and choice of fixative), but also to the actual area of the samples in which neurons are counted. The choice of the areas to be sampled also varies, some workers counting random fields or images taken randomly, others using a systematic sampling method to ensure, for example that

Table 2 Possible sources of variability in reports of age-related enteric neuronal loss. Sources of variability:

Notes:

Tissue sections/wholemounts Use of different markers for the total neuronal population Different counting methods

Most studies have used whole mounts, but some have used tissue sections. Wide range of general neuronal markers used, including histological (Giemsa, Cuprolinic blue), histochemical (NADH diaphorase) and immunolabelling for a range of general neuronal markers (PGP 9.5, Hu C/D). From images or directly from slides. For whole mounts with or without confocal analysis of z stacks (these are needed in regions where the plexus is thicker and neurons may lie on top of each other) Random or systematic sampling methods used A large range of sampling methods have been employed Differing sample sizes Differing total areas have been used for neuronal counts Different methods to ensure standard distension of the Some studies use injection of fluid to maximise distension, others use nicardipine to ensure maximal smooth sample at the time of fixation muscle relaxation Gut growth corrected for or not Many studies correct for changes in intestinal dimensions with age, but some do not. Clearly this is a difficult problem when analysing human samples. The intestinal dimensions of individuals within a cohort at the same age may not all be the same, so application of a mean correction factor may lead to inaccuracies. Measurement of the degree of stretch of wholemount The degree of stretch of individual wholemount preparations is not usually taken into account. Possible preparations changes in dimensions after fixation and processing have also not usually been considered How neuronal numbers expressed Neuronal density has been expressed per unit serosal area, per unit length of gut region, per ganglion and per ganglionic area Ranges of ages studied Ages compared include ‘old’ vs. ‘young’, or variable ranges of ages. The ages studied in relation to the maximum lifespan of the species/ strain/cohort being used has not often been considered, in part because the typical maximal lifespan of the cohort being studied may not be knowna. Some studies do not use older than ‘middle-aged’ animals. The age of the ‘young’ control animals varies. Animal husbandry Particularly diet; differences between results from ad libitum-fed and caloric restriction have been demonstrated. Most studies have used AL feeding regimes, but the type of chow used in AL feeding may vary, for example a ‘breeding’ or a ‘maintenance’ diet may be used. This is often not detailed in publications, but may influence the rate of ageing. Other aspects of husbandry may also affect animal ageing and longevity. a

Some studies however, have used established colonies, such as the National Institute of Ageing (NIA) colony of Fischer 344 rats. See Phillips and Powley, 2007.

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

there is no bias caused by disproportionate sampling of mesenteric and anti-mesenteric regions (which are known to have different densities of neurons). This type of problem is particularly difficult in GI regions in which the plexus is irregular, so that even some systematic sampling methods can introduce an element of bias (Gamage et al., in press; Phillips and Powley, 2007). Choice of neuronal markers used A number of different pan-neuronal markers have been used to study enteric neuronal ageing (see Tables 1 and 2). The reliability of these different markers varies considerably. For example it has been reported that NADHd only labels approximately 80% of guinea-pig myenteric neurons, and PGP9.5 also does not label all myenteric neuronal cell bodies (Phillips et al., 2004a; Phillips and Powley, 2007). Another difficulty in the use of immunolabelling to determine if total or particular subpopulations are reduced in number during ageing is that expression of the marker being studied may be down-regulated, so that the neurons in question may be present, but not detectable by the method being used. This problem has been described for HuC/D expression, which has been shown to be reduced in ageing rat myenteric neurons (Phillips et al., 2004b). Selection and number of time points in ageing studies Another key variable that is important in ageing studies is the time-points that are used. Studies that include larger numbers of time points are increasingly favoured, because the time at which age-associated changes begin, and the rate at which changes occur are important aspects in understanding both the ageing process itself and the development of interventions that may alleviate or delay detrimental changes. Detailed ageing studies thus require a significant time commitment and are expensive, so it is perhaps not surprising that some studies report differences between ‘old’ and ‘young’ ages, or are confined to younger ages that do not approach the maximum lifespan of the species involved. In this respect, it is essential to know the typical survival of the animals being studied; not only do different laboratory animal species, such as rats, mice and guinea-pigs have different lifespans, but the lifespan of different strains varies, and is dependent upon husbandry (e.g. see Merry, 2005; Nadon, 2006). Thus ‘old’ animals in some studies may be equivalent to ‘middle-aged’ animals in others. Effects of diet and environmental enrichment on enteric neuronal survival during ageing The issue of animal husbandry is very important in the study of ageing. Myenteric neuronal loss has been reported to be reduced (da Silva Porto et al., 2012; Thrasivoulou et al., 2006) or eliminated (Cowen et al., 2000; Johnson et al., 1998) by caloric restriction. The widespread effects of caloric restriction on ageing are now well-characterised in a number of species and the intervention is known not only to decrease morbidity but to increase life-span. These effects raise an interesting question in the interpretation of differential changes in AL-fed and calorically restricted animals; the effective physiological age of the latter may in fact be younger than that of the former, so direct comparison of the two groups may not be strictly valid (e.g. see Hoyle and Saffrey, 2012; Merry, 2005). Another factor that has not usually been considered is the type of diet used (e.g. breeding versus maintenance diet) and dietary supplements. The amount of food intake by AL-fed animals is also an issue; this could vary considerably, some ageing rodents may gain a lot of weight and it is possible that they may be obese, thus confounding data believed to be due to ageing. Not all studies

7

have provided detailed information about food intake and body weight of the animals studied. The effects of diet are also not confined to post-weaning period. Prenatal nutrition may also influence neuronal numbers in the adult gut (Santer and Conboy, 1990). It is also possible that other aspects of husbandry, such as provision of an enriched environment that encourages increased exercise, may influence ageing of enteric neurons. Total myenteric neuronal loss in 12 month old rats has been reported to be reduced by exercise, although nitergic neuronal numbers were reduced (Martinez Gagliardo et al., 2008). Lack of mobility of humans is known to have a negative effect on frequency of bowel movements (e.g. see Rao and Go, 2010). Nevertheless, despite all the issues described above, the presence of spaces in the myenteric ganglia of ageing humans (Hanani et al., 2004) and rats (Thrasivoulou et al., 2006) supports the evidence that some enteric neurons are lost during ageing, at least in some animals. Further evidence for enteric neurodegeneration in ageing is described in the following section.

Degenerative changes in the ENS Although the extent of neuronal cell death that may occur in the ENS during ageing is not established, there is clear evidence that neurodegeneration of enteric neurons does occur in older animals. Most of this evidence comes from studies at the light microscope level; very few studies have described ultrastructural changes in the ENS during ageing (Feher and Penzes, 1987). Several groups have demonstrated the presence of swollen and dystrophic nerve fibres in ageing gut. NADPHd positive dystrophic fibres have been described in the myenteric plexus of rats (Phillips et al., 2003) and similar nNOS-IR fibres in mice (Gamage et al., in press). In the rat myenteric plexus, degenerating NADPHd fibres (Phillips et al., 2003) have been detected in rats at 24 months of age. In the mouse distal colon, dystrophic nNOS- IR and calbindinIR nerve fibres were seen occasionally in 12 month animals, but were frequent in animals at 18 and 24 months of age (Gamage et al., in press). Degenerating SP-, VIP- and somatostatinimmunoreactive nerve fibres have also been described in aged rat small intestine at the ultrastructural level (Feher and Penzes, 1987). Ageing is associated with the accumulation of lipofuscin in postmitotic cells (see Grey and Woulfe, 2005). Accumulations of lipofuscin in ageing enteric neurons have been reported in rats (Corns et al., 2002) (Phillips et al., 2004a) and guinea-pigs (Abalo et al., 2005, 2007). Although ultrastructural changes in the ageing ENS have not been studied in detail, some analysis of the ENS at the electron microscope level has been performed. Gabella (1989) has reported that degenerating myenteric neurons are sometimes seen in the ageing guineapig myenteric plexus, but that the fine structure of the surrounding ganglion appears intact. Feher and Penzes (1987) have described the presence of degenerating SP-VIP and somatostatin-IR fibres in the aged rat small intestine. In preliminary studies of ageing myenteric plexus ganglia in the mouse terminal colon, we have observed increased numbers of inclusions, indicative of neurodegeneration, in ageing mouse myenteric neurons (unpublished results).

Alpha-synuclein and enteric neurodegeneration Recent evidence has shown that alpha-synuclein is expressed by enteric neurons (Phillips and Powley, 2007; Phillips et al., 2008). This protein is also expressed in many CNS neurons and its aggregation in Lewy bodies is a hallmark of Parkinson’s disease. Phillips et al. (2008) have demonstrated that alpha-synuclein is

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

expressed by 3% of myenteric neurons in the stomach and 22% of myenteric neurons in the duodenum of Fischer 344 rats, and that it colocalises with nNOS, calbindin and calretinin in myenteric subpopulations. Abundant alpha-synuclein-IR nerve fibres were also found in these tissues. In 18 month- and in 24–25 month-old Fischer 344 rats, alpha-synuclein positive aggregates were observed in dystrophic axons that were also immunoreactive for nNOS, calbindin and calretinin (Phillips et al., 2009), although not all dystrophic axons expressing these markers also expressed alpha-synuclein. Hyper-phosphorylated Tau aggregates were also observed in the ageing myenteric plexus, but in a limited area, the proximal small intestine (Phillips et al., 2009). The expression and aggregation of alpha-synuclein in enteric neurons is not only of interest in the development of neurodegeneration during ageing of the ENS, but also in Parkinson’s disease. Lewy bodies have been described in the ENS of Parkinson’s disease patients, and it has been suggested that the ENS may be one possible site of onset of the disease caused by an ingested pathogen, and that the pathology may spread to the CNS from the intestine (e.g. see Hawkes et al., 2007).

the same animals. Santer (1994) however, observed that the density of NADPHd positive fibres in the tertiary plexus appeared greater in the small intestine of aged rats and an increase in the density of nNOS-IR fibres has been described In the tertiary plexus of the mouse distal colon (Gamage et al., in press). A different approach has been used in the study of changes in neuronal subpopulations in ageing mice by Van Ginneken et al. (2011). These authors performed volume density measurements of labelled neurons in tissue sections from wild type and transgenic (see below) mice. No reduction in the density of nervous structures labelled for βIII-tubulin, substance P or nNOS were observed between 18 month old (the oldest animals studied) and 6 month old animals, although some reduction at 12 months was observed. While more work clearly needs to be done, taken together these data all indicate that if major neuronal losses occur in the ageing gut, they do not necessarily lead to loss of nerve fibres. It is possible, and perhaps even likely, that compensatory mechanisms occur, such as sprouting and or growth of remaining neurites from surviving neurons. In this context it is of interest that extensive neurite regeneration occurs after chemical ablation of the myenteric plexus in adult mice (Hanani et al., 2003), indicating that mechanisms for regeneration exist in the adult ENS.

Changes in nerve fibre density in the ageing gut: evidence for regeneration? Neurogenesis in the ageing ENS As discussed above, an increase in the length and circumference of the gut during adulthood and ageing would be expected to result in a dilution of nerve fibres, unless there is growth and/or sprouting of nerves to compensate for increases in the volume of the smooth muscle, for example. In addition to the possible dilution of nerve fibres due to GI growth, a significant loss of enteric neurons in aged animals, such as the 40–60% losses reported in some studies (see Table 1) would be expected to result in a clearly-detectable loss of nerve fibres. Several studies have investigated changes in nerve fibre density in the ageing gut, although unfortunately no such analysis has been performed in cohorts reported to have major losses of neuronal cell bodies. Santer (1994) observed that the density of NADPHd positive nerve fibres appeared to be greater in the myenteric plexus and muscularis externa of ageing rat small intestine, but did not quantify the changes. Despite a decrease in density of fibres due to smooth muscle growth, an increase in the total innervation of the guinea-pig taenia coli in aged guinea-pigs has been described (Gabella, 2001). Peck et al. (2009), who studied changes in the density of different types of nerve fibre in the circular muscle of the guinea-pig colon during ageing, found an apparent reduction in density of nNOS and SP-IR fibres in samples from older animals. However, once increase in muscle thickness was taken into account, no change in total area of these nerve fibres was measured during ageing, indicating that reduced fibre density was due to gut growth rather than neuronal loss. In human sigmoid colon, a similar result was obtained; regression analysis showed a decrease in the density of nNOS -, VIP- and SP-IR nerve fibres in the circular muscle during ageing (Southwell et al., 2010), but this was accompanied by an increase in circular muscle thickness, indicating that there was no age-related loss of nerve fibres. Bernard et al. (2009) have also reported no change in the density of nerve fibres, in this case despite neuronal loss (see Table 1), and suggest a compensatory mechanism of neuronal growth may occur during ageing. Changes in the density of nerve fibres in the tertiary plexus have also been examined. Abalo et al. (2007) investigated changes in calretinin-IR nerve fibre density in the tertiary plexus of the guinea-pig ileum during ageing, and found no significant change, again despite a loss of myenteric neurons expressing calretinin in

It is now established that enteric neural stem cells are present in the adult ENS (Kruger et al., 2002) and also that adult enteric glia have the potential to form neurons in vitro (Joseph et al., 2011; Laranjeira et al., 2011 and see Gershon, 2011; Metzger, 2010). Studies to investigate if neurogenesis is ongoing in adulthood however, have not detected formation of new neurons except in culture, or under conditions in which damage or stress has occurred (Joseph et al., 2011; Laranjeira et al., 2011 and see Gershon, 2011). Among a range of other paradigms studied, Joseph et al. (2011) investigated if neurogenesis occurred in ageing rats, but found no evidence for the formation of new neurons in 21 month old animals. Thus it appears as if neurogenesis does not typically occur in the ageing ENS. However it is possible that small local regions of damage or stress occur during ageing, which may stimulate generation of new neurons. In this context the appearance of occasional single neurons at the edge of some ganglia in old samples (Abalo et al., 2005; Gabella, 1989; Hanani et al., 2004) is perhaps interesting, as this is a site of neural precursors. Differences in restorative neurogenesis in the samples used for analysis of age-related changes in neuronal number could possibly account for the different degrees of neuronal loss obtained in different studies.

Importance and implications of age-related enteric neuronal loss: are enteric neurons more vulnerable than other neurons to age-related degeneration? Age-related loss of enteric neurons is significant for two very different reasons; first are the obvious implications for age-related changes in gut function (described in the next section), second is that in other parts of the nervous system neuronal loss during ageing does not occur. Although neuronal loss in the CNS was at one time considered to be typical, relatively recent developments in the techniques of unbiased stereology revealed that such loss is limited to neurodegenerative diseases. Enteric neurons may therefore be more vulnerable to ageing than neurons of the CNS and other peripheral ganglia. Thus it is both of local and broader interest if neuronal losses do indeed occur in the ageing ENS. There are several reasons why enteric neurons may be more vulnerable to age-related loss than neurons in other parts of the

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

nervous system. In addition to the continual physical changes (which are also a feature of some other autonomic neurons, such as those that innervate other organs that regularly undergo distension, the heart and bladder), the intestine is exposed to continual changes in its contents, including the intestinal microbiota. As already touched upon, the possible exposure of enteric neurons to pathogens has been suggested as one site for the development of neuropathies such as those seen in PD, which may spread to the CNS via the vagal nerve (Hawkes et al., 2007; Phillips et al., 2009, 2008). This may be just one example of a pathology resulting from expose to ingested pathogens. However, changes in the balance of non-pathogenic microbial populations in the gut during ageing may also influence enteric neurons. The gut microbiota are now known to play a major role in human health (e.g. see Grenham et al., 2011; Nicholson et al., 2012) including gastrointestinal health (see Aziz et al., 2013) Changes in the in the gut microbiota have been described during ageing (e.g. Makivuokko et al., 2010 and see Aziz et al., 2013; Makivuokko et al., 2010; Tiihonen et al., 2010), and these ageassociated changes can be associated with diet, and with negative indicators of health (Claesson et al., 2012). Interactions between the microbiota, the cells of the gut and the rest of the organism are complex, and only now beginning to be understood. In addition to effects within the gut itself, including effects on enteric neurons, the influence of the microbiota exerted via the brain-gut axis is also now an area of great interest (e.g. see Rhee et al., 2009). Here, the effects of the microbiota on enteric neurons and glia are briefly considered, in order to highlight possible areas that may impact upon ageing of the ENS. Once again, these effects are complex, and involve interactions with other intestinal cells, notably enteric glia and the cells of the mucosal immune system. One key aspect of exposure of enteric neurons to microbes is the permeability of the intestinal epithelium, which has an important barrier role (see Neunlist et al., 2013). The intestinal epithelial barrier (IEB) is not fully developed at birth and some permeability during the postnatal period has been suggested to be important for the appropriate development of the immune system. Changes in IEB permeability are associated with several disease states and can lead to intestinal inflammation, which in turn may lead to neuronal damage (see below). Of particular interest are recent reports suggesting that cells of the ENS play an important role in the maintenance of IEB function (Neunlist et al., 2013; Savidge et al., 2007). Once microbes cross the IEB, they can have a number of effects, and may cause an inflammatory response. Inflammatory diseases of the gut are a major cause of morbidity and have been studied intensively. Interaction between enteric neurons, glia and the cells of the mucosal immune system is an area that is of increasing interest in the context of these diseases. Evidence demonstrates close interactions of both enteric neurons and glia with cells of the mucosal immune system, and inflammation may affect both the structure and function of the ENS (see Lomax et al., 2005; Mawe et al., 2009). For example, a loss of enteric neurons has been described in some experimental models of inflammation (see Lomax et al., 2005) and hyperexcitability of AH neurons (IPANs) occurs in inflammation (Mawe et al., 2009). Changes in the microbiota during ageing, which may be related to diet, may therefore increase intestinal inflammation, which may, in turn influence cells of the ENS, possibly resulting in their damage or death. It is also possible that relatively small localised areas of inflammation may occur during ageing, leading to small areas of neuronal loss. Recent studies have also indicated that the microbiota may also have an important influence on the physiological and/or neurochemical properties of enteric neurons. For example, in recent work, McVey Neufeld et al. (2013) found that excitability of intrinsic primary afferent neurons (IPANS) was reduced in germ-

9

free mice. di Giancamillo et al. (2010) presented evidence that there was an increase in the number of galanin- and calcitonin gene-related peptide immunopositive neurons in the ileal submucous plexuses of piglets fed a diet supplemented with probiotics. The total number of neurons, however, was not affected, suggesting an effect on existing neurons. An increase in the number of glial cells in the submucosa was also detected, in the same regions. Whether these changes result from direct or indirect effect of the bacteria used, however, is not clear. Thus it is possible that changes in the microbiota during ageing could impact upon the properties of enteric neurons and glial cells. Clearly this is an important area for future study. Thus there is a complex interrelationship between the gut microbiota, epithelium, mucosal immune system and the ENS, which may not only make enteric neurons particularly vulnerable during ageing, but may also play an important role in the aetiology of other diseases. Age-associated changes in the microbiota pose an interesting question of cause or effect. For example age-associated changes in effector cells may affect gut physiology resulting in delayed transit, which may lead to bacterial overgrowth, in turn leading to a change in the balance of microbiota in the gut. Alternatively, changes in microbiota could influence intestinal cells, including neurons, thus leading to physiological changes including delayed transit. It is perhaps, likely, that ageing involves both these possibilities Finally, if contact with gut microbes and/or mucosal inflammation is indeed linked to enteric neural damage and degeneration during ageing, one would expect submucosal neurons and myenteric neurons that project to the mucosa, IPANs, might be preferentially affected. Extrinsic nerves that innervate the mucosa may also be affected.

Does enteric neuronal loss during ageing lead to age-related GI dysfunction? The potential significance of age-related neuronal loss to the changes in gut function that are common during ageing is clear. However in most studies in which high percentages of neuronal loss have been reported, the issue of possible GI dysfunction in the experimental animals used has not been addressed. It would be expected that a loss of 40–60% of the total myenteric neuronal population would have a marked effect on function, which could be manifest as decreased frequency of defaecation; constipation is common in elderly humans. Observations on the frequency, size and composition of stools in older animals however, have not often been reported in studies that have investigated age-related changes in neuronal number. Gabella (1989); however, reported that there were no obvious changes in feeding habits or faeces production of the guinea-pigs in his study, which exhibited a loss of 40–60% of myenteric neurons in the small intestine. Most physiological and pharmacological studies of GI tissue from ageing animals have not investigated neuronal numbers in the same tissues, and the converse is also the case; few studies have investigated physiological changes in animals in which neuronal numbers have also been studied. However, in a comparison of cholinergic neuromuscular transmission in ileal longitudinal muscle of young and aged ad libitum fed and caloricallyrestricted animals no age-associated impairment of the cholinergic system was observed, and diet did not affect responses to field stimulation (Hoyle and Saffrey, 2012). This work was carried on animals in which a neuronal loss had been demonstrated in AL-fed animals (Thrasivoulou et al., 2006), so it appears as if some cholinergic function at least was unimpaired in animals that exhibited a major neuronal loss. Abalo et al. (2005) found an

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

increase in the force of contraction induced by excitatory stimuli in guinea-pig ileum, which paralleled an age-related decrease in density of excitatory motor neurons, and suggested a compensatory mechanism might be involved. In recent work, we have been performing physiological and pharmacological analyses of the terminal bowel of ageing mice in parallel with morphological and immunohistochemical studies. We observe changes in motility and stool numbers (Patel et al., 2012), but no significant changes in neuronal numbers in the distal colon in mice up to 24 months of age (Gamage et al., in press).

What are the mechanisms of enteric neuronal loss during ageing? Many different mechanisms of cellular ageing have been proposed and it is of course possible, or indeed likely, that ageing involves more than a single mechanism. Some cell types may be more vulnerable than others to particular types of change or damage, however. Several potential mechanisms of neuronal loss have been investigated in the ageing ENS; these are considered below. Oxidative and free radical damage have been proposed to play a major role in neuronal ageing, because neurons are highly metabolically active cells. Such damage may be exacerbated for enteric neurons because of their environment in close proximity to the gut contents, microbiota and local immune system, as described above. In support of the suggestion that oxidative stress may play a role in enteric neurodegeneration, ageing myenteric neurons have been shown to have elevated levels of reactive oxygen species (ROS) (Thrasivoulou et al., 2006). Furthermore the rate of ROS generation in 15 month-old rats was found to be reduced by caloric restriction (Thrasivoulou et al., 2006). Neurons from middle-aged (15 month) AL-fed rats were also more vulnerable to menadione –induced apoptosis than neurons from CR-fed animals at the same age, or 6 month-old AL fed rats (Thrasivoulou et al., 2006). There is evidence that neurotrophic factors protect neurons from oxidative stress (Hardingham and Lipton, 2011; Numakawa et al., 2011) and the neurotropic factors GDNF and NT-3 have been shown to reduce ROS generation by myenteric neurons in samples from both 12–15 month and 24 month old CR-fed, but not AL-fed rats at the same ages (Thrasivoulou et al., 2006). Neurotrophin-3, but not GDNF, also reduces neuronal death in dissociated myenteric plexus in response to hydrogen peroxide treatment (Korsak et al., 2012b). Thus it has been suggested that neurotrophic factors may play a role in protection of enteric neurons throughout life, and that a breakdown in neurotrophic factor support leading to increased susceptibility to free radical damage may contribute to enteric neuronal degeneration, and eventually neuronal death (e.g. see Camilleri et al., 2008 Thrasivoulou et al., 2006). The neurotrophic factors glial cell line-derived neurotrophic factor (GDNF) and neurturin are known to promote the survival and development of enteric neurons during embryonic and early postnatal development (see Saffrey, 2009). Analysis of the expression of these factors and their receptors in the ileum of AL- and caloricallyrestricted rats during ageing show that expression continues in ageing animals, and the levels GDNF and its receptor GFRα-1 are elevated in ageing AL-fed, but not CR animals. (Korsak et al., 2012a). Such an increase could represent a compensatory upregulation in response to elevated stress and/or neurodegeneration. In contrast the expression of the neurturin receptor, GFRα-2b, was reduced in ageing AL-fed animals (Korsak et al., 2012a). Calcium dysregulation has also been proposed to be a factor that contributes to neuronal ageing, and the presence of calciumbinding proteins in some neurons has been suggested to be protective. Reduction in the number of enteric neurons expressing

calcium binding proteins such as calbindin (Thrasivoulou et al., 2006), calretinin (Abalo et al., 2005; Thrasivoulou et al., 2006) and neurocalcin (Corns et al., 2002) in ageing animals could be the result of changes in the expression of these proteins, indicative of calcium dysregulation. The development of alpha-synuclein immunopositive aggregates in calbindin- and calretinin-IR myenteric neurons during ageing (Phillips et al., 2009) suggests that these neurons may in fact be vulnerable to age-related degeneration. Cellular senescence occurs in many cell types during ageing. In addition to well-known phenomenon of replicative senescence, exhibited by mitotic cells, recent evidence shows that post-mitotic neurons exhibit a p21-dependent senescence-associated phenotype, induced by a DNA damage response (Jurk et al., 2012). Recent evidence indicates that this senescence-associated phenotype is also exhibited by a population of myenteric neurons in the ageing mouse small intestine (Jurk et al., 2012). The proportion of neurons affected was reported to be between 20 and 40%, in 22–24 month-old animals. The subpopulation/s of neurons exhibiting the senescence-associated phenotype has not yet been established, but it of interest that recent evidence indicates that an element of the phenotype is secretory, and that in culture at least, senescent fibroblasts can induce damage in adjacent cells (Nelson et al., 2012). This last observation raises the interesting possibility that ageing of enteric neurons could also be affected by ageing of surrounding replicative cells.

Transgenic animal models and the study of ENS ageing A number of transgenic models of ageing and neurodegeneration are now available. Analysis of these animals may provide valuable information about the mechanism of ageing in wild-type animals. ENS changes however, have been studied in only very few of these models. Klotho is a ‘anti-ageing’ peptide that occurs both bound to membranes and in a soluble state, is a co-receptor for FGF and also regulates vitamin D and phosphate metabolism (see Asuzu et al., 2011). Transgenic mice expressing low levels of Klotho exhibit premature ageing phenotypes and have a shortened lifespan (60–70 days), over-expressers have an extended lifespan 20–30% longer than their wild-type littermates (see Asuzu et al., 2011). Analysis of the ENS in mice with reduced Klotho expression have shown that myenteric neuronal numbers were reduced in some regions of the gut, but that gut lengths were also reduced, possibly because of reduced food intake by these animals. These results serve to indicate that caution is necessary in interpretation of the data from transgenic animals, as observations may be due to secondary or compensatory changes, rather than a primary effect of the under-or over-expression of the gene involved. Another recent study has investigated changes in the ENS in Thy1-APP23 transgenic mice, which have been used as a model for Alzheimer’s disease and over-express a form of amyloid precursor protein (Van Ginneken et al., 2011). Although amyloid deposits were observed in the smooth muscle and mucosa of the intestines of these transgenic animals, the density of βIII-tubulin, SP and NOS-IR fibres was found to be increased. The authors suggest that, as in the CNS, under some circumstance a neurotrophic effect of APP fragments may occur in the intestine of these transgenic mice, again demonstrating the difficulty of interpreting data from some transgenic animal models.

Conclusions and future directions The ENS continues to undergo changes throughout life. In part these changes are due to a non-pathological plasticity of the system, which is clearly advantageous, but there is also evidence

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

that ageing is associated with enteric neurodegeneration. The extent of this degeneration is not clear, and it is likely to vary according to environmental conditions, including diet and changes in the microbiota. Diet and the microbiota, together with intestinal inflammation and the brain-gut-axis, which is now increasingly referred to the Q2 brain-gut-enteric microbiota axis (e.g. Cryan and O’Mahony, 2011; Rhee et al., 2009) will clearly be an important area for future study in the context of ENS ageing and GI disorders. It is also increasingly clear that cell interactions that may influence ageing of the ENS involve more cell types and are more complex than previously believed. The relationship between cellular and physiological changes in the aging ENS however, remains poorly understood. It is therefore important that parallel analyses of cellular, molecular and functional changes of the ENS and other intestinal cells during ageing are performed in future work. Understanding the changes that occur in the ENS due to normal cellular ageing is an important area for future research. The increase in the elderly population will result in many more people suffering from age-related GI disorders, such as faecal incontinence and constipation. These conditions not only have a detrimental effect of the quality of life of the elderly, but also incur large health-care costs. Understanding the different contributions made by cellular ageing, lifestyle, diet, the microbiota, co-morbidity and medication to ENS ageing is therefore essential in order that preventative and therapeutic measures can be developed.

Acknowledgements Research in the Author’s laboratory has been supported by BBSRC Grants: SAGE Initiative 108/SAG10013 and BB/G015988/1. References Abalo, R., Jose Rivera, A., Vera, G., Isabel Martin, M., 2005. Ileal myenteric plexus in aged guinea-pigs: loss of structure and calretinin-immunoreactive neurones. Neurogastroenterol. Motil. 17, 123–132. Abalo, R., Vera, G., Rivera, A.J., Martin, M.I., 2007. Age-related changes in the gastrointestinal tract: a functional and immunohistochemical study in guineapig ileum. Life Sci. 80, 2436–2445. Asuzu, D.T., Hayashi, Y., Izbeki, F., Popko, L.N., Young, D.L., Bardsley, M.R., Lorincz, A., Kuro, O.M., Linden, D.R., Farrugia, G., Ordog, T., 2011. Generalized neuromuscular hypoplasia, reduced smooth muscle myosin and altered gut motility in the klotho model of premature aging. Neurogastroenterol. Motil. 23, e309–323. Aziz, Q., Dore, J., Emmanuel, A., Guarner, F., Quigley, E.M., 2013. Gut microbiota and gastrointestinal health: current concepts and future directions. Neurogastroenterol. Motil. 25, 4–15. Belai, A., Burnstock, G., 1999. Distribution and colocalization of nitric oxide synthase and calretinin in myenteric neurons of developing, aging, and Crohn’s disease human small intestine. Dig. Dis. Sci. 44, 1579–1587. Belai, A., Cooper, S., Burnstock, G., 1995. Effect of age on NADPH-diaphorasecontaining myenteric neurones of rat ileum and proximal colon. Cell Tissue Res. 279, 379–383. Bernard, C.E., Gibbons, S.J., Gomez-Pinilla, P.J., Lurken, M.S., Schmalz, P.F., Roeder, J.L., Linden, D., Cima, R.R., Dozois, E.J., Larson, D.W., Camilleri, M., Zinsmeister, A.R., Pozo, M.J., Hicks, G.A., Farrugia, G., 2009. Effect of age on the enteric nervous system of the human colon. Neurogastroenterol. Motil. 21 746–e746. Bitar, K., Greenwood-Van Meerveld, B., Saad, R., Wiley, J.W., 2011. Aging and gastrointestinal neuromuscular function: insights from within and outside the gut. Neurogastroenterol. Motil. 23, 490–501. Bitar, K.N., Patil, S.B., 2004. Aging and gastrointestinal smooth muscle. Mech. Ageing Dev. 125, 907–910. Camilleri, M., Cowen, T., Koch, T.R., 2008. Enteric neurodegeneration in ageing CORRECTED in full. Neurogastroenterol. Motil. 20, 185–196. Chatoor, D.R., Taylor, S.J., Cohen, C.R., Emmanuel, A.V., 2007. Faecal incontinence. British J. Surg. 94, 134–144. Choi, J.H., Lee, C.H., Chung, D.W., Hwang, I.K., Won, M.H., Seong, J.K., Yoon, Y.S., Lee, I.S., 2008. Age-related changes of calbindin D-28k-immunoreactive neurons in the myenteric plexus of gerbil duodenum. J. Vet. Med. Sci. 70, 343–348. Claesson, M.J., Jeffery, I.B., Conde, S., Power, S.E., O’Connor, E.M., Cusack, S., Harris, H.M., Coakley, M., Lakshminarayanan, B., O’Sullivan, O., Fitzgerald, G.F., Deane, J., O’Connor, M., Harnedy, N., O’Connor, K., O’Mahony, D., van Sinderen, D., Wallace, M., Brennan, L., Stanton, C., Marchesi, J.R., Fitzgerald, A.P., Shanahan, F., Hill, C.,

11

Ross, P.W., O’Toole, P.W., 2012. Gut microbiota composition correlates with diet and health in the elderly. Nature 488, 178–184. Corns, R.A., Hidaka, H., Santer, R.M., 2002. Neurocalcin-alpha immunoreactivity in the enteric nervous system of young and aged rats. Cell Calcium 31, 53–58. Cowen, T., Johnson, R.J.R., Soubeyre, V., Santer, R.M., 2000. Restricted diet rescues rat enteric motor neurones from age related cell death. Gut 47, 653–660. da Silva Porto, G., Bertaglia Pereira, J.N., Tiburcio, V.G., Stabille, S.R., Garcia de Faria, H., de Melo Germano, R., de Britto Mari, R., 2012. Effect of caloric restriction on myenteric neuroplasticity in the rat duodenum during aging. Auton. Neurosci. 168, 43–47. de Souza, R.R., Moratelli, H.B., Borges, N., Liberti, E.A., 1993. Age-induced nerve cell loss in the myenteric plexus of the small intestine in man. Gerontology 39, 183–188. de Vries, P., Soret, R., Suply, E., Heloury, Y., Neunlist, M., 2010. Postnatal development of myenteric neurochemical phenotype and impact on neuromuscular transmission in the rat colon. Am. J. Physiol. Gastrointest. Liver Physiol. 299, G539–547. di Giancamillo, A., Vitari, F., Bosi, G., Savoini, G., Domeneghini, C., 2010. The chemical code of porcine enteric neurons and the number of enteric glial cells are altered by dietary probiotics. Neurogastroenterol. Motil. 22, e271–278. Drozdowski, L., Thomson, A.B., 2006. Aging and the intestine. World J. Gastroenterol. 12, 7578–7584. El-Salhy, M., Sandstrom, O., Holmlund, F., 1999. Age-induced changes in the enteric nervous system in the mouse. Mech. Ageing Dev. 107, 93–103. Feher, E., Penzes, L., 1987. Density of substance P, vasoactive intestinal polypeptide and somatostatin-containing nerve fibers in the ageing small intestine of the rats. Gerontology 33, 341–348. Foong, J.P., Nguyen, T.V., Furness, J.B., Bornstein, J.C., Young, H.M., 2012. Myenteric neurons of the mouse small intestine undergo significant electrophysiological and morphological changes during postnatal development. J. Physiol. 590, 2375–2390. Franceschi, M., Di Mario, F., Leandro, G., Maggi, S., Pilotto, A., 2009. Acid-related disorders in the elderly. Best Pract. Res. Clin. Gastroenterol. 23, 839–848. Gabella, G., 1989. Fall in the number of myenteric neurons in aging guinea pigs. Gastroenterology 96, 1487–1493. Gabella, G., 1990. On the plasticity of form and structure of enteric ganglia. J. Auton. Nerv. Syst. 30 (Suppl.), S59–S66. Gabella, G., 2001. Development and ageing of intestinal musculature and nerves: the guinea-pig taenia coli. J. Neurocytol. 30, 733–766. Gallagher, P., O’Mahony, D., 2009. Constipation in old age. Best Pract. Res. Clin. Gastroenterol. 23, 875–887. Gamage, P., Ranson, R.N., Yeoman, M.S., Patel, B.A., Saffrey, M.J., Myenteric neuron Q4 numbers are maintained in aging mouse distal colon. Neurogastroenterol. Motil., in press. Gershon, M.D., 2011. Behind an enteric neuron there may lie a glial cell. J. Clin. Invest. 121, 3386–3389. Gomes, O.A., de Souza, R.R., Liberti, E.A., 1997. A preliminary investigation of the effects of aging on the nerve cell number in the myenteric ganglia of the human colon. Gerontology 43, 210–217. Grey, D.A., Woulfe, J., 2005. Lipofuscin and Aging: A Matter of Toxic Waste. Science of aging knowledge environment: SAGE KE 2005, re1. Grenham, S., Clarke, G., Cryan, J.F., Dinan, T.G., 2011. Brain-gut-microbe communication in health and disease. Front. Physiol. 2, 94. Hanani, M., Fellig, Y., Udassin, R., Freund, H.R., 2004. Age-related changes in the morphology of the myenteric plexus of the human colon. Auton. Neurosci. 113, 71–78. Hanani, M., Ledder, O., Yutkin, V., Abu-Dalu, R., Huang, T.Y., Hartig, W., Vannucchi, M.G., Faussone-Pellegrini, M.S., 2003. Regeneration of myenteric plexus in the mouse colon after experimental denervation with benzalkonium chloride. J. Comp. Neurol. 462, 315–327. Hardingham, G.E., Lipton, S.A., 2011. Regulation of neuronal oxidative and nitrosative stress by endogenous protective pathways and disease processes. Antioxid. Redox Signal. 14, 1421–1424. Hawkes, C.H., Del Tredici, K., Braak, H., 2007. Parkinson’s disease: a dual-hit hypothesis. Neuropathol. Appl. Neurobiol. 33, 599–614. Hoyle, C.H., Saffrey, M.J., 2012. Effects of aging on cholinergic neuromuscular transmission in isolated small intestine of ad libitum fed and caloricallyrestricted rats. Neurogastroenterol. Motil. 24, 586–592. Hoyle, C.H.V., Burnstock, G., 1989. Neuromuscular Transmission in the Gastrointestinal Tract. American Physiological Society, Bethesda, Maryland, USA. Johnson, R.J., Schemann, M., Santer, R.M., Cowen, T., 1998. The effects of age on the overall population and on sub-populations of myenteric neurons in the rat small intestine. J. Anat. 192 (Pt 4), 479–488. Joseph, N.M., He, S., Quintana, E., Kim, Y.G., Nunez, G., Morrison, S.J., 2011. Enteric glia are multipotent in culture but primarily form glia in the adult rodent gut. J. Clin. Invest. 121, 3398–3411. Jurk, D., Wang, C., Miwa, S., Maddick, M., Korolchuk, V., Tsolou, A., Gonos, E.S., Thrasivoulou, C., Jill Saffrey, M.J., Cameron, K., von Zglinicki, T., 2012. Postmitotic neurons develop a p21-dependent senescence-like phenotype driven by a DNA damage response. Aging cell. Kirkwood, T.B., 2004. Intrinsic ageing of gut epithelial stem cells. Mech. Ageing Dev. 125, 911–915. Korsak, K., Dolatshad, N.F., Silva, A.T., Saffrey, M.J., 2012a. Ageing of enteric neurons: oxidative stress, neurotrophic factors and antioxidant enzymes. Chem. Central J. 6, 80.

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015

12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

M. Jill Saffrey / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Korsak, K., Silva, A.T., Saffrey, M.J., 2012b. Differing effects of NT-3 and GDNF on dissociated enteric ganglion cells exposed to hydrogen peroxide in vitro. Neurosci. Lett. 517, 102–106. Kruger, G.M., Mosher, J.T., Bixby, S., Joseph, N., Iwashita, T., Morrison, S.J., 2002. Neural crest stem cells persist in the adult gut but undergo changes in selfrenewal, neuronal subtype potential, and factor responsiveness. Neuron 35, 657–669. Laranjeira, C., Sandgren, K., Kessaris, N., Richardson, W., Potocnik, A., Vanden Berghe, P., Pachnis, V., 2011. Glial cells in the mouse enteric nervous system can undergo neurogenesis in response to injury. J. Clin. Invest. 121, 3412–3424. Lomax, A.E., Fernandez, E., Sharkey, K.A., 2005. Plasticity of the enteric nervous system during intestinal inflammation. Neurogastroenterol. Motil. 17, 4–15. Makivuokko, H., Tiihonen, K., Tynkkynen, S., Paulin, L., Rautonen, N., 2010. The effect of age and non-steroidal anti-inflammatory drugs on human intestinal microbiota composition. Br. J. Nutr. 103, 227–234. Marese, A.C., de Freitas, P., Natali, M.R., 2007. Alterations of the number and the profile of myenteric neurons of Wistar rats promoted by age. Auton. Neurosci 137, 10–18. Martinez Gagliardo, K., Clebis, N.K., Stabille, S.R., De Britto Mari, R., De Sousa, J.M., De Souza, R.R., 2008. Exercise reduces inhibitory neuroactivity and protects myenteric neurons from age-related neurodegeneration. Auton. Neurosci. 141, 31–37. Mawe, G.M., Strong, D.S., Sharkey, K.A., 2009. Plasticity of enteric nerve functions in the inflamed and postinflamed gut. Neurogastroenterol. Motil. 21, 481–491. McVey Neufeld, K.A., Mao, Y.K., Bienenstock, J., Foster, J.A., Kunze, W.A., 2013. The microbiome is essential for normal gut intrinsic primary afferent neuron excitability in the mouse. Neurogastroenterol. Motil. 25, 183–e188. Meciano Filho, J., Carvalho, V.C., de Souza, R.R., 1995. Nerve cell loss in the myenteric plexus of the human esophagus in relation to age: a preliminary investigation. Gerontology 41, 18–21. Merry, B.J., 2005. Dietary restriction in rodents—delayed or retarded ageing? Mech. Ageing Dev. 126, 951–959. Metzger, M., 2010. Neurogenesis in the enteric nervous system. Arch. Ital. Biol. 148, 73–83. Nadon, N.L., 2006. Exploiting the rodent model for studies on the pharmacology of lifespan extension. Aging cell 5, 9–15. Nelson, G., Wordsworth, J., Wang, C., Jurk, D., Lawless, C., Martin-Ruiz, C., von Zglinicki, T., 2012. A senescent cell bystander effect: senescence-induced senescence. Aging cell 11, 345–349. Neunlist, M., Van Landeghem, L., Mahe, M.M., Derkinderen, P., des Varannes, S.B., Rolli-Derkinderen, M., 2013. The digestive neuronal-glial-epithelial unit: a new actor in gut health and disease. Nat. Rev. Gastroenterol. Hepatol. 10, 90–100. Nicholson, J.K., Holmes, E., Kinross, J., Burcelin, R., Gibson, G., Jia, W., Pettersson, S., 2012. Host-gut microbiota metabolic interactions. Science 336, 1262–1267. Numakawa, T., Matsumoto, T., Numakawa, Y., Richards, M., Yamawaki, S., Kunugi, H., 2011. Protective action of neurotrophic factors and estrogen against oxidative stress-mediated neurodegeneration. J. Toxicol., 405194. Patel, B., Patel, N., Fidalgo, S., Ranson, R., Saffery, J., Yeoman, M., 2012. Age-related changes in colonic motility, faecal output and the properties of faecal pellets in the mouse. Neurogastroenterol. Motil. 24 129–129. Peck, C.J., Samsuria, S.D., Harrington, A.M., King, S.K., Hutson, J.M., Southwell, B.R., 2009. Fall in density, but not number of myenteric neurons and circular muscle nerve fibres in guinea-pig colon with ageing. Neurogastroenterol. Motil. 21, 1075–e1090. Phillips, R.J., Hargrave, S.L., Rhodes, B.S., Zopf, D.A., Powley, T.L., 2004a. Quantification of neurons in the myenteric plexus: an evaluation of putative panneuronal markers. J. Neurosc. Methods133, 99–107. Phillips, R.J., Kieffer, E.J., Powley, T.L., 2003. Aging of the myenteric plexus: neuronal loss is specific to cholinergic neurons. Auton. Neurosci. 106, 69–83.

Phillips, R.J., Kieffer, E.J., Powley, T.L., 2004b. Loss of glia and neurons in the myenteric plexus of the aged Fischer 344 rat. Anat. Embryol. 209, 19–30. Phillips, R.J., Pairitz, J.C., Powley, T.L., 2007. Age-related neuronal loss in the submucosal plexus of the colon of Fischer 344 rats. Neurobiol. Aging 28, 1124–1137. Phillips, R.J., Powley, T.L., 2001. As the gut ages: timetables for aging of innervation vary by organ in the Fischer 344 rat. J. Comp. Neurol. 434, 358–377. Phillips, R.J., Powley, T.L., 2007. Innervation of the gastrointestinal tract: patterns of aging. Auton. Neurosci. 136, 1–19. Phillips, R.J., Walter, G.C., Ringer, B.E., Higgs, K.M., Powley, T.L., 2009. Alphasynuclein immunopositive aggregates in the myenteric plexus of the aging Fischer 344 rat. Exp. Neurol. 220, 109–119. Phillips, R.J., Walter, G.C., Wilder, S.L., Baronowsky, E.A., Powley, T.L., 2008. Alphasynuclein-immunopositive myenteric neurons and vagal preganglionic terminals: autonomic pathway implicated in Parkinson’s disease? Neuroscience 153, 733–750. Rao, S.S., Go, J.T., 2010. Update on the management of constipation in the elderly: new treatment options. Clin. Interv. Aging 5, 163–171. Rhee, R.h., Pothoulakis, C., Mayer, E.A., 2009. Principles and Clinical Implications of the brain-gut-enteric microbiota axis. Nat. Rev Gastroenterol. Hepatol. 6, 306–314. Saffrey, M.J., 2004. Ageing of the enteric nervous system. Mech. Ageing Dev. 125, 899–906. Sandstrom, O., El-Salhy, M., 2001. Age-related changes in neuroendocrine system of the gut. A possible role in the pathogenesis of gastrointestinal disorders in the elderly. Minireview based on a doctoral thesis. Upsala J. Med. Sci. 106, 81–97. Santer, R.M., 1994. Survival of the population of NADPH-diaphorase stained myenteric neurons in the small intestine of aged rats. J. Auton. Nerv. Syst.49, 115–121. Santer, R.M., Baker, D.M., 1988. Enteric neuron numbers and sizes in Auerbach’s plexus in the small and large intestine of adult and aged rats. J. Auton. Nerv. Syst. 25, 59–67. Santer, R.M., Conboy, V.B., 1990. Prenatal undernutrition permanently decreases enteric neuron number and sympathetic innervation of Auerbach’s plexus in the rat. J. Anat. 168, 57–62. Savidge, T.C., Sofroniew, M.V., Neunlist, M., 2007. Starring roles for astroglia in barrier pathologies of gut and brain. Lab. Invest. 87, 731–736. Southwell, B.R., Koh, T.L., Wong, S.Q., King, S.K., Ong, S.Y., Lee, M., Farmer, P.J., Peck, C.J., Sutcliffe, J.R., Stanton, M.P., Keck, J., Cook, D.J., Chow, C.W., Hutson, J.M., 2010. Decrease in nerve fibre density in human sigmoid colon circular muscle occurs with growth but not aging. Neurogastroenterol. Motil. 22 (439–445), e106. Takahashi, T., Qoubaitary, A., Owyang, C., Wiley, J.W., 2000. Decreased expression of nitric oxide synthase in the colonic myenteric plexus of aged rats. Brain Res. 883, 15–21. Thrasivoulou, C., Soubeyre, V., Ridha, H., Giuliani, D., Giaroni, C., Michael, G.J., Saffrey, M.J., Cowen, T., 2006. Reactive oxygen species, dietary restriction and neurotrophic factors in age-related loss of myenteric neurons. Aging cell 5, 247–257. Tiihonen, K., Ouwehand, A.C., Rautonen, N., 2010. Human intestinal microbiota and healthy ageing. Ageing Res. Rev. 9, 107–116. Van Ginneken, C., Schafer, K.H., Van Dam, D., Huygelen, V., De Deyn, P.P., 2011. Morphological changes in the enteric nervous system of aging and APP23 transgenic mice. Brain Res. 1378, 43–53. Wade, P.R., Cowen, T., 2004. Neurodegeneration: a key factor in the ageing gut. Neurogastroenterol. Motil. 16 (Suppl. 1), S19–S23. Wiskur, B., Greenwood-Van Meerveld, B., 2010. The aging colon: the role of enteric neurodegeneration in constipation. Curr. Gastroenterol. Rep. 12, 507–512. Wu, M., Van Nassauw, L., Kroese, A.B., Adriaensen, D., Timmermans, J.P., 2003. Myenteric nitrergic neurons along the rat esophagus: evidence for regional and strain differences in age-related changes. Histochem. Cell Biol. 119, 395–403.

i Please cite this article as: Jill Saffrey, M., Cellular changes in the enteric nervous system during ageing. Dev. Biol. (2013), http://dx.doi. org/10.1016/j.ydbio.2013.03.015