Correlations between behavior and hormone concentrations or gut microbiome imply that domestic cats (Felis silvestris catus) living in a group are not like 'groupmates'.

Domestic cats (Felis silvestris catus) can live in high densities, although most feline species are solitary and exclusively territorial animals; it is possible that certain behavioral strategies enable this phenomenon. These behaviors are regulated by hormones and the gut microbiome, which, in turn, is influenced by domestication. Therefore, we investigated the relationships between the sociality, hormone concentrations, and gut microbiome of domestic cats by conducting three sets of experiments for each group of five cats and analyzing their behavior, hormone concentrations (cortisol, oxytocin, and testosterone), and their gut microbiomes. We observed that individuals with high cortisol and testosterone concentrations established less contact with others, and individuals with high oxytocin concentrations did not exhibit affiliative behaviors as much as expected. Additionally, the higher the frequency of contact among the individuals, the greater the similarity in gut microbiome; gut microbial composition was also related to behavioral patterns and cortisol secretion. Notably, individuals with low cortisol and testosterone concentrations were highly tolerant, making high-density living easy. Oxytocin usually functions in an affiliative manner within groups, but our results suggest that even if typically solitary and territorial animals live in high densities, their oxytocin functions are opposite to those of typically group-living animals.

Introduction

During evolution, ecological factors, including food availability, influence the development of group and solitary living behavior in animals [1]. Concentrated food distribution has led to selection for communal feeding, which has resulted in cooperative sociality [1]. Moreover, such sociality could have been enhanced by human interventions that affected ecological conditions, such as availability of new food resources in garbage dumps and destruction of pre-existing habitats [2]. Although most wild felids live alone [3], domesticated cats (Felis silvestris catus) live in high densities and interact with each other [4]. Incidentally, as humans began to settle down and start cultivation, cats acquired a new niche where food resources, such as rodents, were abundant, which led to an ability to live in high densities. In this process of acquiring a new niche, temperamental selection for communal feeding must have occurred in the felids, although not to an extent similar to that in the canids [5].

When solitary and exclusively territorial animals begin to live in groups due to domestication, it is natural to wonder whether their group-formation strategies are similar to those of typical group-living animals. They may be ’groupmates,’ similar to other mammals that live in groups, or they may be in conflict with each other, in spite of living in high densities, due to their inherent solitary nature. Therefore, investigating how cats establish groups (compared to the group formation of other animal species) can help clarify the effects of ecological environment or nature of species on the process of group formation. Moreover, if cats live as ’groupmates,’ then it would be possible to understand the mechanisms of animal group formation by exploring the factors influencing the development of ’groupmates’ among cats.

Previously, the social behavior of animals living in groups has been studied on the basis of endocrine activity, such as glucocorticoids (GCs), testosterone, and oxytocin [6-17]. The primary physiological function of GCs is to increase the metabolism of glucose (energy) as per the requirement of behavioral responses. This energy production is necessary when an animal is faced with some threat, and it has been demonstrated that individuals with higher GC concentrations exhibit enhanced aggression or fear responses in various species [18-21]. Testosterone, a sex hormone belonging to the androgen family, is positively correlated with aggression [22-26]. Notably, cortisol and testosterone interactively regulate aggressive behaviors [27,28]. In addition, the cortisol concentrations in domestic cats are lower than those in wild cats [29]. Furthermore, a direct association has been established between cortisol and aggression in feral female cats [30]. A previous study has also shown that cortisol can increase testosterone secretion in vitro [31]. However, to date, no survey has explored the components of social behavior that may be related to GC concentrations in group-living cats. Oxytocin, a peptide hormone, is well-known for its role in influencing reproductive behaviors, such as mating and maternal care, and regulates diverse social behaviors related to ’tend-and-defend’ within a group [32-34]. Therefore, if domestic cats co-habit as ‘groupmates,’ oxytocin would probably induce a ’tending’ behavior among them.

Changes in food resource niches can modulate gut microbiome [35]. Incidentally, there exists a brain-gut-axis, in which gut microbes interact with the gut as well as the brain; one of the pathways via which gut microbes influence brain function is through endocrine activity, such as GCs, sex steroids, and neuropeptides [36]. The microbes can produce these hormones directly [37,38] or indirectly [39]. Notably, the microbiome has a great influence on the hypothalamic-pituitary-adrenal (HPA) axis [40,41] and the hypothalamic-pituitary-gonadal (HPG) axis [42]. Recent empirical studies have revealed that the gut microbiome can alter host sociality by modulating oxytocin secretion from the hypothalamus [43]. Therefore, changes in a host’s niche can modulate gut microbiome and influence host sociality via such endocrine pathways.

Another notable finding in this context is the similarities in gut microbiome among groupmates that can arise due to microbial transmission through social interactions, or via shared environments in human and non-human primates [44-46]. A microbiome can generate biochemical signals that are used in the host’s social communications in insects such as drosophila [47-49], or it can directly affect the host’s nervous system in rodents [50-52], which, in turn, can affect host social behavior. Such observations suggest that the exchange of microbiomes and biochemical similarity can enhance group formation, resulting in close connections within groups.

The aim of the present study was to investigate the nature of interactions among cats within a group, and how such interactions are associated with various internal factors. Hence, we analyzed the relationships between the gut microbiome and concentrations of different hormones, such as cortisol, testosterone, and oxytocin, and the potential associations with social behavior of the group-living cats. Our four hypotheses were as follows: 1. individuals with high cortisol and testosterone concentrations will be less socially tolerant, and they will exhibit more aggression-related behaviors than affiliative behaviors; 2. individuals with high oxytocin concentrations will exhibit more affiliative behaviors than aggressive responses; 3. there is a relationship between the composition of gut microbiomes and hormone concentrations in an individual’s body; 4. the gut microbiomes of individuals that have frequent contact among them are similar. Incidentally, to the best of our knowledge, this is the first attempt to reveal the role of oxytocin in conspecific social behaviors in a solitary mammal.

Methods

Subjects

Cats living in a shelter (Tanpoponosato, Kanagawa, Japan) participated in this experiment. There were 10 males and five females, with a mean age of 4.2 ± 2.3 years. All the cats were neutered, and they were divided into three groups of five cats each, randomly. All cats were housed in one room at Azabu University during the experiment. Further details regarding the participating cats are provided in Supplementary Information (S1 Table).

Experimental setting

Each group of five cats was housed in a room (4 m × 7.5 m) for two weeks. More than five beds were set up in the room so that the cats could choose where to rest. There were five litter trays, two food bowls, and two water bowls. The cats were able to eat food and drink water at any time. Two cameras (HX-A1H, Panasonic, Japan) and two infrared lights were set up at the top corners of the room. The infrared lights were switched on during the night, enabling observations in the dark. Every alternate day, the behavior of the cats from 21:00 h at night to 7:00 h the next morning were analyzed. Their urine and feces samples were collected every morning (between 7:00 h and 10:00 h). The experiment was conducted one group at a time, and the three groups stayed in the same room. In addition, all the cats ate the same food. All cats were fed a complete and balanced commercially available kibble diet. The composition of the cat’s gut microbiome was not affected by room environment or food. All protocols are carried out in accordance with relevant guidelines and regulations. All experimental procedures were approved by the Animal Ethics Committee of Azabu University (#180410–1).

Behavioral analysis

Based on observations of the first group throughout the day, interaction among cats was quite low during the day, and they were more active at night. Therefore, behavioral analyses were henceforth conducted every alternate day from 21:00 h to 7:00 h the next morning for two weeks (70-h observation for each cat). We focused on each cat and recorded its behaviors and those of partners. The recorded behaviors and their definitions are listed in Supplementary Information (S2 Table). Active (subject cat initiating the behavior) and passive (subject cat receiving the behavior) behaviors were marked. BORIS v.7.0.8 (https://www.boris.unito.it/) was used for behavioral analysis. To check the inter-observer reproducibility, the observers analyzed the same video for 30% of the total video time. The kappa coefficient, which indicates reproducibility, was 0.86.

Assay of urinary hormone concentrations

The urine samples of the cats were collected using a two-tiered litter box immediately after urination. The collected urine samples were centrifuged for 15 min (at 4°C and 3000 rpm) and immediately stored in a freezer at ˗80°C. Sixty-three urine samples, i.e., 4.2 ± 2.4 samples/individual (A minimum of one and a maximum of nine urine samples were collected per individual) were collected during the entire observation period, and all samples were assayed for cortisol concentrations. Oxytocin and creatinine concentrations were analyzed in 62 samples (4.1 ± 2.4 samples/individual); one sample was excluded due to its low quantity. Testosterone concentrations were assayed for 57 samples (3.8 ± 2.0 samples/individual). The assay protocols were based on previous studies [53-56]. The average of the urine hormone concentrations for each individual was used as the hormone baseline for each individual.

Cortisol concentrations

Cortisol concentrations were measured using enzyme-linked immunosorbent assay (ELISA). The undiluted urine samples were dispensed into the wells of the ELISA-plate. The primary antibody was anti-cortisol antibody (ab1949; Abcam Plc., UK) that had been diluted 200,000-fold, and the secondary antibody was mouse IgG-Fc fragment antibody (A90-131A, Bethyl Laboratories, USA) that had been diluted 500-fold. The horseradish peroxidase (HRP) used in this case was Cortisol-3-CMO-HRP (FKA403, Cosmo bio, Japan) that was diluted 1 million-fold. Standard and urine samples were dispensed in 15-μL volumes, and primary antibody, secondary antibody, and HRP in 100-μL volumes. The obtained values of the cortisol concentrations were adjusted for creatinine correction.

Testosterone concentrations

Testosterone concentrations were also measured using the ELISA technique. Based on the concentration, either undiluted urine samples were dispensed into the wells of the ELISA-plate, or the samples were diluted two-fold with a phosphate buffer containing 0.1% bovine serum albumin, before loading. The primary antibody was anti-testosterone 3 CMO antibody (ab35878, Abcam Plc., UK) that had been diluted 25,000-fold, and the secondary antibody was mouse IgG-Fc fragment antibody (A90-131A, Bethyl Laboratories, USA) that had been diluted 500-fold. The HRP used in this process was Testosterone-3-CMO-HRP (FKA101, Cosmo bio Co., Ltd., Japan) that was diluted 1 million-fold. Standard and urine samples were dispensed in 25-μL volumes, and primary antibody, secondary antibody, and HRP in 100-μL volumes. The obtained values of the testosterone concentrations were adjusted for creatinine correction.

Oxytocin concentrations

The commercially available oxytocin ELISA kit (ADI-901-153A-0001, Enzo Life Sciences, Inc., USA) was used for the assay. The urine samples were diluted 50-fold with the assay buffer in the kit and dispensed into the wells of the ELISA-plates. Standard and urine samples were dispensed in 15-μL volumes, and primary antibody and HRP in 50-μL volumes. The obtained values of the oxytocin concentrations were adjusted for creatinine correction.

Creatinine concentrations

The creatinine standard and the urine samples were diluted 100-fold with distilled water and dispensed into a 96-well microplate (AS ONE Co., Ltd., Osaka) at 100 μL each, followed by 50 μL of 1 M NaOH and 50 μL of 1 g/dL trinitrophenol. The plate was left at room temperature (22–26°C) for 20 min, and the absorbance was measured at a wavelength of 490 nm using a microplate reader (MODEL 680XR, Bio-Rad Laboratories, Inc., USA). The samples were used undiluted.

Analysis of gut microbes

Fecal samples were collected from different individuals only when the cats that excreted them were observed. The fecal samples were collected from only eight individuals. For the first group of cats, the feces were collected within 3 h of defecation, stored in a refrigerator at 4°C, processed into a glycerol stock within 24 h, and stored in a ˗80°C freezer. For the second and third groups, approximately 1.0 g of feces was collected within 15 min of defecation using a sterile cotton swab; thereafter, it was placed in a 15 mL tube containing a reagent mixture of 1.0 mL phosphate-buffered saline and 2.0 mL glycerol and dissolved using a bamboo skewer. Although different collection methods were used, the time of the sampling did not change the main composition of the microbiome as long as we collected the part of the feces that is not exposed to the air near the center (Sinha et al. 2016). The use of the swab did not change the composition of the microbiome (Sinha et al. 2016). It was then stored in a ˗80°C freezer. Subsequently, we transferred the intact feces into a tube containing 2.5 mm-diameter cell-grinding beads and analyzed gut microbiomes at the Bioengineering Lab. Co., Ltd.

DNA was extracted from the crushed sample using the MPure Bacterial DNA Extraction Kit (MP Bio Japan, Japan). Then, the concentration of the extracted DNA solution was measured using Synergy H1 (BioTek) and QuantiFluor dsDNA System, and the library was prepared. The library was prepared using the two-step tailed PCR method. The V3–V4 region of the 16S ribosomal RNA (rRNA) gene was amplified using PCR. In the 1st PCR, a solution was prepared by adding 1.0 μl of 10XEx Buffer, 0.8 μl of deoxynucleotide Triphosphates (dNTPs, 2.5mM each), 0.5 μl of Forward primer (10 μM), 0.5 μl of Reverse primer (10 uM), 2.0 μl of Template DNA (0.5 ng/μl), 0.1 μl of ExTaqHS (TaKaRa) (5 U/μl), and 5.1 μl of deionized distilled water (DDW). The 16S Amplicon PCR Forward Primer (5′-ACACTCTTTCCCTACACGACGCTCTTCCGATCT-NNNNN- CCTACGGGNGGCWGCAG-3′) and 16S Amplicon PCR Reverse Primer (5′-GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT-NNNNN-GACTACHVGGGTATCTAATCC-3′) were used [57]. PCR amplification was performed by pre-denaturation at 94°C for 2 min, followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 30 s, and final extension at 72°C for 5 min. The PCR product was washed using AMPure XP beads (BECKMAN COULTER). In the 2nd PCR reaction, a solution was prepared by adding 1.0 μl of 10XEx Buffer, 0.8 ul of dNTPs (2.5mM each), 0.5 μl of Forward primer (10μM), 0.5 μl of Reverse primer (10μM), 2.0 μl of PCR product (max 5 ng/ul), 0.1 μl of ExTaqHS (TaKaRa) (5U/μl), and 5.1 μl of DDW. The 16S Amplicon PCR Forward Primer (5′-AATGATACGGCGACCACCGAGATCTACAC-Index2-ACACTCTTTCCCTACACGACGC-3′) and 16S Amplicon PCR Reverse Primer (5′-CAAGCAGAAGACGGCATACGAGAT-Index1-GTGACTGGAGTTCAGACGTGTG-3′) were used. PCR amplification was performed by pre-denaturation at 94°C for 2 min, followed by 10 cycles of 94°C for 30 s, 60°C for 30 s, 72°C for 30 s, and final extension at 72°C for 5 min. The PCR product was washed using AMPure XP beads (BECKMAN COULTER). The quality of libraries was evaluated using a Fragment Analyzer and a dsDNA 915 Reagent Kit (Advanced Analytical Technologies, Inc., USA) [58]. The purified products were sequenced using the Illumina MiSeq System (Illumina Inc., San Diego, CA, USA). We obtained 428,762 total reads (53,595 ± 4589).

For data analysis, we extracted only those sequences whose start reading was an exact match to the primer used, using fastq_barcode_spliltter in the Fastx toolkit [59]. The primer sequences of the extracted sequences were removed. We then used sickle tools to remove sequences with a quality value < 20, and discarded sequences < 150 bases in length and their paired sequences. The paired-end merge script FLASH [60] was used to merge the sequences that passed the quality filtering. The merging conditions were 420 bases for the merged fragment length, 280 bases for the read fragment length, and 10 bases for the minimum overlap length. The sequences that passed all the filtering were checked for chimeras using the uchime algorithm of usearch [61]. The reads that were not determined to be chimeras were used in subsequent analyses. In addition, if there were non-bacterial reads, they were removed. The creation of OTUs and phylogenetic estimation were performed using QIIME workflow scripts with no reference and all parameters set to default conditions. The number of reads used in the QIIME analysis was 224,878 (28,110 ± 1764). The genetic data were registered in DNA Data Bank of Japan (https://www.ddbj.nig.ac.jp/index.html: 191 accession number PRJDB12856).

Statistical analysis

Spearman’s rank correlation coefficient was used to examine the correlation between the hormonal concentrations and the behaviors of the cats, and different hormones. To investigate the similarity in gut microbes among the cats, a hierarchical cluster analysis was performed based on the percentage of similar microbes in the cats at the genus level. In addition, the UniFrac distance was calculated by constructing a phylogenetic tree using representative sequences from each OTU and UniFrac Principal Coordinate Analysis (PCoA) calculated. To assess the relationship between gut microbiome similarity and contact frequency among the cats, the correlation between weighted UniFrac distance and each behavior was examined using the Spearman’s rank correlation coefficient. The correlations between the components of each axis by PCoA of gut microbe OTUs and hormones were analyzed. In addition, the Mann-Whitney test was used to examine the sex-based differences in hormone concentrations, and Spearman’s rank correlation coefficient was used to examine the associations between hormone concentrations and age. JMP v14.2.0 (SAS Institute, Cary, NC, US) was used for all analyses.

Results

Correlation between hormone concentrations and behaviors of cats

Cortisol

We observed negative correlations between cortisol concentration and contact among cats, as well as the food sharing behavior (S3 Table). Significant negative correlations (p < 0.05) were specifically observed between cortisol concentration and active following (rs = ˗0.556, p = 0.032), playing (active: rs = ˗0.580, p = 0.023; passive: rs = ˗0.679, p = 0.005), sharing of food (rs = ˗0.681, p = 0.005), and passive sniffing (rs = ˗0.539, p = 0.038).

Testosterone

The results of correlation between testosterone concentration and cat behaviors are presented in S4 Table. Significant positive correlations (p < 0.05) were observed between testosterone concentration and active escape behavior (rs = 0.560, p = 0.030).

Oxytocin

We observed negative correlations between oxytocin concentration and contact among cats, as well as food sharing behavior, similar to in the results for cortisol (S5 Table). Significant negative correlations (p < 0.05) were observed between oxytocin concentration and active allo-grooming (rs = ˗0.678, p = 0.006), passive following (rs = ˗0.660, p = 0.007), passive playing (rs = ˗0.597, p = 0.019), sharing of food (rs = ˗0.597, p = 0.034), and sniffing (active: rs = ˗0.596, p = 0.019; passive: rs = ˗0.761, p = 0.001).

Furthermore, we analyzed correlations among concentrations of the different hormones (Fig 1). There was a significant positive correlation (p-value < 0.05) between concentrations of cortisol and testosterone (Fig 1A; rs = 0.636, p = 0.011). Incidentally, there was no effect of sex (S1 Fig; cortisol: Z = 0.412, p = 0.680; testosterone: Z = ˗0.177, p = 0.860; oxytocin: Z = 0.503, p = 0.596) or age (S2 Fig; cortisol: rs = 0.265, p = 0.361; testosterone: rs = ˗0.147, p = 0.615; oxytocin: rs = 0.087, p = 0.766) on any of the hormones.

Fig 1

Correlations among different hormones (cortisol, testosterone, and oxytocin).

Fig 1A shows the correlation between cortisol and testosterone, 1B shows the correlation between cortisol and oxytocin, and 1C shows the correlation between testosterone and oxytocin. Each group is color coded. Each point represents the average hormone concentration for each individual. The gray circle is a 95% confidence ellipse.

Gut microbes

Similarity of the proportion of gut microbes among the cats

First, to investigate the compositions of the gut microbiomes of individuals, the hierarchical clusters of the proportion of gut microbes observed at the genus level of the cats are shown in Supplemental Information (S3 Fig). Next, to investigate whether the compositions of the microbiomes are affected by contact with other individuals, we analyzed the correlations among UniFrac distances, which indicates similarity of gut microbiomes between individuals, and the interactions among the cats. Negative correlations were observed with respect to some of the behaviors, namely sharing a bed, entering bed, grooming, and sniffing (S6 Table). Hence, the greater the similarity in gut microbiome, the greater the sharing bed (Fig 2A, rs = ˗0.679, p < 0.001), entering bed (Fig 2B, rs = ˗0.395, p = 0.037), and sniffing (Fig 2C, rs = ˗0.446, p = 0.018).

Fig 2

Correlations among cat behaviors and UniFrac distances of their gut microbiomes.

Fig 2A shows the correlation between ‘sharing bed’ (s/h) and UniFrac distance, 2B shows the correlation between ‘entering bed’ (times/h) and UniFrac distance, and 2C shows the correlation between ‘sniffing’ (times/h) and UniFrac distance. The points indicate data in each pair, e.g., Cat 10 and Cat 11, Cat 10 and Cat 12. The gray circle is a 95% confidence ellipse.

Correlations of gut microbiome with hormone concentrations as well as behaviors of the cats

The correlations between each component of the axis based on PCoA of the microbes and hormone levels in the cats (S7 Table) as well as those between the PCoA and cat behavior (S8 Table) were analyzed. PCoA2 was correlated negatively with cortisol concentration (rs = ˗0.714, p = 0.047) and positively with rubbing (total: rs = 0.819, p = 0.013), following (total: rs = 0.747, p = 0.033), sharing of food (rs = 0.733, p = 0.039), and sniffing (rs = 0.738, p = 0.037) behaviors. Additionally, PCoA3 was correlated positively with entering bed (active: rs = 0.762, p = 0.028; passive: rs = 0.781, p = 0.022; total: rs = 0.762, p = 0.028), and was correlated negatively escaping behavior (rs = ˗0.903, p = 0.002). PCoA5 was correlated positively with cortisol concentration (rs = 0.762, p = 0.028), and negatively with following behavior (active: ˗0.866, p = 0.005). Moreover, PCoA8 was correlated with attacking behavior (rs = ˗0.714, p = 0.047).

Discussion

At the beginning of the study, it was hypothesized that cats with high concentrations of cortisol and testosterone would be less tolerant in their social interactions. The results of our study are consistent with our hypothesis. We observed that cats with high cortisol concentrations as well as those with high testosterone concentrations had less contact with other individuals; moreover, cats with high testosterone concentrations had a greater tendency to escape than cats with low testosterone concentrations. However, the correlation between oxytocin concentration and contact behaviors among cats was inconsistent with our hypothesis that high oxytocin concentrations would make them socially affiliative. The result is the first to reveal the correlation between cat–to–cat interaction and oxytocin concentration. Furthermore, cats with similar gut microbiomes exhibited high degrees of interaction with other cats during the study period, and correlations were also observed between gut microbiomes of the cats and their behavioral patterns, as well as cortisol concentrations.

Hormones and social behaviors

Previous studies have reported positive correlations between oxytocin concentration and affiliative contact with partners, such as allo-grooming, in several group-living species[32,33,62,63]. On the contrary, in this study, we observed a negative correlation between oxytocin concentration and affiliative contact among the cats living in the same space. Oxytocin functions in an affiliative manner, with respect to in-group individuals, but in an exclusionary manner, for out-group members [34,62], which implies that individuals with high oxytocin concentrations rarely exhibit affiliative behaviors with out-group members. Based on the results, even if cats spend time together and share the same space, they might not be able to form tightly connected groups since each cat might be consider the other cats out-group individuals. The experimental period of two weeks may be too short to enable the formation of a tight relationship group and observing them for a longer period could result in a change in their relationships. Although their relationships were possibly established since the cats who participated in the experiment had been living in the same room at the shelter before the experiment, investigations over a longer duration could offer greater insights on group-formation in cats.

Notably, the function of oxytocin varies between species. A previous study showed that eye contact in response to oxytocin is different between chimpanzees and bonobos―while in bonobos oxytocin increases eye contact, in chimpanzees oxytocin appears to have the opposite effect, and reduce eye contact [64]. The characteristics of species and the ecological environment surrounding them may cause different functions of oxytocin. It is possible that the characteristics of the cats in the experiment, such as sex and age, as well as the ecological environment in which they have been living, influence oxytocin function.

In the present study, we observed that cats who made contact with the other group members without any fear had a lower cortisol concentration. The results are consistent with those of studies on silver fox domestication, in which individuals were bred to show no fear or anxiety reactions towards humans [21,65]. Domesticated individuals with an advanced selection for being human-friendly exhibited higher tolerance toward human approach, and their GC secretion was inhibited. Therefore, the selection for fearlessness led to the inhibited development of the HPA axis, which is consistent with our results. Hence, the inhibition of the GC secretion system most probably contributed to the cohabitation as well as the high tolerance toward the group members in cats. In future, further studies may be performed to elucidate the biological mechanisms underlying such correlation.

Several previous studies have demonstrated that testosterone is associated with aggression in some species [22-25]. Although we did not find any association between testosterone and aggressive behavior in the cats, the individuals with high testosterone concentrations showed a high tendency to escape. The results also suggest that in spayed and castrated cats, testosterone is involved in developing tolerance towards other individuals, even though it does not lead to any aggressive behavior. Moreover, due to castration, the testosterone concentrations declined rapidly in the male cats [66]; hence, there was no difference between testosterone concentrations in the two sexes in the present study. Despite castration, testosterone secretions from the adrenal cortex may influence the temperaments of individual cats from this result.

In the present study, we investigated the relationship between basal hormone levels and cat behavior. Previously, studies have reported that several hormones are responsive in nature, especially cortisol [67]. Moreover, the psychological factors of an individual, such as implicit power motive and coping style [68-70], as well as the behavioral ecology [71] of the species can influence the hormonal variations with respect to a certain behavior. For instance, in the case of bonobos and chimpanzees, the behavioral ecology of the species leads to differences in endocrine changes with respect to competition [14]. Therefore, in future, the investigation of responsive hormones may offer a more comprehensive understanding of the influence of the behavioral ecology of cats on the relationship between behavior and basal hormone concentrations.

Gut microbiome

In the present study, the more interactions there were between individuals, the more similar their gut microbiome was. When animals are housed together and the environment is shared, there is increased similarity in their gut microbes owing to the increased chances of direct or indirect contact [72]. Previously, a study on baboons demonstrated that microbes may be transmitted through contact [44]. Incidentally, allo-grooming, which involves the exchange of biochemicals from external secretions, has a stronger influence on transmission of gut microbes than any other physical contact behavior. However, we did not observe any relationship between allo-grooming and similarity in the gut microbes in the cats. This could be due to the small sample size and limited time frame of the study. In addition, this study examined the gut microbiome of cats that stayed in the same room and ate the same food, but it did not completely eliminate the effects of environment, because the experiments were conducted at different times. Therefore, it needs to be re-examined using the time series gut microbiomes of a larger sample size living in the same room at the same period. A key question to be addressed is whether the sharing of microbiomes through physical contact produces a fitness benefit or cost to the host. In this respect, there is a positive aspect in which the microbiome can recolonize in individuals that have lost beneficial microbes due to disease or use of antibiotics, through microbiome transmission from other members. On the contrary, there is a negative aspect in which highly interactive individuals have high chances of being exposed to pathogenic microbes.

Gut microbes reportedly affect brain function, thereby causing phenotypic changes in individuals, such as behavioral alterations. Gut microbes are involved in the development of the HPA and HPG axes [40,42] and generation of oxytocin-expressing neurons [43]. One of the pathways via which the gut microbes influence brain functions is through endocrine activity, such as secretion of GCs, testosterone, and oxytocin. Since we observed a correlation between gut microbes and cortisol secretion, as well as between gut microbes and behavioral responses of the cats, it is highly likely that the gut microbes influence hormone secretions and behavior mechanisms of individuals. However, in the current study, we did not analyze the relationship between the microbial species and behavioral responses or hormone concentrations in detail. In future, an extensive research with a large sample size and a detailed analysis, together with empirical experiments using germ-free mice, may reveal which microbes specifically influence behavior and hormone concentrations. Most mammalian gut microbiome studies have focused on fecal microbiota, but it is unclear how well fecal samples reflect the microbiota of the intestinal region, and these should also be examined.

Limitation and future perspective

A follow-up study observing the cats over a period of several months might provide more comprehensive information. In addition, our study was limited to correlations between hormones and behaviors, so that causality is not known. Since the subject cats were of different or unknown ages and backgrounds, it may have been impossible for them to live as ’group mates’. Hence, in future studies, it is necessary to examine the factors that can affect the ability of the cats to consider each other as ‘in-group’ members and form affiliative relationships among them; for instance, the experience of spending time together from their juvenile period can facilitate tight group formation among the cats [73]. Moreover, the sex composition of the members of groups was not consistent. Experiments with male-only and female-only groups could facilitate the elucidation of the adaptive significance of gender in cats living in groups.

Hormones regulate social behavior by binding to their specific receptors in the brain. Hormones can affect social behavior in different ways, depending on the number of receptors as well as the region of the brain where the receptors are present. For example, the expression patterns of oxytocin receptors differ greatly between prairie voles (Microtus ochrogaster) and montane voles (Microtus montanus), which has been suggested to be one the factors influencing their monogamous/polygamous nature [74]. Therefore, future studies should focus on the expression patterns of hormone receptors in domesticated cats to uncover the sociality of the cats living in groups.

In summary, we observed that cortisol, testosterone, and oxytocin concentrations are correlated with the social behavior of the cats, which are considered solitary animals, and the gut microbial composition is related to social behavior as well as endocrine concentrations. Changes in the endocrine system may lead to temperament changes, which, in turn, enable the cats to share space with other cats. Furthermore, this study sheds new light on the role of oxytocin in solitary animals, and further research is required to understand the mechanisms underlying such a relationship. Since behavior is greatly influenced by external environmental factors, such as temperature and food resources, examining changes in social relationships among cats that occur as a result of variations in the external environment could provide further insights into the factors that influence sociality among the individuals.

Sex differences in cortisol, testosterone, and oxytocin concentrations.

S1A Fig indicates cortisol, S1B indicates testosterone, and S1C indicates oxytocin. Each point shows average hormone concentration for each individual.

(TIF)

Click here for additional data file.

Correlation between age and cortisol, testosterone, and oxytocin concentrations.

S2A Fig indicates cortisol, S2B indicates testosterone, and S2C indicates oxytocin. Each point shows average hormone concentration for each individual. The gray circle is a 95% confidence ellipse.

(TIF)

Click here for additional data file.

Similarity in the proportion of gut microbes.

The hierarchical clusters are shown based on the percentages of gut microbes at the genus level. The colors of each cat are color-coded for each group (Orange: Group 1, Blue: Group 2, Green: Group 3).

(TIF)

Click here for additional data file.

Profile of cats.

This table shows each group of cats, their sex, and age. The colors are divided by groups, with the first group shown in orange, the second group in blue, and the third group in green.

(CSV)

Click here for additional data file.

Analyzed behaviors and their definitions.

(CSV)

Click here for additional data file.

Correlation between cortisol concentrations and behaviors of cats.

The results of Spearman’s rank correlation coefficient between cortisol concentrations and cat behaviors are listed. The letters in brackets following each behavior are: a for active, p for passive, and t for the sum of passive and active behavior. Those with p-values < 0.05 have been marked in bold. The data are color-coded according to the value of the correlation coefficient (Blue: ˗0.8 to ˗0.6, Light blue: ˗0.6 to ˗0.4, Light orange: 0.4 to 0.6, Orange: 0.6 to 0.8). The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Correlation between testosterone concentrations and behaviors of cats.

The results of Spearman’s rank correlation coefficient between cortisol concentrations and cat behaviors are listed. The letters in brackets following each behavior are: a for active, p for passive, and t for the sum of passive and active behavior. Those with p-values < 0.05 have been marked in bold. The data are color-coded according to the value of the correlation coefficient (Blue: ˗0.8 to ˗0.6, Light blue: ˗0.6 to ˗0.4, Light orange: 0.4 to 0.6, Orange: 0.6 to 0.8). The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Correlation between oxytocin concentrations and behaviors of cats.

The results of Spearman’s rank correlation coefficient between oxytocin concentrations and the cat behaviors are listed. The letters in brackets following each behavior are: a for active, p for passive, and t for the sum of passive and active behavior. Those with p-values < 0.05 are shown in bold. The data are color-coded according to the value of the correlation coefficient (Blue: 521 ˗0.8 to ˗0.6, Light blue: ˗0.6 to ˗0.4, Light orange: 0.4 to 0.6, Orange: 0.6 to 0.8). The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Correlations of UniFrac distances of gut microbiome with interactions among the individuals.

Those with p-values < 0.05 are shown in bold. For sharing food and playing, statistical processing was not possible due to the small number of individuals that exhibited the behaviors. The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Correlations of the components of each component of the axis based on principal coordinate analysis (PCoA) axis with cortisol, oxytocin, and testosterone concentrations.

Those with p-values < 0.05 are shown in bold. The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Correlations of the components of each component of the axis based on principal coordinate analysis (PCoA) axis with the behaviors of the cats.

The letters in brackets following each behavior are: a for active, p for passive, and t for the sum of passive and active behavior. Those with p-values < 0.05 are shown in bold. The rs are correlation coefficients.

(CSV)

Click here for additional data file.

Dataset_Behaviors and hormones.

(CSV)

Click here for additional data file.

Dataset_List of operational taxonomic units (OTUs), which are classification units in clustering based on 16SrRNA gene sequences.

(CSV)

Click here for additional data file.

Dataset_Weighted unifrac distance and behaviors.

(CSV)

Click here for additional data file.

Dataset_Weighted unifrac pcoa and hormones and behaviors.

(CSV)

Click here for additional data file.

26 Dec 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Chun Wie Chong

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Acknowledgments Section of your manuscript:

“This work was supported by the Japan Society for the Promotion of Science and Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (Grant numbers: #20J14760 to H.K., 23 #18H02489 and #19K22823 to M.N., and #19H00972 to T.K.).”

We note that you have provided additional information within the Acknowledgements Section that is not currently declared in your Funding Statement. Please note that funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“This work was supported by the Japan Society for the Promotion of Science (https://www.jsps.go.jp/) and Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (https://www.mext.go.jp/) The grant numbers are #20J14760 to H.K., #18H02489 and #19K22823 to M.N., and #19H00972 to T.K.

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The manuscript addresses an interesting relationship between behaviour, hormones and gut microbiota of group-living cats. However, the methodology and statistics used in this study is not well-explained and lack sufficient detail to evaluate the relevance of the findings.

First, the analysis of gut microbiota section must have a detailed information about the laboratory and bioinformatic processing. Even if the processing was done by a company all the information about the processing steps need to be addressed. There is no information about extraction method, PCR settings, library building, primers, sequencing strategy, if biological or technical replicates were used, how many blanks and what kind were used… Moreover, the faecal samples were collected using two different method? Was this controlled in the analysis?

First of all, the gut microbiota analysis section should have detailed information about the laboratory and bioinformatics processing. Even if the processing was done by a company, all information about the processing steps must be addressed in the methodology. There is no information on the extraction method, PCR setup, library construction, primers, sequencing strategy, if biological or technical replicas were used, how many blanks and what type were used... Moreover, faecal samples were collected using two different methods. Was this controlled in the analysis?

Second, it is not sufficient to write that the “data were processed using QIMME”, the author should include detailed information about how the sequences were analysed (e.g. demultiplexing, primer removal, filtering, denoising, contaminant removal…). In addition, the authors should make the bioinformatic pipeline available.

Third, it is not clear to me how the faecal samples were collected. In the experimental setting section, the authors say that 8 faecal samples were collected, but from which individuals? Was it possible to identify who owned each stool sample? Otherwise, how was it possible to correlate a specific microbial community with a specific behaviour? And finally, have you considered in the analysis that some faecal samples belong to cats that share a room (5 rooms and 8 faecal samples)? Therefore, some samples are not independent. How was this controlled in the statistical analysis?

Fourth, since the relevant correlations (and p-values) are already included in the text and the tables do not provide any additional information, all the tables should be moved to supplementary material.

Lastly, the discussion part should be re-organized to clarify the findings of this study and its consequences.

Reviewer #2: This study provides insight between behavior of shelter cats correlated to hormonal and gut microbiome profiling. The conclusions obtained were valuable as the findings shed light on the hormonal status of originally solitary or group-living animals. It is hopeful that the conclusions drawn from this study can be an example to help behaviorists incorporate more measurable parameters which will reduce subjectivity when relying only on visual observations. Despite the gender-neutral assumption, it might be worthwhile to have gender-specific groupings to further understand its impact on behavior for future studies, in addition to increased sample size and longer duration. The methods used were sound and results obtained provided clarity on the hypotheses.

Reviewer #3: 1. Interesting. The authors addressed notable correlation between hormones and behaviours. While mounting studies had established the correlation between gut microbiome and behaviours, the present study lacking clarity in term of study design for determining the gut microbiota constituents. Microbiota composition varies inter-individually, and effected by various environmental factors. The recent study only sampled the cats stool at one point of time. It is suggested to sample the stool similarly like urine, and sequence the stool concurrently i.e. time series analysis for the gut microbiome.

2. The grouping strategy was vaguely described.

3. While the hormonal analyses were conducted for 50 plus samples, the microbiome analysis was conducted only on 8 cats subjects, with Group 1 only represented by 1 subjects.

4. Sample preparation (microbial genomics library preparation) was vaguely described.

5. Microbial genomics analysis was vaguely described.

6. The raw sequencing data were not publicly available. Data availability. We recommend that all data and related meta-data underlying the findings reported in a submitted manuscript will be deposited in an appropriate public repository (GenBank, European Nucleotide Archives, National Center for Biotechnology and Information, National Bioscience Data Center, DNA data bank of Japan).

7. Is there any specific reason to opt for UniFrac distance?

8. The manuscript was written in standard English, with minor grammatical errors.

9. Data presentation using PCoA were OK, but making individual comparison for microbiome analysis was vary confusing. I would suggest for the authors to conduct group comparison for the microbiome analysis. In fact, inter group correlation between behaviours also can be conducted.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

23 Jan 2022

24.01.2022

Chun Wie Chong

Academic Editor

PLOS ONE

Dear Dr. Chun Wie Chong

Thank you for inviting us to submit a revised draft of our manuscript (PONE-D-21-34568) titled “Correlation between behavior and hormone or gut microbiome implies that domestic cats (Felis silvestris catus) living in a group are not like ‘groupmates’”. We appreciate the time and effort you and each of the reviewers have dedicated to providing insightful feedback on how to strengthen our paper. Thus, it is with great pleasure that we resubmit our article for further consideration. We have incorporated changes that reflect the detailed suggestions you have graciously provided. We hope that our edits and the responses we provide below satisfactorily address all issues and concerns you and the reviewers have noted. Please find our point-by-point responses to the reviewers' comments below.

I am looking forward to hearing from you.

Sincerely,

Hikari Koyasu

Human-Animal Interaction and Reciprocity Laboratory,

Azabu University

Reviewer #1:

The manuscript addresses an interesting relationship between behaviour, hormones and gut microbiota of group-living cats. However, the methodology and statistics used in this study is not well-explained and lack sufficient detail to evaluate the relevance of the findings.

Reply: We thank the reviewer for all the detailed comments and suggestions. We found them very useful as we approached our revision.

First, the analysis of gut microbiota section must have a detailed information about the laboratory and bioinformatic processing. Even if the processing was done by a company all the information about the processing steps need to be addressed. There is no information about extraction method, PCR settings, library building, primers, sequencing strategy, if biological or technical replicates were used, how many blanks and what kind were used… Moreover, the faecal samples were collected using two different method? Was this controlled in the analysis?

Reply: We have added detailed information about the bioinformatic processing of gut microbiomes (lines 146-175).

In addition, as mentioned by the reviewer, the fecal samples were collected using two methods, and it was best to collect them in one consistent way. The sampling method for the first group was for 16S rRNA analysis, in addition to the storage of bio-active microbiomes (glycerol stock). However, the application of this method would have involved high efforts and exceeded our storage space, so that we changed the collection method. The time of the sampling did not change the composition of the microbiome as long as we collected the part of the feces that was not exposed to the air near the center (Sinha et al. 2016). Moreover, the use of the swab did not change the composition of the microbiome (Sinha et al. 2016). Therefore, targeting a small number of bacteria in the fecal microbiome that are sensitive to oxygen may not eliminate the effect of differences in sampling methods, but by comparison of the overall structure, we determined that there were small effects on the analysis.

Second, it is not sufficient to write that the “data were processed using QIMME”, the author should include detailed information about how the sequences were analysed (e.g. demultiplexing, primer removal, filtering, denoising, contaminant removal…). In addition, the authors should make the bioinformatic pipeline available.

Reply: We have added bioinformatic information and information about the analysis of the gut microbiome（lines 176-191).

Third, it is not clear to me how the faecal samples were collected. In the experimental setting section, the authors say that 8 faecal samples were collected, but from which individuals? Was it possible to identify who owned each stool sample? Otherwise, how was it possible to correlate a specific microbial community with a specific behaviour? And finally, have you considered in the analysis that some faecal samples belong to cats that share a room (5 rooms and 8 faecal samples)? Therefore, some samples are not independent. How was this controlled in the statistical analysis?

Reply: We collected the fecal samples from different individuals only when we observed which cats excreted the feces. Moreover, the experiment was conducted for one group at a time, and the three groups stayed in the same room. In addition, all cats were eating the same food. We have reduced the influence of the similarity of the gut microbiome through the room environment and food as much as possible. We have rewritten the method description on the fecal collection and experimental setting (lines 146, 96-98), and individuals for which fecal samples were collected are shown in Table S1 to address the reviewer’s comment. We hope that the revised section is now clarified.

However, it was not possible to completely eliminate the effects of the environment because each group was in the specific environment at different times. Since we were only able to collect samples from one cat in the first group and two cats in the second group, it was difficult to take into account that there were different groups in this study. Therefore, we have added to the discussion that it is necessary to examine the gut microbiome of more cats living in the same room at the same time (lines 325-329).

Fourth, since the relevant correlations (and p-values) are already included in the text and the tables do not provide any additional information, all the tables should be moved to supplementary material.

Reply: We thank the reviewer for the suggestion. All tables have been moved to the supplementary material.

Lastly, the discussion part should be re-organized to clarify the findings of this study and its consequences.

Reply: We have rewritten the discussion to clarify the findings and their consequences.

Reviewer #2:

This study provides insight between behavior of shelter cats correlated to hormonal and gut microbiome profiling. The conclusions obtained were valuable as the findings shed light on the hormonal status of originally solitary or group-living animals. It is hopeful that the conclusions drawn from this study can be an example to help behaviorists incorporate more measurable parameters which will reduce subjectivity when relying only on visual observations. Despite the gender-neutral assumption, it might be worthwhile to have gender-specific groupings to further understand its impact on behavior for future studies, in addition to increased sample size and longer duration. The methods used were sound and results obtained provided clarity on the hypotheses.

Reply: We thank the reviewer for the useful comments. We agree on the importance of conducting further research with gender-specific groupings. The differences in frequency of behaviors that occur within males and females, and between males and females should be investigated by addition of further cats. We have added this information to the discussion (lines 351-352).

Reviewer #3:

1. Interesting. The authors addressed notable correlation between hormones and behaviours. While mounting studies had established the correlation between gut microbiome and behaviours, the present study lacking clarity in term of study design for determining the gut microbiota constituents. Microbiota composition varies inter-individually, and effected by various environmental factors. The recent study only sampled the cats stool at one point of time. It is suggested to sample the stool similarly like urine, and sequence the stool concurrently i.e. time series analysis for the gut microbiome.

We thank the reviewer for raising this important point. For this gut microbiome analysis, we used feces collected as close to the end of the two-week experimental period as possible. Although only suggestive, this shows an association between the frequency of interactions among individuals during two weeks and the similarity of the gut microbiome as close to the end of that period as possible.

As pointed out by the reviewer, microbiome composition varies inter-individually and is affected by various environmental factors, including food, the presence of other individuals, interaction with others, and so on. Sampling at various time points and examination of the relationship between changes of gut microbiomes in the time series and the frequency of contact with other individuals would be required. In addition, the possibility of transmission of bacteria through a shared environment has not been completely eliminated, and this is an issue that needs to be investigated as a future task. We have added these concerns to the discussion (lines 325-329).

2. The grouping strategy was vaguely described.

Reply: As for the grouping, it was completely random. Since the number of males and females was not consistent, we have added to the discussion that it will be necessary to conduct experiments in groups with a consistent number of males and females, as well as in groups with only males or only females to understand the adaptive significance of group formation (lines 351-352).

3. While the hormonal analyses were conducted for 50 plus samples, the microbiome analysis was conducted only on 8 cats subjects, with Group 1 only represented by 1 subjects.

Reply: This was a matter of concern to us. We could not collect many feces from the cats, but this was due to the fact that we only collected feces when we could see who had defecated at the time of defecation, and the cats often defecated when humans were not around. We have added that future analyses of time series with larger sample sizes are needed (lines 325-329).

4. Sample preparation (microbial genomics library preparation) was vaguely described.

Reply: We thank the reviewer for pointing out this point. We have added detailed information about the sample preparation, such as microbial genomics library preparation (lines 156-175).

5. Microbial genomics analysis was vaguely described.

Reply: We have added detailed information about the microbial genomics analysis (lines 176-191).

6. The raw sequencing data were not publicly available. Data availability. We recommend that all data and related meta-data underlying the findings reported in a submitted manuscript will be deposited in an appropriate public repository (GenBank, European Nucleotide Archives, National Center for Biotechnology and Information, National Bioscience Data Center, DNA data bank of Japan).

Reply: We thank the reviewer for the suggestion. The data has been deposited in the DNA Data Bank of Japan (https://www.ddbj.nig.ac.jp/index.html: accession number PRJDB12856).

7. Is there any specific reason to opt for UniFrac distance?

Reply: We opted for the UniFrac distance because it enables us to compare the differences in the overall microbiome between two samples while considering phylogeny. The UniFrac distance takes into account the distance between bacterial phylogeny, which enables principal coordinate analysis.

8. The manuscript was written in standard English, with minor grammatical errors.

Reply: Our manuscript was again corrected by native speakers.

9. Data presentation using PCoA were OK, but making individual comparison for microbiome analysis was vary confusing. I would suggest for the authors to conduct group comparison for the microbiome analysis. In fact, inter group correlation between behaviours also can be conducted

Reply: We thank the reviewer for the suggestion. We agree on the relevance of conducting an intergroup comparison of the gut microbiome, but this was not possible due to the insufficient number of cases (only one sample in the first group and two samples in the second group). When we show that different groups have different gut microbiome compositions , we cannot exclude the possibility of similarity from the shared environment. Therefore, we examined the relationship between the frequency of contact between individuals and the degree of similarity of their gut microbiome. To avoid any confusion, we have reorganized the results section (lines 243-250).

Submitted filename: Response to reviewers.docx

Click here for additional data file.

2 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Chun Wie Chong

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

Reviewer #4: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #3: No

Reviewer #4: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #3: No

Reviewer #4: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

Reviewer #4: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #3: Yes

Reviewer #4: No

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I believe that the ms has improved significantly, however, I still have some doubts about the reliability of the results coming from the analysis of microbial data. The authors do not mention the use of controls and, in addition, they did not perform any filtering steps to minimise the introduction of non-real reads. Therefore, the reliability of these results is highly compromised. With regard to statistical analysis, more than one urine sample was collected from an individual to assess hormone levels, so the samples are not independent. How did you statistically analyse this dependency between the samples? Additionally, I think the authors should also modify the discussion so that the correlation between hormone levels and behaviours is easier to understand. For example: why is oxytocin in one subsection and cortisol and testosterone in another subsection? Also, at the end of the cortisol and testosterone section you mention oxytocin. So it is necessary to clarify the subsections or if they prefer to merge them all into one. It may be easier for the reader to understand the correlation of hormones and behaviour if the authors explain them all together. Since the discussion is mainly based on the correlation of hormones and affiliative contact with peers (tolerance towards other individuals), I would recommend discussing the results according to behaviour. What are the hormone levels related to affiliative behaviour? Otherwise it is very repetitive.

L28-29 Is it only the environment that shapes it? Genes have no effect? I would say that environment is one of the variables that affect behaviour. This sentence should be modified and more references added.

L33: What do you mean domesticated cats? Animals that live with humans as pets? or feral cats that live in cities. Please clarify it. I don't think cats LIVE in high densities, but rather in large groups. We can find high densities of cats in cities and towns, but not all of these cats live together as a group.

L46-47: Add References

L52. Delete this sentence, you already mentioned it in the previous sentence.

L79: did not measure the AMOUNT of specific intestinal microbiota, but rather the composition

L81-82: It is not clear to me what “solitary mammal within the same species” refers to.

L85: Indicate the name of the shelter and the locality

L86: Have these animals received any health treatment, such as antibiotics?

L113: You already mentioned how many samples you collected in L96. You only need to write it once.

L123-124: Explain why the concentrations for creatinine correction were adjusted and how it was done.

L117-125-134: There are no references on the measurement of cortisol, testosterone… Explain how these protocols were established and add references if based on previously published protocols.

L165: Add reference for primers.

L178 - Fastx tool kit needs a reference

L179: Explain the function and the package used to perform quality control. Add the corresponding references

L181:Add references for FLASH and uchimeras

L184-185: I guess you removed the chimeric sequences and then grouped them together and finally added the taxonomy. But this is not clear to me in the text, you should explain each step well and add the function used and its references.

L188-190: Just mention it in the statistical analysis section

L280: Add examples considered as affiliative behaviours in your study

L287-289: It is not clear what additional information you provide from the previous sentence

L300: Are cortisol and testosterone levels correlated? Are both related to fearlessness?

L303: Add a reference

L304-307: It is not clear which results are from this study and which are from other studies. Please clarify this and add references when referring to other studies.

Reviewer #3: Although the author has made corrections according to the recommendations given, I still see weaknesses in the study design for this study. With an unbalanced sample size for each group of cats, it is too difficult to draw conclusions. For example, the composition of the bacteria identified in this study is not discussed in depth. So it is too far -fetched for the authors to draw comparative conclusions involving the gut microbiome. Finally, I can't see the crosstalk between hormones and the composition of the cat's gut microbiota. This makes this study not strong.

Reviewer #4: Dear authors,

I appreciate the edits already completed by the authors from reviewer 1-3. However, this publication is still not at an acceptable standard for publication. The premise of the study is excellent, and it is a very interesting paper. However, authors must complete major edits, as I have begun to outline below:

The use of colloquial English ('we', "attempted to", "close to significant" etc.) must be corrected.

Abbreviations need only be explained once when first used. References must be added to statements on existing literature in the introduction, and species specific references included as opposed to blanket statements. The analytical methods used to assess hormone concentrations must be referenced. Standard units must be used (e.g. µl not ul), kits and reagents must have the manufacturing company and country of origin specified in brackets. Supplementary tables provide little extra information, where are the notes indicating what rs denotes? Why is there a ? in front of these rs scores? Consistency must be addressed e.g. "p-value" and "p". Figure's require legends which thoroughly describe the research presented. In an example, Figure 1A-C, 62 samples were analysed for hormone concentrations, yet there are 14 data points? What do the grey circle denote? Also authors must note that they are analysing the faecal microbiome, not the gastrointestinal microbiome.

Additionally, here are some specific edits I began writing:

Line 46: Please provide references for this statement

Line 59-61: Usually this is just called the gut-brain axis (GBA). References must be included at the end of this sentence.

Line 62-65: Please state which model (human, rodent, in vitro) these studies were conducted in, as there is currently no studies which have assessed this in domestic cats.

Line 67-72: Again, please state which species you are referring too. These statements form a more compelling argument for the group-living of domestic cats, if they are referring to research conducted in prides of lions as opposed to a school of dolphins.

Line 85: Please add a statement declaring where the cats were housed during the study. E.g. Adult domestic shorthair cats were housed at the XX facility/shelter, Japan for the duration of the study.

Line 89: two weeks

Line 90: litter tray?

Line 97: As diet is such an important factor for the microbiome, it would be great to have the dietary information here. Even if only to add in a sentence such as: “Cats were fed a complete and balanced commercially available kibble/can diet”

Line 113: For future studies, collect urine from individuals and note which cat produced it. This is a far more accurate way of collecting data, rather than expressing the sample as a number per individuals.

Cortisol/Creatinine/Testosterone/Oxytocin Concentrations: Were these methods derived from existing published methods? If so, please insert references.

Best wishes

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

Reviewer #4: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

17 May 2022

Reviewer #1:

I believe that the ms has improved significantly, however, I still have some doubts about the reliability of the results coming from the analysis of microbial data. The authors do not mention the use of controls and, in addition, they did not perform any filtering steps to minimise the introduction of non-real reads. Therefore, the reliability of these results is highly compromised.

Reply: We appreciate the reviewer for all the detailed comments and suggestions. We found them very useful as we undertook our revision.

Instead of using the PhiX control, we have shifted the reading frame by adding 0-5 random N in front of the primer sequence. As for the non-real reads that the referee pointed out, it seems to be due to the following three issues: low quality reads, chimeras, and non-bacterial reads. We removed sequences with a quality value < 20, and all sequences were checked for chimeras, and those not determined to be chimeric were extracted and used for subsequent analyses. The information has been provided in the manuscript. For the processing of non-bacteria reads, non-bacteria reads are removed after annotation. This has been added to the methods: In addition, if there were non-bacterial reads, they were removed (lines 197-198).

With regard to statistical analysis, more than one urine sample was collected from an individual to assess hormone levels, so the samples are not independent. How did you statistically analyse this dependency between the samples?

Reply: Thank you for pointing this out. When we checked the factors affecting all urinary hormones using a generalized linear mixed model, we found that these hormone samples were not affected by date and that individual effects were significant. Therefore, the average of the urine hormone concentrations for each individual was used as the hormone baseline for each individual. We have added an additional explanation for the urinary hormone data: The average of the urine hormone concentrations for each individual was used as the hormone baseline for each individual (lines 121-122).

Additionally, I think the authors should also modify the discussion so that the correlation between hormone levels and behaviours is easier to understand. For example: why is oxytocin in one subsection and cortisol and testosterone in another subsection? Also, at the end of the cortisol and testosterone section you mention oxytocin. So it is necessary to clarify the subsections or if they prefer to merge them all into one. It may be easier for the reader to understand the correlation of hormones and behaviour if the authors explain them all together. Since the discussion is mainly based on the correlation of hormones and affiliative contact with peers (tolerance towards other individuals), I would recommend discussing the results according to behaviour. What are the hormone levels related to affiliative behaviour? Otherwise it is very repetitive.

Reply: We appreciate the reviewer's concerns on this point. As the reviewer pointed out, the topic of oxytocin in the cortisol/testosterone section could cause confusion; therefore, the section has been reorganized. Since the study is based on the hypothesis that hormones modify cat behaviors, discussions have been made based on hormones, and we would like to maintain this approach.

L28-29 Is it only the environment that shapes it? Genes have no effect? I would say that environment is one of the variables that affect behaviour. This sentence should be modified and more references added.

Reply: As the reviewer points out, the environment is one of the factors that affect behavior. We have corrected the relevant sentence: During evolution, ecological factors, including food availability, influence the development of group and solitary living behavior in animals [1] (lines 27-28).

L33: What do you mean domesticated cats? Animals that live with humans as pets? or feral cats that live in cities. Please clarify it. I don't think cats LIVE in high densities, but rather in large groups. We can find high densities of cats in cities and towns, but not all of these cats live together as a group.

Reply: Domesticated cat in this context refers to the domesticated cat as a species, which includes both cats living in homes with humans and feral cats. While many species of felids live alone and have exclusive territories, domestic cat species are able to live at relatively high densities in specific spaces. The definition of a group varies. Therefore, in the present context, we do not use the word 'group' but 'high density' because it is unclear what kind of group the cats form.

L46-47: Add References

Reply: We have added references.

e.g.,

M Stöwe; T Bugnyar; C Schloegl; B Heinrich; K Kotrschal; E Möstl. Corticosterone excretion patterns and affiliative behavior over development in ravens (Corvus corax). Horm Behav 53, 208–216 (2008)

FB de Waal; F Aureli; PG Judge. Coping with crowding. Sci Am 282, 76–81 (2000)

DA Gust; TP Gordon; MK Hambright; ME Wilson. Relationship between social factors and pituitary-adrenocortical activity in female rhesus monkeys (Macaca mulatta). Horm Behav 27, 318–331 (1993)

MN Barbosa; MT da S Mota. Behavioral and hormonal response of common marmosets, Callithrix jacchus, to two environmental conditions. Primates 50, 253–260 (2009)

C Crockford; RM Wittig; K Langergraber; TE Ziegler; K Zuberbuhler; T Deschner. Urinary oxytocin and social bonding in related and unrelated wild chimpanzees. Proc Biol Sci 280, 20122765 (2013)

L52. Delete this sentence, you already mentioned it in the previous sentence.

Reply: Thank you for the suggestion. The previous sentence was about various species, and the relevant sentence described cats. We have added an explanation that these describe a variety of species or cats.

L79: did not measure the AMOUNT of specific intestinal microbiota, but rather the composition

Reply: The reviewer’s comment is correct. We have corrected the sentence: 3. there is a relationship between the composition of gut microbiomes and hormone concentrations in an individual’s body (lines 80-81)

L81-82: It is not clear to me what “solitary mammal within the same species” refers to.

Reply: ‘Solitary mammal within the same species’ means that ‘role of oxytocin in conspecific social behaviors in a solitary mammal.’ To clarify, we have edited the sentence (lines 83).

L85: Indicate the name of the shelter and the locality

Reply: We have added the name of the shelter and the locality (line 87).

L86: Have these animals received any health treatment, such as antibiotics?

Reply: The cats in the experiment were not treated with antibiotics or other treatments.

L113: You already mentioned how many samples you collected in L96. You only need to write it once.

Reply: Thank you for pointing this out. We deleted the sentence with the same information in the experimental setting.

L123-124: Explain why the concentrations for creatinine correction were adjusted and how it was done.

Reply: All urinary hormone concentrations vary greatly, depending on the concentration of urine. Since the daily excretion of creatinine is constant (Shaffer PA, 1908), the hormone concentration considering the concentration of urine can be calculated based on the amount of creatinine in that urine (Munro CJ et al., 1991).

Shaffer PA., The excretion of creatin and creatinine in health and disease., American Journal of Physiology, 1908;10:1-10.

Munro CJ, Stabenfeldt GH, Cragun JR, Addiego LA, Overstreet JW, Lasley BL., Relationship of serum estradiol and progesterone concentrations to the excretion profiles of their major urinary metabolites as measured by enzyme immunoassay and radioimmunoassay., Clinical Chemistry, 1991;37:838-844.

L117-125-134: There are no references on the measurement of cortisol, testosterone… Explain how these protocols were established and add references if based on previously published protocols.

Reply: These protocols for hormone assays were based on previous studies. Related references have been added.

L165: Add reference for primers.

Reply: We have added an appropriate reference for the primers.

Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., & Glöckner, F. O. (2013). Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic acids research, 41(1), e1-e1.

L178 - Fastx tool kit needs a reference

Reply: We have added an appropriate reference.

A Gordon; GJ Hannon; Others. Fastx-toolkit. FASTQ/A short-reads preprocessing tools (unpublished) http://hannonlab cshl edu/fastx_toolkit 5 (2010)

L179: Explain the function and the package used to perform quality control. Add the corresponding references.

Reply: Fragment Analyzer and dsDNA 915 Reagent Kit (Advanced Analytical Technologies, Inc.) were used to check the quality of the libraries. We have added a relevant sentence and reference: The quality of libraries was evaluated using a Fragment Analyzer and a dsDNA 915 Reagent Kit (Advanced Analytical Technologies, Inc., USA).

N Susai; T Kuroita; K Kuronuma; T Yoshioka. Analysis of the gut microbiome to validate a mouse model of pellagra. Bioscience of Microbiota, Food and Health advpub, 2021–2059 (2022)

L181:Add references for FLASH and uchimeras

Reply: We have added the relevant references.

FLASH: T Magoč; SL Salzberg. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics 27, 2957–2963 (2011)

UCHIME: RC Edgar; BJ Haas; JC Clemente; C Quince; R Knight. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200 (2011)

L184-185: I guess you removed the chimeric sequences and then grouped them together and finally added the taxonomy. But this is not clear to me in the text, you should explain each step well and add the function used and its references.

Reply: We have changed the sentences: The sequences that passed all the filtering were checked for chimeras using the uchime algorithm of usearch[61]. The reads that were not determined to be chimeras were used in subsequent analyses. In addition, if there were non-bacterial reads, they were removed. The creation of OTUs and phylogenetic estimation were performed using QIIME workflow scripts with no reference and all parameters set to default conditions. The number of reads used in the QIIME analysis was 224,878 (28,110 ± 1764) (lines 195-200).

L188-190: Just mention it in the statistical analysis section

Reply: We have removed the sentence and moved it to the statistical analysis section.

L280: Add examples considered as affiliative behaviours in your study

Reply: We considered allo-grooming, entering bed and sharing bed as affiliative behaviors in the present study. Since the affiliative behavior in the previous studies is mainly allo-grooming, we have described it.

L287-289: It is not clear what additional information you provide from the previous sentence

Reply: We have revised this part for enhanced clarity: The experimental period of two weeks may be too short to enable the formation of a tight relationship group and observing them for a longer period could result in a change in their relationships. Although their relationships were possibly established since the cats who participated in the experiment had been living in the same room at the shelter before the experiment, investigations over a longer duration could offer greater insights on group-formation in cats (lines 290-294).

L300: Are cortisol and testosterone levels correlated? Are both related to fearlessness?

Reply: Yes, there was a positive correlation between cortisol and testosterone concentrations. Yes, both are related to fearlessness.

L303: Add a reference

Reply: This is a suggestion based on the results of the present study and there is no previous study reporting such results. A comparison should be made to examine whether testosterone is associated with any of the behaviors in castrated and uncastrated cats. To clarify, we have clearly stated that this sentence is derived from the results of the present study.

L304-307: It is not clear which results are from this study and which are from other studies. Please clarify this and add references when referring to other studies.

Reply: Same as in the previous point, we have made them clearer.

Reviewer #3:

Although the author has made corrections according to the recommendations given, I still see weaknesses in the study design for this study. With an unbalanced sample size for each group of cats, it is too difficult to draw conclusions. For example, the composition of the bacteria identified in this study is not discussed in depth. So it is too far -fetched for the authors to draw comparative conclusions involving the gut microbiome. Finally, I can't see the crosstalk between hormones and the composition of the cat's gut microbiota. This makes this study not strong.

Reply: We appreciate you pointing this out. The fecal sample size in this study is a a major source of concern. Although it would have been ideal to collect feces from all individuals, limited feces were collected from all individuals within the two weeks. Some of the cats were afraid of people and would not defecate when people were present, because they were from a shelter. Therefore, data are limited due to the small sample size, and the results of the present study only suggest that there is a relationship between gut microbiome and behaviors and hormones. Nevertheless, the preliminary results could guide and facilitate future studies on cat microbiomes and the correlation with behavior.

Reviewer #4:

Dear authors,

I appreciate the edits already completed by the authors from reviewer 1-3. However, this publication is still not at an acceptable standard for publication. The premise of the study is excellent, and it is a very interesting paper. However, authors must complete major edits, as I have begun to outline below:

Reply: Thank you very much for providing such critical comments. In the following sections, you will find our responses to each of your points and suggestions. We are grateful for the time and energy you expended in reviewing this manuscript.

The use of colloquial English ('we', "attempted to", "close to significant" etc.) must be corrected. Abbreviations need only be explained once when first used.

Reply: We have revised the entire manuscript accordingly, with checks by native speakers.

References must be added to statements on existing literature in the introduction, and species specific references included as opposed to blanket statements.

Reply: We have added species-specific references (cats/other felids) and noted the associated species.

M Franchini; A Prandi; S Filacorda; EN Pezzin; Y Fanin; A Comin. Cortisol in hair: a comparison between wild and feral cats in the north-eastern Alps. Eur J Wildl Res 65, 90 (2019)

H Finkler; J Terkel. Cortisol levels and aggression in neutered and intact free-roaming female cats living in urban social groups. Physiol Behav 99, 343–347 (2010)

G Genaro; CR Franci. Cortisol influence on testicular testosterone secretion in domestic cat: An in vitro study. Pesqui Vet Bras 30, 887–890 (2010)

The analytical methods used to assess hormone concentrations must be referenced. Standard units must be used (e.g. µl not ul), kits and reagents must have the manufacturing company and country of origin specified in brackets.

Reply: We have added the appropriate references. In addition, standard units have been corrected and countries of origin for kits and reagents added.

Supplementary tables provide little extra information, where are the notes indicating what rs denotes? Why is there a ? in front of these rs scores? Consistency must be addressed e.g. "p-value" and "p". Figure's require legends which thoroughly describe the research presented. In an example, Figure 1A-C, 62 samples were analysed for hormone concentrations, yet there are 14 data points? What do the grey circle denote?

Reply: Thank you for pointing this out. The ‘rs’ represent correlation coefficients. The text in the supplemental table was garbled and a ‘?’ showed a negative value. We have corrected these points. In addition, explanations have been added to describe the result presented comprehensively.

Also authors must note that they are analysing the faecal microbiome, not the gastrointestinal microbiome.

Reply: Thank you for pointing this out. We have added the sentence: Most mammalian gut microbiome studies have focused on fecal microbiota, but it is unclear how well fecal samples reflect the microbiota of the intestinal region, and these should also be examined (lines 352-354).

Additionally, here are some specific edits I began writing:

Line 46: Please provide references for this statement

Reply: We have added the relevant references.

e.g.,

M Stöwe; T Bugnyar; C Schloegl; B Heinrich; K Kotrschal; E Möstl. Corticosterone excretion patterns and affiliative behavior over development in ravens (Corvus corax). Horm Behav 53, 208–216 (2008)

FB de Waal; F Aureli; PG Judge. Coping with crowding. Sci Am 282, 76–81 (2000)

DA Gust; TP Gordon; MK Hambright; ME Wilson. Relationship between social factors and pituitary-adrenocortical activity in female rhesus monkeys (Macaca mulatta). Horm Behav 27, 318–331 (1993)

MN Barbosa; MT da S Mota. Behavioral and hormonal response of common marmosets, Callithrix jacchus, to two environmental conditions. Primates 50, 253–260 (2009)

C Crockford; RM Wittig; K Langergraber; TE Ziegler; K Zuberbuhler; T Deschner. Urinary oxytocin and social bonding in related and unrelated wild chimpanzees. Proc Biol Sci 280, 20122765 (2013)

Line 59-61: Usually this is just called the gut-brain axis (GBA). References must be included at the end of this sentence.

Reply: We have added the relevant references.

JF Cryan; KJ O’Riordan; CSM Cowan; KV Sandhu; TFS Bastiaanssen; M Boehme; MG Codagnone; S Cussotto; C Fulling; AV Golubeva; KE Guzzetta; M Jaggar; CM Long-Smith; JM Lyte; JA Martin; A Molinero-Perez; G Moloney; E Morelli; E Morillas; R O’Connor; JS Cruz-Pereira; VL Peterson; K Rea; NL Ritz; E Sherwin; S Spichak; EM Teichman; M van de Wouw; AP Ventura-Silva; SE Wallace-Fitzsimons; N Hyland; G Clarke; TG Dinan. The Microbiota-Gut-Brain Axis. Physiol Rev 99, 1877–2013 (2019)

Line 62-65: Please state which model (human, rodent, in vitro) these studies were conducted in, as there is currently no studies which have assessed this in domestic cats.

Reply: Thank you for the suggestion. For enhanced clarity, we have added the models in the studies.

Line 67-72: Again, please state which species you are referring too. These statements form a more compelling argument for the group-living of domestic cats, if they are referring to research conducted in prides of lions as opposed to a school of dolphins.

Reply: These references were mainly about insects and rodents. We have identified the species examined in the studies.

Line 85: Please add a statement declaring where the cats were housed during the study. E.g. Adult domestic shorthair cats were housed at the XX facility/shelter, Japan for the duration of the study.

Reply: All cats were housed in one room at Azabu University during the experiment. We have added information on where the animals were maintained during the experiment: All cats were housed in one room at Azabu University during the experiment.(lines 89-90).

Line 89: two weeks

Reply: We have corrected (line 93).

Line 90: litter tray?

Reply: Yes, those were litter trays. We have changed ‘toilets’ to ‘litter trays (line 94).’

Line 97: As diet is such an important factor for the microbiome, it would be great to have the dietary information here. Even if only to add in a sentence such as: “Cats were fed a complete and balanced commercially available kibble/can diet”

Reply: We have added a sentence about diet: All cats were fed a complete and balanced commercially available kibble diet (lines 100-101).

Line 113: For future studies, collect urine from individuals and note which cat produced it. This is a far more accurate way of collecting data, rather than expressing the sample as a number per individuals.

Reply: Thank you for your comment. In the present study, we recorded which cats excreted and the time. A minimum 1 - maximum 9 urine samples were collected per individual. We have added the sentence in the hormonal assay section: Sixty-three urine samples, i.e., 4.2 ± 2.4 samples/individual (A minimum of one and a maximum of nine urine samples were collected per individual) were collected during the entire observation period… (lines 116-117).

Cortisol/Creatinine/Testosterone/Oxytocin Concentrations: Were these methods derived from existing published methods? If so, please insert references.

Reply: We have added the relevant references.

T Nagasawa; M Ohta; H Uchiyama. The Urinary Hormonal State of Cats Associated With Social Interaction With Humans. Front Vet Sci 8, 680843 (2021)

K Uetake; A Goto; R Koyama; R Kikuchi; T Tanaka. Effects of single caging and cage size on behavior and stress level of domestic neutered cats housed in an animal shelter. Anim Sci J 84, 272–274 (2013)

K Carlstead; JL Brown; W Strawn. Behavioral and physiological correlates of stress in laboratory cats. Appl Anim Behav Sci 38, 143–158 (1993)

K Kojima; S Hguchi; N Tashiro. Changes in urinary cortisol associated with hypthalamically elicited restlessness. Stress Med 11, 61–65 (1995)

Submitted filename: Response_to_reviewers_v3_final.docx

Click here for additional data file.

25 May 2022

Correlations between behavior and hormone concentrations or gut microbiome imply that domestic cats (Felis silvestris catus) living in a group are not like ‘groupmates’

PONE-D-21-34568R2

Dear Dr. Koyasu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Chun Wie Chong

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

30 May 2022

PONE-D-21-34568R2

Correlations between behavior and hormone concentrations or gut microbiome imply that domestic cats (Felis silvestris catus) living in a group are not like ‘groupmates’

Dear Dr. Koyasu:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Chun Wie Chong

Academic Editor

PLOS ONE