Date of first publication: 31-03-2018

Colin Altena, University of Amsterdam

 

Abstract

Psychiatric disorders are a prevalent and burdening pathology for both the sufferer and societal healthcare. Research has shown that a single treatment with antibiotics increases the risk of depression, and that there is comorbidity between Irritable Bowel Syndrome and psychiatric disorders. Findings like these suggest a link between the gut microbiome and mental wellbeing. This review gives a summary of current research on the effect of intestinal microbiotal differences on behaviour, using rodent models. It is shown that imbalances in the gut microbiome influence anxiety- and depression-like behaviour. Specific microbiome transplants are able to causally reduce or increase symptoms of anxiety and depression. This field of research provides us with strong indications that there is a bidirectional link between the gut microbiome and anxiety- and depressive-like behaviour.

 

Introduction

Environmental influences are known to have complex interactions with human genes, which in turn influence behaviour. One such influential factor is the effect of the micro-organisms that co-exist in the environment, and that are present on and inside the human body. The presence of micro-organisms preceded human evolution, with micro-organisms living in symbiosis with animals long time prior. Animal intestinal micro-organisms have had their own evolutionary selection process in the intestinal environment, in which they consume nutrients and excrete metabolites. Some of these metabolites are known to be precursors or essential nutrients for human metabolism and development (Clarke, Grenham, Scully, Fitzgerald, Moloney et al., 2013). Recent research indicates that disturbances in the human gut microbiome can result in a wide range of pathologies.

Different research models have been used to investigate the role of the gut microbiome on behaviour, each with their own pros and cons. Germ-free models are animal models in which the animal is bred in a sterile environment and bears no living micro-organisms. This germ-free model can show the direct effect of the absence or presence of (specific) micro-organisms, but these animals may suffer from stunted nervous system development (Luczynski, Neufeld, Oriach, Clarke, Dinan et al., 2016). Antibiotic models are animal models in which the lab animal receives a mix of various antibiotics. This antibiotics model provides a more naturally developed nervous system, but side effects of the antibiotics and the presence of non-bacterial micro-organisms can interfere with the experiment. Probiotic models are animal models in which specific micro-organisms are introduced to an animal. The benefit of probiotic models lies in the fact that the effect of specific micro-organisms can be causally tested. Most studies mentioned use ‘Specific Pathogen Free’ (SPF) animals for control conditions. These animals are bred to be general lab animals that are free of specific hazardous pathogenic organisms. This review uses a combination of these abovementioned models to give an overview of the behavioural effects of differences in gut microbiota.

Bacterial disturbances in the microbiome can increase risk of depression; a single treatment with antibiotics increases the susceptibility to depression in human patients (Lurie, Yang, Haynes, Mamtani & Boursi, 2015). Additional research shows that there is comorbidity between Irritable Bowel Syndrome and psychiatric disorders (Mayer, Savidge, Shulman, 2014). These findings suggest a link between gut microbiome health and mental wellbeing. However, it is not clear what types of behaviours are influenced by the gut microbiome, and how these influences may contribute to the development of mental illnesses. Therefore this review will summarize and discuss the current view on gut microbiome-behaviour interactions using rodent models.

Mental illnesses are complex disorders with symptoms ranging from anxiety, depression and apathy to susceptibility to addiction. To investigate the role of the gut microbiome on mental disorders, it is therefore valuable to quantify its effect on this spectrum of behavioural symptoms. Due to the extensive metabolic microbiome-host interactions it is also unlikely that differences in gut microbiome only affect a single type of behaviour. The following paragraphs will therefore summarize the effects of differences in gut microbiome on these varying types of behaviour. Studies using either rat models or mouse models seem to provide contradicting information in various cases, therefore these animal models will be considered separately.

 

Anxiety-like behaviour

Anxiety disorders are a prominent mental disorder with very burdening effects for both individuals suffering from the disorder as the society as a whole. Anxiety can also occur as a symptom of other psychiatric disorders, making it a strong indicator of mental discomfort. Rodents are frequently used to study anxiety and anxiety-like behaviour. Anxiety is mainly quantified by the proportion of time the rodent freezes, flees, explores or resides in adverse locations. Tests used to quantify anxiety are mainly the ‘elevated plus maze’, the ‘open field test’ and the light/dark-box. These tests use the innate instinct of rodents to avoid open and illuminated spaces to avoid risk of (aerial) predation. In these tests the time the animal spends in aversive locations (open or illuminated areas) is measured as an indication of how anxious the animal is in the specific condition. The more time the animal spends in an open (central) or illuminated area, the less anxious the animal is assumed to be. The effects of microbiome alterations on these behavioural tests in mice and rats will now be considered separately.

 

Mice

Research by Zheng, Zeng, Zhou, Liu, Fang et al (2016) showed a decrease in anxiety in mice depleted of their microbiome, indicated by less aversion against open spaces. This was tested by placing germ-free Kunming mice and SPF control mice in an open field test. The germ-free mice had a significantly larger motion distance in the innermost 25 percent of the open field compared to the SPF mice, indicative of a reduction in anxiety. The same research group showed a link between fecal transplantation and the transfer of a pathological condition. The researchers performed a fecal transplantation from depressed human patients to germ-free mice. The mice receiving a fecal transplant from depressed human patients showed decreased motion distance in the innermost 25 percent of the open field compared to mice receiving fecal transplantation from healthy human patients (Zheng et al., 2016). Using the NMRI mouse strain it was also shown that absence of the gut microbiome leads to more motion distance in open areas, and the presence of a correlation between fecal transplants and behavioural changes (Heijtz, Wang, Anuar, Qian, Björkholm et al., 2011). This was tested by comparing the central motion distance of germ-free NMRI mice and SPF mice in the open field test. In the open field test the germ-free mice showed an increased motion distance in the centre of the open field compared to SPF mice. In addition, an elevated plus maze experiment showed a considerable increase in open arm visits and visits to the open arm’s end in the germ-free mice group compared to the SPF group. After re-colonization of the germ-free NMRI mice with the microbiome of the SPF mice, the open arm visits decreased (Heijtz et al. 2011). These studies using germ-free mice models show a reduction in anxiety-like behaviour compared to regular mice, which is reversed by re-inoculation of the microbiome by fecal transplantation.

Desbonnet, Clarke, Traplin, O’Sullivan, Crispie et al. (2015) showed a positive correlation between antibiotics-treatment in mice and the time spent in averse (anxiety-triggering) conditions. Antibiotics-treated mice and untreated controls were used for a light/dark box experiment. The mice treated with antibiotics spent more time in the aversive light part of the box compared to controls (Desbonnet et al., 2015). Another light/dark box experiment conducted by Bercik, Denou, Collins, Jackson, Lu et al. (2011) showed that fecal transplants from anxiety-resistant mice decreased anxious behaviour in recipient mice, compared to recipient mice receiving fecal transplants from anxiety-prone mice. This was tested by using two different mouse strains, the anxiety-prone BALB/C mice and the anxiety-resistent NIH Swiss mice. Germ-free variants of each strain received a fecal transplant from the other strain. The transplanted animals and original controls were subjected to a step-down avoidance test, in which the innate aversion against heights is measured by step-down latency. Anxiety-sensitive germ-free BALB/C mice receiving a fecal transplant from anxiety-resistant NIH Swiss mice spent more time in the aversive light chamber, compared to germ-free mice that received a fecal transplant from anxiety-prone BALB/C mice. Germ-free NIH Swiss (anxiety-resistant) mice that received a fecal transplantation from anxiety-sensitive BALB/c mice also displayed a large increase in step-down latency in a step-down avoidance test. Germ-free BALB/c mice receiving a fecal transplant from NIH Swiss mice showed a significant decrease in step-down latency compared to BALB/c controls (Bercik et al. 2011).

From abovementioned studies it can be concluded that germ-free mice show less anxiety-like behaviour in the elevated plus maze and open field compared to controls with regular gut microbiome (Zheng et al., 2016; Heijtz et al. 2011). Mice treated with antibiotics, which reduce gut microbiome diversity, also decrease anxiety-like symptoms as measured using the light/dark-box (Desbonnet et al., 2015). It is also shown that microbiomal transfer by fecal transplantation induces a similar behavioural profile in the recipient as shown by the fecal donor. Mice receiving fecal transplantations from depressed human patients showed increased signs of anxiety-like behaviour (Zheng et al., 2016). Anxiety-prone mice receiving fecal transplants from an anxiety-resistant strain showed a decrease in anxiety-like behaviour, whereas anxiety-resistant mice receiving fecal transplants from an anxiety-prone strain increased anxiety-like behaviour (Bercik et al. 2011).

 

Rats

Germ-free rats seem to have increased anxiety compared to SPF rats (Crumeyrolle-Arias, Jaglin, Bruneau, Vancassel, Cardona et al., 2014). This increase in anxiety-like behaviour was accompanied by a 2.8-fold increase in corticosterone after testing. In a comparison between germ-free and SPF stress sensitive F344 rats, the germ-free rats spent less time in the centre during the open field test and defecated more compared to the SPF rats (Crumeyrolle-Arias et al., 2014). Kelly, Borre, O’Brien, Patterson, Aidy et al. (2016) showed that fecal transplants from depressed human patients increased anxiety in rats compared to rats receiving transplants from healthy human patients. First the Sprague-Dawley rats received a mix of antibiotics to reduce their original gut microbiome, after which the rats received a fecal transplantation from either depressed or healthy humans patients. The rats receiving a fecal transplant from a depressed human patient showed decreased visits to open arms in the elevated plus maze compared to rats receiving fecal transplants from healthy human subjects. These rats also showed a decrease in time spent in the center of the open field in the open field test compared to rats receiving healthy fecal transplants. These findings were accompanied by a reduction in intestinal micro-organism abundance and diversity in the humans suffering from depression and in the rats receiving their fecal transplant (Kelly et al., 2016).

In rats, anxiety seems to increase in the absence of the gut microbiome (Crumeyrolle-Arias et al., 2014), as shown by behavioural and physiological markers. Fecal transplants from depressed human patients increased anxiety-like behaviour in rats compared to rats receiving a fecal transplant from a healthy subject. These findings suggest a link between microbiome transfer and pathology transfer from donor to host.

 

Compared to mice, rats seem to be affected differently by changes in the gut microbiome. Studies show an increase in anxiety-like behaviour in the absence of gut micro-organisms in rats, whereas mice show a decrease in anxiety-like behaviour in the absence of gut micro-organisms. Currently, only one study using germ-free rats in this setup has been conducted; the body of evidence for studies using mice is higher. It is therefore debatable whether the rat study overthrows the results found in mice. More homogenous results are found in fecal transplant studies. Mice receiving fecal transplants from either depressed humans or from anxiety-prone mice became more anxious compared to controls. Rats receiving a fecal transplant from depressed humans also became more anxious compared to controls. Inversely, stress-sensitive mice and rats display a decrease in anxiety-like behaviour when receiving a fecal transplant from a healthy human (rats/mice study) or a anxiety-resistant mouse (mice study) subject. Both these findings are indications that behavioural pathology is transmitted upon fecal transfer, and that the pathology can be reduced by a fecal transplant from a healthy subject.

Depressive-like behaviour

Due to the suspected influence of the microbiomal tryptophan synthesis on serotonergic networks, microbial imbalances may make individuals more sensitive for depressive-like behaviour (Clarke et al., 2013; Mayer et al., 2014). There is a large amount of overlap between depression and anxiety-like behaviour; therefore many of the previous mentioned anxiety-tests have also been linked to depression. Tests specifically trying to quantify depressive-like behaviour are the ‘forced swimming test’ and the ‘tail suspension test’. In both tests the subjects are placed in an adverse environment/situation, in which immobility time is measured. The amount of immobility/passiveness is used as a measurement for depressive-like behaviour. The rationale of these tests is that the more depression-prone an animal is, the less likely it is to try to escape and the more likely it is to gives up in despair (reflected by an increase in immobility).

 

Mice

Germ-free mice show less signs of depression compared to SPF mice, as tested by Zheng et al. (2016). Germ-free and SPF Kunming mice were tested using the forced swimming test. Germ-free Kunming mice showed less immobility time in the forced swimming test compared to SPF Kunming mice. After fecal transplantation from depressive human patients to germ-free mice, there was an increase in immobility time in the forced swimming test, indicating an increase in depression caused by the fecal transplant. Germ-free Kunming mice receiving a fecal transplant from depressive human patients also showed more immobility time in the tail suspension test compared to controls, but no tail suspension test was conducted prior to the fecal transplantation (Zheng et al., 2016).

 

Rats

Fecal transplantation from depressed human patients induced physiological markers of depression in research conducted by Kelly et al. (2016). Rats first received an antibiotics treatment to reduce endogenous microbial diversity. Afterwards they received a fecal transplant from depressed human patients or healthy controls. The rats receiving a fecal transplant from depressed human patients showed an increased plasma kynurenine/tryptophan ratio, which is a physiological indicator for depression (Kelly et al., 2016). These findings did not translate in behavioural changes, no significant results were found using the forced swimming test.

In mice, the absence of micro-organisms decreases the prevalence of depressive-like symptoms. Transplanting the microbiome of depressive human patients to mice increases the prevalence of depressive-like symptoms compared to mice that received a fecal transplant from a healthy human subject. Rats receiving a fecal transplant from depressed human subjects showed increased physiological markers of depression, but did not show a behavioural change. This study currently is the only of its kind and uses a mere 13 rats, so these behavioural effects should be studied more extensively.

 

 

Reward sensitivity

Reward insensitivity or apathy has a high comorbidity with depression, but also occurs in individuals that do not suffer from depression. Reward sensitivity imbalances can have an alternative pathological profile compared to depression, mostly linked to dopaminergic networks (Kiraly, Walker, Calipari, Labonte, Issler et al., 2016). Therefore, it might be relevant to consider reward sensitivity separately. Even though symptoms of reward insensitivity often overlap with symptoms of depression, there are available tests that target reward insensitivity specifically. Tests like sucrose consumption and self-administration of preferable food pallets or drugs are known markers of reward (in)sensitivity in rodents.

 

Mice

Research by Kiraly et al. (2016) indicates higher reward sensitivity in mice with depleted microbiota compared to controls. Mice were treated with a combination of antibiotics to deplete the endogenous microbiome. A conditioned place preference test was conducted to test the effect of the gut microbiome on reward sensitivity in mice. Antibiotics-treated mice and control mice received two cocaine injections in two days at a specific location in an experimental setup. The time the mice spent at the site of injection was measured the day after the last injection. Antibiotics-treated mice showed a significant increase in time spent at the site of injection (Kiraly et al., 2016).

 

Rats

Research by Kelly et al. (2016) indicates that the microbiome-profile of depressed patients causes a decrease in reward-sensitivity in rats. Antibiotics-treated Sprague-Dawley rats receiving a fecal transfer from depressed human subjects consumed less sucrose in the sucrose preference test compared to rats that received a fecal transplant from healthy human subjects, with no reduction in overall water intake. There was a reduction in micro-organism abundance and diversity in the group receiving the ‘depressed’ fecal transplant compared to subjects receiving a healthy fecal transplant. (Kelly et al., 2016).

Fecal transplants from depressed human subjects to microbiome-deficient rats seem to decreased reward-sensitivity, displayed by a decrease in sucrose consumption. The absence of a gut microbiome in mice however seems to increase reward sensitivity, measured by an increased motivation for cocaine in microbiome-depleted mice.

 

Discussion

This review set out to give an overview of the behavioural effects that are caused by differences in gut microbiota in rodents. Categories of behaviour (anxiety, depression and reward sensitivity) and rodent species have been separately considered with the following results. Germ-free mice showed significant decreases in anxiety- and depression-like behaviour and increased reward sensitivity compared to control mice, indicating the regulating role of the microbiome in these types of behaviour. Germ-free rats showed contradicting results compared to mice in tests on anxiety- and depression-like behaviour and reward sensitivity. Germ-free rats showed an increase in anxiety-like behaviour and suffered from decreased sensitivity to rewards, as measured by sucrose consumption. However, both mice and rats showed an increase in aversion towards open fields, illuminated areas and an increase in immobility-time after receiving a fecal transplant from human subjects suffering from depression, compared to controls receiving a fecal transplant from healthy human subjects. This indicates that whole-microbiome transplants not only introduce donor micro-organisms to the recipient, but also introduce the donor’s behavioural traits to the recipient.

It therefore can be concluded that alterations in gut microbiome composition seem to influence sensitivity to depressive- and anxiety-like behaviour and reward sensitivity in both rats and mice. Studies using fecal transplants from depressed human patients show a strong causal relationship between imbalanced microbiome profiles (as defined by a reduction in abundance and diversity of intestinal micro-organisms) and adverse changes in behaviour. This causal link between intestinal microbiome alterations and mental health confirm the suspected relationship between antibiotics exposure and an increase in depression-susceptibility (Lurie et al., 2015) and relationship between Irritable Bowel Syndrome and psychiatric disorders (Mayer et al., 2014).

The discrepancy between the effects of the ‘germ-free’ status on behaviour of rats and mice could be explained in various ways. First, it has to be stated that the body of evidence is severely limited in this relatively new field of research. In particular studies using rat models are scarce, with the present studies using a modest amount of test subjects. The limited availability of evidence makes the results more prone to errors, making replication a necessity. Second, the behavioural tests used in abovementioned studies only test a very small component of behaviour, whereas pathologies like anxiety and depression are complex and can be expressed in differing context-dependant ways. It is debatable whether a single-trial test can really quantify such a complex type of behaviour. Third, the genetic makeup in general could also play a role. Rats tend to show more complex behaviour compared to mice, which might make rats more susceptible to disruptions. It is also possible that the gut microbiome as of itself has a regulatory role in stress-susceptibility. The microbiome forms a complex ecosystem of interactions between metabolites that in turn affects the host. Not all interactions between host and microbiome metabolites are known, but a potential route of effect could be the tryptophan-serotonin synthesis pathway (Clarke et al., 2013) the GABA and catecholamines metabolism or the metabolites and cytokines that are released during the immune response induced by micro-organisms (Mayer et al., 2014). Balanced microbiomal ecosystems with higher microbial diversity could be more resistant to disruptions in these pathways, whereas imbalanced ecosystems might be more susceptible to disruption. This way the microbiome as a whole could pose as a ‘buffering’ or regulating agent between host and environment. The effects of fecal transfers between anxious/depressive subjects and non-anxious/-depressive animals also supports the theory that a specific gut microbiome profile can make an animal either more susceptible or resistant to anxiety- and depression-like behaviour. Additional indications for this theory are the fact that apart from the microbiome influencing behaviour, behaviour and environmental factors seem to affect the microbiome as well. Stress (chronic social defeat) alters the gut microbiome (Szyszkowicz, Wong, Anisman, Merali, Audet et al., 2017; Bharwani, Mian, Foster, Surette, Bienenstock et al., 2015). Therefore it is likely that there is a bidirectional link between the brain and gut microbiome, with both entities influencing each other.

Several of the abovementioned studies investigated whether there are specific microbial species that contribute to stress-susceptibility. These studies seem to find differing and even contradicting markers (differences in the microbiome at the phylum, family and genus level) in the microbiome profile, suggesting the behavioural alterations cannot be contributed to a single set of micro-organisms. A finding that all papers did report was that microbiome richness and diversity is positively correlated with stress-resistance, and associated anxiety-resistance. It is possible that any microbial imbalance could lead to behavioural effects, independent of specific alterations in sub-strains of bacteria. There are indications that the gut microbiome influences the synthesis of tryptophan and consequently also serotonin, which is a known neurotransmitter that plays a role in depression (Clarke et al., 2013; Mayer et al., 2014; Kelly et al., 2016). This would explain why any disturbance in the gut microbiome would lead to serotonin-related pathologies like depression. The exact mechanisms of action from microbiome to altered behaviour are not fully known and the serotonin hypothesis has yet to be confirmed.

A critical note on the germ-free models has to be made, as the absence of a microbiome in pre-mature animals stunts the development of the nervous system (Luczynski et al., 2016). Abovementioned studies using these models therefore could give skewed or non-generalizable results on the effects of microbiomal manipulations. The antibiotic model also comes with its caveats, as antibiotics have side-effects (especially when multiple are used simultaneously). Probiotic models are animal models in which specific micro-organisms are introduced to an animal. Additional research using experimental designs characterized by probiotic and adult animals could specify the effects of microbial manipulations on a more naturally developed nervous system. In this case, naturally developed adult animals might give more reliable results, and specific beneficial micro-organisms could be pin-pointed by administering these organisms as probiotics.

The strong correlation between microbial manipulation and depressive- and anxiety-like behaviour also has implications for human studies. It is advised to measure alterations in gut microbiome in subjects undergoing pharmaceutical and/or behavioural therapy for depression or anxiety disorders. Extra control-conditions should also be considered for microbiotal confounding variables in pharmaceutical research, knowing the large effect of the microbiome on stress, anxiety and depression resistance. If supplementation or transplantation of specific micro-organisms can alter behaviour and mental states, new forms of interventional therapy could be designed. All things considered, gut-brain-axis research shows great promises for new routes of treatment in a wide range of psychiatric disorders.

Following from these findings it can be concluded that there is a strong correlation between gut microbiome alterations and behavioural changes, especially in the domain of anxiety- and depressive- like behaviour. Microbiomal transplants seem to not only carry over the micro-organisms to the donor, but also induce donor-like behaviour in the transplant-recipient. The intestinal microbial ecosystem therefore seems to be a likely causal factor determining depression- and anxiety-like behaviour.

 

 

References

Bercik, P., Denou, E., Collins, J., Jackson, W., Lu, J., Jury, J., et al. (2011). The intestinal microbiota affect central levels of brain-derived neurotropic factor and behavior in mice. Gastroenterology (2011) 141:599–609.

Bharwani, A., Mian, M.F., Foster, J.A., Surette, M.G., Bienenstock, J., Forsythe, P. (2015). Structural & functional consequences of chronic psychosocial stress on the microbiome & host. Psychoneuroendocrinology (2016) 63:217–227.

Clarke, G., Grenham, S., Scully, P., Fitzgerald, P., Moloney, R.D., Shanahan, G., et al. (2013). The microbiome-gut-brain axis during early life regulates the hippocampal serotonergic system in a sex-dependent manner. Molecular Psychiatry (2013) 18:666-673.

Crumeyrolle-Arias, M., Jaglin, M., Bruneau, A., Vancassel, S., Cardona, A., Daugé, V., et al. (2014). Absence of the gut microbiota enhances anxiety-like behavior and neuroendocrine response to acute stress in rats. Psychoneuroendocrinology (2014) 42:207—217

Desbonnet, L., Clarke, G., Traplin, A., O’Sullivan, O., Crispie, F., Moloney, R., et al. (2015). Gut microbiota depletion from early adolescence in mice: Implications for brain and behaviour. Brain, Behavior, and Immunity (2015) 48:165–173.

Heijtz, R.D., Wang, S., Anuar, F., Qian, Y., Björkholm, B., Samuelsson, A., et al. (2011). Normal gut microbiota modulates brain development and behaviour. Proceedings of the National Academy of Sciences of the United States of America (2011) 108:3047-3052.

Kelly, J.R., Borre, Y., O’Brien, C., Patterson, E., Aidy, S.E., Deane, J., et al. (2016). Transferring the blues: Depression-associated gut microbiota induces neurobehavioural changes in the rat. Journal of Psychiatric Research (2016) 82:109-118

Kiraly, D.D., Walker, D.M., Calipari, E.S., Labonte, B., Issler, O., Pena, C.J., et al. (2016). Alterations of the host microbiome affect behavioral responses to cocaine. Scientific Reports (2016) 6: 35455.

Luczynski, P., McVey Neufeld, K., Oriach, C. S., Clarke, G., Dinan, T. G., & Cryan, J. F. (2016). Growing up in a Bubble: Using Germ-Free Animals to Assess the Influence of the Gut Microbiota on Brain and Behavior. International Journal of Neuropsychopharmacology, 19(8), pyw020. doi:10.1093/ijnp/pyw020

Lurie, I., Yang, Y., Haynes, K., Mamtani, R., Boursi, B. (2015). Antibiotic exposure and the risk for depression, anxiety, or psychosis: A nested case-control study. Journal of Clinical Psychiatry 2015 76:1522–1528.

Mayer, E.A., Savidge, T., Shulman, R.J. (2014). Brain–gut microbiome interactions and functional bowel disorders. Gastroenterology (2014) 146:1500–1512.

Szyszkowicz, J.K., Wong, A., Anisman, H., Merali, Z., Audet, M.C. (2017). Implications of the gut microbiota in vulnerability to the social avoidance effects of chronic social defeat in male mice. Brain, Behavior, and Immunity (2017) 66:45–55

Zheng, P., Zeng, B., Zhou, C., Liu, M., Fang, Z., Xu, X., et al. (2016). Gut microbiome remodeling induces depressive-like behaviors through a pathway mediated by the host’s metabolism. Molecular Psychiatry (2016) 21:786–796.