Effects of five carbohydrate sources on cat diet digestibility, postprandial glucose, insulin response, and gut microbiomes

Abstract

Cat obesity has become a serious problem that affects cats’ lives and welfare. Knowing how to control obesity in pet cats and its mechanism is urgently needed. Here, by feeding 30 cats different diets for 28 d, we explored the effects of 5 cat foods with potato, sweet potato, cassava, rice, and wheat as the main carbohydrate sources on the glycolipid metabolism of pet cats. The results showed that dietary carbohydrate sources did not affect the normal growth performance and stool scores of cats. Notably, we found that the starch gelatinization degree of sweet potato and cassava cat food were higher than those of other groups, while the rice diets had the highest digestibility, but the difference was not significant (P > 0.05). Furthermore, cats fed cassava diets had lower postprandial glucose responses. The mean glucose value, maximum glucose value, AUC_{0–360 min}, AUC_{≤30 min}, and AUC_{≥30 min} in the cassava group were lower than those in other dietary groups (P > 0.05). In addition, we found that the carbohydrate source had a minimal effect on serum biochemical immune indices, but the blood lipid indices, such as TG, TC, HDL, and LDL of cats fed the cassava diet were maintained at a low level compared with other groups (P > 0.05). In addition, diets with different carbohydrate sources affect the gut microbial composition, and sweet potato and cassava diets tend to increase the diversity of gut microbiota with a higher Shannon index and Simpson index. The abundance of Fusobacterium, Veillonella, and Actinobacillus was significantly higher in sweet potato diet-fed cats (P < 0.05), while the abundance of Delftia, Shinella, Rothia, and Hydrogenophage was highest in cassava diet-fed cats (P < 0.05). Collectively, this study revealed that cassava and sweet potato diets have a better effect on feeding value, controlling blood glucose and blood lipids, and improving the intestinal flora of pet cats, which is worth developing dietary formulations to alleviate pet obesity.

This study showed that cat food with cassava and sweet potato as the main source of carbohydrates has a better effect on controlling blood glucose and blood lipids and improving the intestinal flora of pet cats.

Introduction

In recent years, the pet industry has developed rapidly and the number of pets has been increasing. As good friends of humans, pets are considered to bestow positive health benefits on pet owners for improving mood and emotional state (Arhant-Sudhir et al., 2011; Barroso et al., 2021). Obesity is a prevalent disease in cats and is related to several diseases, such as insulin resistance and type 2 diabetes (Nelson and Reusch, 2014). It may also lead to dermatopathy, lameness, and inflammation, thus reducing both mental and physical quality of life (Hoelmkjaer and Bjornvad, 2014). Many factors contribute to cat obesity, such as breed preference (Kienzle and Moik, 2011), dietary habits, amounts of exercise, spay/neuter status, age, gender, and underlying medical conditions (Alexander et al., 2011; Courcier et al., 2012).

Cat obesity is mainly managed through dietary and exercise interventions. Although no studies have reported the effect of exercise and diet control on weight loss in cats, studies conducted on overweight dogs have shown that dietary caloric restriction is more effective than physical exercise (Chapman et al., 2019). This indicates that dietary regulation can alleviate obesity better. The nutrition needed by cats includes protein, carbohydrates, fat, minerals, and vitamins. Carbohydrates are the main source of energy for many body functions and an important component of pet food, accounting for 25% to 50% of dry foods (Carciofi et al., 2008; Asaro et al., 2018; Berman et al., 2022). Carbohydrates include various sugars, starches, cell wall components, and storage of nonstarch polysaccharides (Chandel, 2021). Starch is the most important carbohydrate in the cat diet. It is mainly derived from corn, wheat, rice, sorghum, and potatoes in modern commercial cat food. The rate and degree of starch digestion and absorption can affect postprandial blood glucose and insulin levels (Appleton et al., 2004; de-Oliveira et al., 2008). This indicates that it may help to reduce the reaction of postprandial glucose and insulin by adjusting the source and proportion of carbohydrates in cat food, thus controlling blood sugar and obesity and improving animal welfare.

Previous studies have mainly focused on the effects of different carbohydrate diets on glucose response and digestibility in cats. However, their effects on blood biochemical indices and the intestinal microflora composition of cats are still unknown. Therefore, the objective of this research was to investigate the effects of different sources of carbohydrates (potato, cassava, sweet potato, rice, and wheat) on cat diet digestibility, postprandial response of glucose, serum biochemical immune index, and microbiomes.

Materials and Methods

Animals and study design

All procedures and housing were approved by the Zhejiang University Animal Care and Use Committee. Thirty young cats (16 females, 14 males), including 6 American shorthairs, 5 Siam cats, 14 British Shorthairs, and 5 Ragdoll cats with a mean body weight of 3.14 ± 0.18 kg and a mean age of 0.60 ± 0.02 yr, were used. The cats were divided into 5 groups based on their varieties, body weight, and sex. The study lasted a total of 34 d, starting with a 7-d food adaptation phase and followed by a 28-d formal trial. Cats were housed in groups and socialized during the dietary adaptation phase. After the start of the formal experiment, the cats were housed individually in cages with a floor area of 0.28 to 0.37 m² and a height of 0.76 m. The ambient temperature was controlled at 20 to 22 °C, the humidity was 45% to 60 %, and the cats drank freely throughout the experiment.

Diets

A total of five diets were tested, and the composition of each diet is shown in Table 1. The diets were designed and structured to meet the nutritional requirements of young cats based on AAFCO recommendations and the total energy of the five experimental diets was 4200 kcal ME/kg. Each diet contains one of the following carbohydrates as its only source of starch: potatoes, cassava, sweet potato, rice, and wheat. Cats were fed diets with different carbohydrate sources. All diets were formulated and manufactured by Shanghai Full Pet Pet Product Co., Ltd. All cats were provided with 100 g of cat food on the first day. Then, we weighed and recorded daily feed intake to adjust the daily feeding amount for the following days. The feeding time was fixed at 0930 hours.

Table 1.

Ingredient composition of experimental diets for cats

	Potato	Cassava	Sweet potato	Rice	Wheat
Chicken powder	30.00	30.00	30.00	30.00	30.00
Potato flour	22.80	-	-	-	-
Cassava starch	-	19.70	-	-	-
Dried sweet potato	-	-	32.00	-	-
Rice	-	-	-	24.70	-
Wheat	-	-	-	-	28.90
Fish meal	16.20	20.00	10.70	6.30	7.90
Pea	15.00	15.00	3.10	15.00	15.00
Beet pulp	4.00	4.00	4.00	4.00	2.00
Cat premix	4.00	4.00	4.00	4.00	4.00
Beer yeast	3.00	3.00	3.00	3.00	3.00
Cheese powder	2.00	2.00	2.00	2.00	2.00
Isolated Soy Protein	2.00	1.30	9.90	10.00	6.20
Beef bone meal	1.00	1.00	1.00	1.00	1.00
Poultry fat	13	13	14	13	13
Cellulose	-	-	0.30	-	-

All values are expressed as % inclusion as fed.

Determination of growth performance

The feed and water supply were cut off at 2100 hours the day before the formal experiment and the end of the test, and the body weight was weighed at 0830 hours the next morning. Then we recorded the initial body weight (IW), final weight (FW), daily feeding amount, and remaining feed amount of each cat to calculate the weight gain (WG), average daily gain (ADG), and average daily feed intake (ADFI).

Digestibility protocol

Feces were collected at the end of the experiment, weighed and kept frozen (–80 °C) until analysis. The content of starch in feces was determined by acid hydrolysis-3,5-Dinitrosalicylic acid (DNS) colorimetry as previously described (Wood et al., 2012). The starch content in each diet was determined according to ISO 6493: 2000 (Animal feeding stuffs-Determination of starch content-Polarimetry method, MOD). Then, the starch digestibility of different diets was calculated according to the following formula:

$${\displaystyle \begin{array}{l}{\displaystyle \mathbf{S}\mathbf{t}\mathbf{a}\mathbf{r}\mathbf{c}\mathbf{h}\mathbf{d}\mathbf{i}\mathbf{g}\mathbf{e}\mathbf{s}\mathbf{t}\mathbf{i}\mathbf{b}\mathbf{i}\mathbf{l}\mathbf{i}\mathbf{t}\mathbf{y}=\frac{\mathbf{S}\mathbf{t}\mathbf{a}\mathbf{r}\mathbf{c}\mathbf{h}\mathbf{c}\mathbf{o}\mathbf{n}\mathbf{t}\mathbf{e}\mathbf{n}\mathbf{i}\mathbf{n}\mathbf{d}\mathbf{i}\mathbf{e}\mathbf{t}-\mathbf{S}\mathbf{t}\mathbf{a}\mathbf{r}\mathbf{c}\mathbf{h}\mathbf{c}\mathbf{o}\mathbf{n}\mathbf{t}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{i}\mathbf{n}\mathbf{f}\mathbf{e}\mathbf{c}\mathbf{e}\mathbf{s}}{\mathbf{S}\mathbf{t}\mathbf{a}\mathbf{r}\mathbf{c}\mathbf{h}\mathbf{c}\mathbf{o}\mathbf{n}\mathbf{t}\mathbf{e}\mathbf{n}\mathbf{t}\mathbf{i}\mathbf{n}\mathbf{d}\mathbf{i}\mathbf{e}\mathbf{t}}\times 100\mathrm{\%}}\end{array}}$$

Fecal score

At the end of the experiment, we performed a quick on-site scoring of each cat’s feces according to The WALTHAM Feces Scoring System (Cavett et al., 2021). Grade 1 = Bullet like crumbles with little pressure; Grade 1.5 = Hard and dry, stool cracks when pressed; Grade 2 = Well formed, does not leave a mark when picked up; Grade 2.5 = Well formed with a slightly moist surface, leaves a mark when picked up; Grade 3 = Moist, beginning to lose form, leaving a definite mark when picked up; Grade 3.5 = Very moist, still with some definite form; Grade 4 = Most or all form is lost, no real shape; Grade 4.5 = Liquid stool with slight consistency; Grade 5 = Completely liquid stool.

After the stool score was determined, we collected the fresh feces in the collection tube immediately after the cats defecated, and this work was completed within 10 to 15 min. Then, we quickly stored the fecal samples at –80 °C for subsequent laboratory analysis.

Postprandial response tests

On the 28th day of the experiment, the postprandial response test was performed to evaluate the postprandial blood glucose response of cats. Blood glucose was measured with a blood glucose test strip before feeding (baseline sample, time 0) and 15, 30, 60, 120, 180, 240, 300, and 360 min after feeding. The blood glucose value was then recorded to form a blood glucose curve, and the area under the curve (AUC) was used to quantify the blood glucose response.

Determination of serum biochemical and immune indices

After completing the postprandial glycemic response test, blood was collected to determine serum biochemical immunological indicators. The scalp needles and syringes were inserted into the forelimb veins to collect blood from each cat, and if the forelimb veins fell, the hind veins were used for blood collection. Blood samples (1.5 mL) were obtained by syringe and transferred to heparin sodium, tubes and then centrifuged (2000 r × 5 min), and the upper serum was collected to determine biochemical and immune indices. Anesthesia was not administered during blood collection. Serum insulin, blood glucose (GLU), fructosamine (FRUC), β-hydroxybutyric (β-HB), total protein (TP), immune protein M (IgM), triglyceride (TG), total cholesterol (TC), high-density lipoprotein (HDL), low-density lipoprotein (LDL), and apolipoprotein A (Apo A) were determined by an automatic biochemical analyzer.

16S rDNA sequencing and gut microbiota analysis

The total RNA of intestinal flora was extracted from fecal samples and 16S rDNA was amplified. The samples were sequenced on an Illumina NovaSeq platform (LC-Bio Technology Co., Ltd, Hangzhou, China). After obtaining the original off-machine data, quality control and chimera filtering were carried out to obtain high-quality data. The noise of the amplified sub-data was reduced based on QIIME2, and the quasi-OTU table was then constructed using the concept of amplicon sequence variants (ASVs). Further diversity analysis, species classification annotation, and difference analysis were carried out.

Statistical analyses

The analysis and drawing of the test data were carried out by GraphPad Prism 8.0.

One-way ANOVA was performed among multiple groups of data. The Kruskal–Wallis test was used to analyze the differential flora among groups. Results are expressed as the mean ± sem. P < 0.05 indicates that the difference is significant, and P > 0.05 indicates that the difference is not significant.

Results

Growth performance and stool scores

The growth performance and fecal score of cats fed experimental diets are shown in Table 2. The IW of cats between different diets ranged from 3027.83 ± 304.63 to 3245.33 ± 285.19 g. We then compared the FW, WG, ADG, and ADFI among the five groups after the 28-d experimental period. The results showed that there was no significant difference in IW, FW, WG, ADG, and ADFI among the cats in each diet group, but the WG, ADG, and ADFI in the cassava group were higher than those in the other four groups (P > 0.05). In addition, the stool scores of cats in each group were not significantly different and were all in the normal range (P > 0.05). These results demonstrate that feeding these five carbohydrate diets will not hurt cats, and all cats in each group are in a normal physiological state.

Table 3.

Growth performance and fecal score of cats fed experimental diets

Item	Potato	Cassava	Sweet potato	Rice	Wheat
IW(g)	3197.67 ± 140.05	3245.33 ± 285.19	3027.83 ± 304.63	3100.33 ± 313.27	3131.00 ± 325.75
FW(g)	3392.17 ± 204.03	3557.17 ± 276.53	3232.50 ± 338.41	3356.00 ± 261.17	3251.67 ± 354.42
WG(g)	194.50 ± 4.36	311.83 ± 6.55	204.67 ± 4.45	255.67 ± 2.91	120.67 ± 6.22
ADG(g/d)	6.95 ± 3.27	11.14 ± 3.15	7.31 ± 1.61	9.13 ± 3.45	4.31 ± 1.88
ADFI(g/d)	63.34 ± 4.36	73.34 ± 6.55	64.14 ± 4.45	69.13 ± 2.91	64.14 ± 6.22
FC	2.92 ± 0.33	3.17 ± 0.31	2.67 ± 0.11	2.92 ± 0.35	2.67 ± 0.31

IW: initial weight; FW: final weight; WG: weight gain; ADG: average daily gain; ADFI: average daily feed intake; FC: fecal score⁺.

Starch gelatinization and digestibility

The starch gelatinization degree and digestibility of five experimental diets are shown in Table 3. The results showed that the starch gelatinization degree of the sweet potato diet was the highest at 90.2%, followed by the cassava diet at 87.3%. This result suggests that the starches from both diets were more susceptible to enzymatic hydrolysis and digestion. We further measured the starch digestibility of different diets, and surprisingly found that the starch digestibility of the rice diet was the highest at 99.00%, followed by the wheat group at 98.72% and the sweet potato group at 98.60%. However, the difference among groups was not significant (P > 0.05). This shows that the rice diet group has better digestion and absorption of starch and high feeding value.

Table 2.

Chemical composition, degree of starch gelatinization of five experimental diets, and diets starch digestibility of cats fed with different diets

		Potato	Cassava	Sweet potato	Rice	Wheat
Crude Protein(%)	Theoretical value	36	36	36	36	36
	Measured values	34.83	35.04	33.98	36.64	36.65
Crude fat(%)	Theoretical value	15	15	15	15	15
	Measured values	14.41	14.70	14.41	13.27	14.24
Crude fiber(%)	Theoretical value	-	-	-	-	-
	Measured values	3.63	3.3	3.93	4.17	3.73
Ca(%)	Theoretical value	1.76	1.73	1.64	1.53	1.45
	Measured values	1.27	1.22	1.20	1.25	1.24
P(%)	Theoretical value	1.23	1.20	1.14	1.19	1.18
	Measured values	1.35	1.28	1.19	1.14	1.15
Carbohydrate(%)	Theoretical value	20	21	21	21	18.5
	Measured values	20.2	19.4	19.3	22	19.6
Degree of starch gelatinization (%)	Measured values	80.80	87.30	90.20	72.40	80.30
Starch digestibility (%)	Measured values	97.95 ± 0.39	98.60 ± 0.11	98.22 ± 0.27	99.00 ± 0.47	98.72 ± 0.47

The above indicators are presented in percentage data.

Postprandial response of glucose

The postprandial blood glucose response of diets with five different carbohydrate sources is shown in Figure 1 and Table 4. Compared with baseline values (0 min), the cassava diet stimulated a significant increase in the glucose response occurring 180 min after a meal (P < 0.05) (Figure 1B); the sweet potato diet stimulated a significant increase in the glucose response occurring 240 and 360 min after a meal (P < 0.05) (Figure 1C); and the wheat diet stimulated a significant increase in the glucose response occurring 180 and 360 min after a meal (P < 0.05) (Figure 1E).

Figure 1.

Postprandial glycemic response curves of cats fed diets with different carbohydrate sources. Values are mean ± sem of the six cats per diet. Blood glucose values at each time point were compared with the baseline (0 min). *P < 0.05.

Table 4.

Basal glucose, mean glucose, maximum glucose, time to peak glucose, and area under blood glucose response curve of postprandial 0-360 min in cats fed experimental diets

	Potato	Cassava	Sweet potato	Rice	Wheat
Basal glucose (md/dL)	91.67 ± 4.60	92.50 ± 4.34	93.00 ± 4.14	92.33 ± 3.68	89.83 ± 2.81
Mean glucose (md/dL)	106.89 ± 2.82	101.02 ± 1.76	104.81 ± 2.28	105.19 ± 3.20	101.37 ± 2.43
Maximum glucose (md/dL)	117.83 ± 2.08	113.50 ± 3.02	119.00 ± 2.46	117.17 ± 3.89	114.67 ± 4.14
Time to glucose Peak (min)	235.00 ± 55.00	140.00 ± 20.00	280.00 ± 29.66	205.00 ± 49.24	190.00 ± 45.61
AUC 0 to 360min (md/dL min)	38421.67 ± 1188.55	36810.33 ± 716.58	38096.50 ± 935.47	38280.17 ± 1320.36	37240.17 ± 1073.56
AUC ≤ 30 min (md/dL min)	3016.67 ± 56.47	2800.33 ± 68.24	2871.50 ± 32.09	2892.67 ± 121.77	2922.67 ± 86.46
AUC ≥ 30 min (md/dL min)	35405.00 ± 1188.15	34010.00 ± 692.28	35225.00 ± 919.05	35387.50 ± 1282.31	34317.50 ± 1030.48

We further calculated the basal glucose, mean glucose, maximum glucose, time to glucose peak, the total area under the curve (AUC_{0-360 min}), immediate meal response (AUC_{≤30 min}), and later meal response (AUC_{≥30 min}) according to the postprandial blood glucose response curve. The results showed that the basal glucose value of cats fed the wheat diet was the lowest, which was 89.83 ± 2.81 mg/dL (P > 0.05). The mean glucose and maximum glucose values of cats in the cassava diet group were lower than those of the other dietary groups, and the time to reach the peak was shortest in the cassava group and longest in the sweet potato group (P > 0.05). The blood glucose response (measured by AUC) showed no significant difference among the groups, with AUC_{0-360 min}, AUC_{≤30 min}, and AUC_{≥30 min} being lower in the cassava group than in the other dietary groups, while those in the potato dietary group were the highest (P > 0.05).

Serum biochemical immune indices

The serum biochemical immune indices of cats fed diets of different carbohydrate sources are shown in Figure 2. First, the content of serum insulin in the potato diet group was higher than that in the other dietary groups, but the difference was not significant (P > 0.05). Blood glucose-related indices included GLU, FRUC, and β-HB. The GLU content of cats in the potato diet group was higher than that in the other diet groups, and the contents of FRUC and β-HB were similar (P > 0.05). Blood lipid-related indices included TG, TC, HDL, LDL, and Apo A. The results showed that the contents of serum TG, TC, HDL, and LDL in the cassava diet group were lower than those in the other groups, although the difference was not significant (P > 0.05). The Apo A content among the five groups was very similar, with no significant difference (P > 0.05). We further tested the contents of immune indices and found that there was no significant difference in TP and IgM between the groups (P > 0.05). Combined with the results of each index, the physiological status of cats in each dietary group was similar, and the related indices of blood lipids in the cassava group were maintained at a low level.

Figure 2.

Serum biochemical immunity indices of cats fed diets with different carbohydrate sources. Values are mean ± sem of the six cats per diet. GLU, glucose; FRUC, fructosamine; β-HB, β-hydroxybutyric; TG, triglyceride; Tchol, total cholesterol; HDL, high-density lipoprotein; LDL, ow-density lipoprotein; Apo A, apolipoprotein A; TP, total protein.

ASV distribution Venn diagram, α-diversity, and composition of the gut microbiota at the phylum and genus levels

The distribution of ASVs in each group is shown in Figure 3. The Venn diagram showed that the cat fecal ASVs in the sweet potato group were the most abundant, followed by the cassava group, and the number of ASVs shared by the cassava and sweet potato groups was 444. In addition, the sweet potato group’s α-diversity (Shannon index and Simpson index) was higher than those of the other groups, but the difference was not significant (P > 0.05). This indicates that the intestinal flora of cats fed the sweet potato diet and cassava diet were similar, while the sweet potato diet tended to increase microbial community diversity.

Figure 3.

Effect of diets with different carbohydrate sources on intestinal microbiota composition in cats. (A) Venn diagram of ASV distribution. (B) Shannon index. (C) Simpson index. (D) Bar diagram of phylum level microbial composition. (E) Bar diagram of genus level microbial composition. *P < 0.05.

We further analyzed the microbial abundance at both the phylum and genus levels. At the phylum level, the abundances of Firmicutes, Actinobacteriota, Bacteroidota, and Proteobacteria were higher. In sweet potato diet-fed cats, the abundance of Bacteroidota and Proteobacteria was increased compared with the abundance in the other groups of cats. The gut microbiota at the genus level mainly included Collinsella, Megasphaera, Bifidobacterium, Solobacterium, and Catenibacterium.

The abundance of differential microbiota at the genus level

We used the Kruskal–Wallis test to analyze the differential flora among groups at the genus level (Figure 4). A total of 11 bacterial lineages were identified as significant under the condition that the P value < 0.05. Three species, Fusobacterium, Veillonella, and Actinobacillus, were more abundant in sweet potato diet-fed cats than in other groups, while Haemophilus and Neisseria were more abundant than rice and wheat diet-fed cats. The abundance of Delftia, Shinella, Rothia, and Hydrogenophage in cassava diet-fed cats was significantly higher than that in rice and wheat diet-fed cats, and the abundance of Acidovorax was significantly higher than that in rice diet-fed cats. In addition, the abundance of Olsenella was significantly higher in wheat diet-fed cats than in potato and cassava diet-fed cats.

Figure 4.

Effects of diets with different carbohydrate sources on the abundance of genus level bacteria in cats.

Discussion

Approximately 63% of cat owners reported their cats being overweight or obese, and this proportion continues to grow (Wallis and Raffan, 2020). Obesity in cats may lead to a variety of related diseases, a decline in quality of life, and a shorter life expectancy (Wall et al., 2019). Preventing and alleviating cat obesity are urgently needed. Humans may lose weight through bariatric surgery, but this is not a suitable solution for pets due to morality and practicality. In addition, exercising may aid weight loss, but the effects are less obvious than diet management, and it is challenging for both owners and pets to exercise regularly (Swift et al., 2014; Vitger et al., 2016; Chapman et al., 2019). Thus, dietary management, such as dietary caloric restriction or offering a low-energy and high-protein diet, remains the most recommended and feasible method to treat cat obesity (Blanchard et al., 2004; German et al., 2007). Carbohydrates are the primary energy source of cat food. It is reported that excessive amounts of dietary carbohydrates may cause an increase in insulin and eventually lead to obesity (Buffington, 2008; Verbrugghe and Hesta, 2017). Here, we tested the effects of five diets with potato, cassava, sweet potato, rice, and wheat as the sole sources of starchy carbohydrates on glucose and lipid metabolism in cats.

First, we found the source of carbohydrates did not affect the normal growth performance and the fecal scores of cats, which is by previous findings (de-Oliveira et al., 2008). This may be because these five carbohydrates are in the normal range, in accordance with the nutritional needs of cats. In addition, time may also be one of the reasons as our experiment lasted for 28 d and the normal growth of cats did not change significantly in a short period, which needs to be further confirmed by long-term observations. Another meaningful finding was that sweet potato and cassava diets showed better digestibility potential. Our results demonstrated that the sweet potato diet had the highest starch gelatinization degree, followed by the cassava diet. This may be due to the different structures of starches from different carbohydrate sources. Starch will be gelatinized, and its ordered structure will be irreversibly transformed into a disordered structure when heated in water (Liu et al., 2009). The starch gelatinization degree affects starch digestibility and food quality and is an important index by which to evaluate the processing quality of pellet feed (Wang and Copeland, 2013; Nayak et al., 2014). In general, the higher the gelatinization of starch, the easier it is to be hydrolyzed by enzymes, which is conducive to digestion and absorption. However, we found that the starch digestibility of the rice diet was the highest. Rice has a small starch granular structure, with a large surface area for digestive enzyme action during digestion. Therefore, rice-based diets are more easily digestible and do not depend on the granularity or gelatinization of the raw material (Bazolli et al., 2015). This result suggests that we need to pay attention to the performance of the rice group in other aspects.

Carbohydrates are one of the major nutrients affecting the postprandial glucose response (Papakonstantinou et al., 2022). However, a previous study suggested that starch source and levels had minimal effects on postprandial glucose and insulin responses in adult lean cats (de-Oliveira et al., 2008; Asaro et al., 2018). These findings support our results that the blood glucose response was not significantly different among the five groups. This is probably because cats are carnivores (Kienzle, 1993). The salivary amylase and intestinal amylase activity in cats are low (McGeachin and Akin, 1979). Therefore, the postprandial glucose response in cats is insensitive to diets with different carbohydrate sources. In addition, we found that the mean glucose, peak glucose, AUC_{0-360 min}, AUC_{≤30 min}, and AUC_{≥30 min} were lower in cats fed the cassava diet than in the other groups. We speculate that cats fed cassava diets had a lower postprandial glucose response. Obesity in cats may predispose them to type 2 diabetes, reduce insulin sensitivity, disrupt glucose homeostasis, and lead to dyslipidemia (Kumar et al., 2012; Osto and Lutz, 2015). By comparing the results from serum biochemical immune indices, we hope to determine whether the cassava diet may help to control obesity. We found that the contents of FRUC and β-HB were similar among the five dietary groups and that the contents of serum TG, TC, HDL, and LDL in the cassava diet group were lower than those in the other groups. From the results, it is clear that the related indices of blood lipids in the cassava group were maintained at a low level. These results are consistent with those of previous studies wherein obese cats had greater serum concentrations of GLU, TG, and INS than ideal cats, but FRUC concentrations did not differ between groups (Zapata et al., 2017; Williams et al., 2019).

The gut microbial community is involved in the cats’ body metabolism and physiological activities, such as the digestion and absorption of nutrients and the induction and shaping of the body’s immune system, affecting the health of the host (Mondo et al., 2019; Wernimont et al., 2020). Age, diet, intestinal diseases, and interactions between pets and humans are the main factors that affect the composition of gut microbiota (Alessandri et al., 2020). Among them, diet largely alters the diversity and function of the mammalian gut microbiome (David et al., 2014). The microbial diversity of the gut microbiota of obese cats was significantly reduced, suggesting dysbiosis in the gut microbiota of obese cats (Ma et al., 2022). Here, we found that the sweet potato diet tended to increase microbial community diversity. Meanwhile, the ASVs shared by the cassava group and sweet potato group were as high as 444. We may consider that both cassava and sweet potato diets may help relieve obesity. In addition, we found that Firmicutes, Actinobacteriota, Bacteroidota, and Proteobacteria were the main bacterial phyla, while Collinsella, Megasphaera, Bifidobacterium, Solobacterium, and Catenibacterium were the major bacterial genera of cats fed diets with different carbohydrate sources. The same results were also found in the study by Handl and Minamoto et al. (Handl et al., 2011; Minamoto et al., 2012). The composition of intestinal flora is different between obese cats and skinny cats. A previous study reported that compared with obese cats, Fusobacteria and Veillonella showed increased abundance in the lean group (Kieler et al., 2016; Li and Pan, 2020). Here, we identified similar results that Fusobacterium, Veillonella, and Actinobacillus were more abundant in the sweet potato group, while Delftia, Shinella, Rothia, and Hydrogenophage were more abundant in the cassava group than in the other groups. A recent study performed whole-genome shotgun metagenomic sequencing of obese cats and found that microbial composition differences occurred at the most abundant genera levels, with large differences in the abundance of Bacteroides, Lactimicrobium, and Phascolarctobacterium between normal and obese felines (Ma et al., 2022). These results are not entirely consistent, so more research is needed to identify marker microbes in obese cats. Based on the above results, it can be considered that cassava and sweet potato diets tend to increase the diversity of intestinal microflora and reduce obesity in pet cats.

In conclusion, our data showed that cat food with potato, cassava, sweet potato, rice, and wheat as the main carbohydrate sources, did not have a negative impact on pet cats and that the cats in each dietary group were in normal physiological conditions. Different carbohydrate diets will affect the diversity, richness, and population structure of intestinal microflora. Among them, cassava and sweet potato diets could better control blood glucose, and blood lipid levels, and improve gut flora. Our research does provide a valuable reference for the research and development of pet food. However, the optimal proportion of sweet potato and cassava in the diet and its specific regulatory mechanism are not clear and need to be further explored. In addition, attention should be given to dog obesity in the future. These findings are of great significance for revealing the regulatory mechanism of glucose and lipid metabolism in obese pets and the development of diets to reduce pet obesity.

Glossary

Abbreviations

ADFI

average daily feed intake

ADG

average daily gain

Apo A

apolipoprotein A

ASVs

amplicon sequence variants

AUC

area under the curve

FC

fecal score+

FRUC

fructosamine

FW

final weight

GLU

glucose

HDL

high-density lipoprotein

IW

initial weight

LDL

low-density lipoprotein

TC

total cholesterol

TG

triglyceride

TP

total protein

WG

body weight gain

β-HB

β-hydroxybutyric

Data Availability

Data are available from the authors upon reasonable request.

Conflict of Interest Statement

The authors have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Yingchun Wang, Han Dan, and Yang Ren are employees of Shanghai Full Pet Pet Product Co., Ltd. The funders and employees had no role in the design of the study; analysis, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. No other author has reported a potential conflict of interest relevant to this article.

Author contributions

TZ designed the research. SZ conducted the research, collected and assembled the data. SZ mainly wrote the manuscript draft. TZ, YR, YQ, and YC edited and revised the manuscript. HD participated in the feeding experiment and collected samples. All authors contributed to the article and approved the submitted version.