← Science Hub
The Microbiome

Independent Research: How Different Fibres Shape the Canine Microbiome and Metabolome

A peer-reviewed study of 18 healthy adult dogs tested 12 foods containing different amounts and types of dietary fibre. By combining faecal metagenomics with metabolomics, researchers showed that diet altered both microbial composition and the chemical environment of the gut, with metabolite responses often stronger and more individualised than changes in the microbiome itself.

Peer reviewed Open access
Independent Research: How Different Fibres Shape the Canine Microbiome and Metabolome
Year
2024
Study type
Peer-reviewed controlled dietary intervention study
Sample
18 healthy adult dogs
Duration
12 sequential 7-day feeding periods
Measured
Faecal microbiome composition, microbial functional pathways and faecal metabolome
01

The research question

Dietary fibre is not one single substance.

Different fibres have different chemical structures and can be used by different members of the gut microbial community.

This means that two diets containing similar amounts of fibre may not necessarily have the same effect on the microbiome or on microbial metabolism.

Bhosle and colleagues therefore asked:

How do different types and amounts of dietary fibre alter the canine gut microbiome and metabolome?

The study also addressed a deeper question:

Does the same fibre produce the same metabolic response in every dog?

02

What did the researchers do?

Eighteen healthy adult dogs took part.

The group included 15 beagles and three mixed-breed dogs, with a mean age of 9.4 years.

Each dog was fed a sequence of 12 different test foods.

Each food was provided for seven days before the dog moved to the next diet.

The foods varied in:

Total fibre content

Soluble and insoluble fibre

Starch content

Specific fibre sources

The diets contained approximately 5% to 13% fibre on an as-fed basis.

Faecal samples were collected at the end of each feeding period.

Researchers analysed those samples using two complementary approaches.

Metagenomic sequencing

This provided detailed information about microbial species and their potential functional pathways.

Metabolomics

This measured a broad range of chemical compounds present in the faecal samples.

The researchers could therefore compare:

which microorganisms were present

with

which metabolites were present

across the different diets.

03

What did they measure?

The study examined several levels of the gut ecosystem.

Microbial species

Researchers measured changes in the relative abundance of individual microbial species.

Microbial functional pathways

Metagenomic data were used to examine the metabolic capabilities encoded within the microbial community.

Faecal metabolites

A large number of compounds were measured, including:

short-chain fatty acids

acylglycerols

fibre-associated sugars

polyphenols

and other products of microbial and dietary metabolism.

Dietary composition

The researchers compared microbiome and metabolome responses with:

total fibre

insoluble fibre

soluble fibre

starch

and specific fibre sources.

Individual response

The team also examined whether the same dietary change produced similar responses in every dog.

04

What was observed?

The study produced several particularly important findings.

Diet altered the metabolome more strongly than the microbiome

Food explained approximately 33.7% of variation in faecal metabolite profiles.

By comparison, diet explained around 15.4% of variation in microbiome composition.

This means the chemical activity within the gut changed more dramatically than the overall microbial community structure.

Individual dogs had highly individual microbiomes

Individual identity explained more microbiome variation than diet.

Around 28.6% of microbiome variation was associated with differences between individual dogs.

This shows how strongly personalised the canine microbiome can be.

Specific fibres were associated with specific microbes

Fourteen microbial species were significantly enriched in response to at least one particular fibre source.

The response was therefore not simply a case of “more fibre equals more beneficial bacteria”.

The type of fibre mattered.

Metabolites changed alongside microbial responses

Fibre-associated changes included differences in compounds such as short-chain fatty acids, acylglycerols, fibre-bound sugars and polyphenols.

For eight of the 12 foods, larger differences in microbiome composition were associated with larger differences in metabolite profiles.

Different dogs responded differently to the same fibre

One of the most interesting findings was that the metabolic response to fibre depended partly on the microbial community already present in the individual dog.

The same fibre could therefore produce different microbial and metabolite responses in different animals.

05

What does this study tell us?

This study is a particularly useful demonstration of why the gut microbiome should not be treated as a static list of bacteria.

Changing the nutritional substrate altered both:

which microorganisms were relatively abundant

and

what chemical compounds were present

But the chemical response was often greater than the compositional response.

That reinforces an important distinction:

Microbiome composition tells us who is there.

Metabolomics helps tell us what is happening.

The study also provides strong evidence for a substrate-led view of the microbiome.

Different fibres produced different microbial and metabolic responses.

The effect depended not only on the fibre being supplied but also on the microbial ecosystem already present in the individual dog.

A useful way to express the relationship is:

Dietary substrate + Existing microbiome → Microbial activity → Metabolites

This is a much more nuanced model than assuming that a particular ingredient will produce exactly the same microbiome response in every dog.

06

What doesn't this study tell us?

The study has several limitations that are important when interpreting the results.

Small population

Only 18 dogs were included.

Although each dog generated many samples, the number of individual animals remained relatively small.

Dogs were research-colony animals

The dogs were owned by Hill’s Pet Nutrition and lived under controlled conditions.

Their responses may not fully represent the wider companion-dog population.

Short feeding periods

Each test food was fed for seven days.

Some microbiome adaptations may take longer to stabilise.

No washout period between most foods

Dogs moved directly from one test food to the next.

Carryover effects from the previous diet therefore cannot be completely excluded.

Sequential rather than randomised order

All dogs received foods in the same sequence.

This makes it more difficult to separate diet effects from possible time or order effects.

Industry involvement

The study was funded by Hill’s Pet Nutrition, two authors were Hill’s employees, and the dogs and diets were provided by the company.

That does not invalidate the research, but it is relevant context when assessing independence.

Metabolite associations do not prove direct production

Finding a metabolite alongside a microbial species does not necessarily mean that species produced the compound.

Multiple organisms and metabolic pathways may contribute.

Original research

Study details

Study
Response of the gut microbiome and metabolome to dietary fiber in healthy dogs
Authors
Amrisha Bhosle
Matthew I. Jackson
Aaron M. Walsh
Eric A. Franzosa
Dayakar V. Badri
Curtis Huttenhower
Institutions
Broad Institute of MIT and Harvard

Harvard T. H. Chan School of Public Health

Harvard Chan Microbiome in Public Health Center

Hill’s Pet Nutrition
Reference
Bhosle A, Jackson MI, Walsh AM, Franzosa EA, Badri DV, Huttenhower C. Response of the gut microbiome and metabolome to dietary fiber in healthy dogs. mSystems. 2024;10(1):e00452-24.

Published online 23 December 2024.

Eighteen healthy adult dogs were fed 12 sequential test foods differing in starch, fibre quantity and fibre source. Faecal samples were analysed using metagenomic sequencing and metabolomics.

DOI: 10.1128/msystems.00452-24
View source ↗
Scientific transparency

No single study provides every answer. We show methods and limitations alongside findings so the evidence can be understood in context.