"Without it, young mice face long-lasting consequences, including several signs of inflammatory bowel diseases (IBD)."
The first level popular press article does not say that they develop symptoms reliably, just signs. Which could mean clinical measurements which are associated with IBD rather than symptoms. If you read the abstract this appears to be the case. Patterns of gene expression, microbial populations, etc. Not sure they looked at actual disease at all. Hard to ask a mouse if it has a stomach ache.
A further possibility is that it may well not be 100% in the mice. Would have to read the whole article, which isn't trivially available... sadly.
Another is that the symptoms are much more reliable in the strain of extremely inbred lab mice they used (all lab mice are extremely inbred). Wild mice or humans may be more robust via diversity.
Another is that the very uniform prison like environment lab mice live in makes the outcome more reliable. Non-breastfed human children are likely encountering the gut bacteria of lots of other humans. These mice are likely only encountering their mother, and may even be separated from her early. Would have to read the paper.
I think the hypothesis on why infant immune systems aren't fully active immediately is that it is going through some initialization steps where it learns to ignore self and constant environmental factors. It cannot do that and simultaneously react to threats, so the mother's milk supplies a stream of antibodies which ought to be good against whatever diseases are floating around (because the mother already had them, or gets them along with the child). However it is also entirely possible that it doesn't turn on earlier mainly because it was never a problem until quite recently. Any baby that did not get breast fed was almost certain to die anyway.