They all but disappeared during the Dark Ages of Medieval Europe and were rediscovered in France in the 1700s. Sir William Gage introduced the gages to England in the 1720s and subsequently both lost the varietal labels and named them after himself. The trees are weak to moderate in vigor and extremely narrow and upright. At their tree-ripe perfection in late July and August, the gages feature a green, yellow, or golden skin and a sugary sweet taste with slight tangy undertones that is arguably the most intensely rich-tasting fruit on the planet. True green gage plums are hard to find but worth the search.This species originated in China 2,000 years ago, was introduced to Japan in the 1600s, and subsequently brought to the U.S. by horticulturists John Kelsey and Luther Burbank.Burbank used this stock to breed the Satsuma, the Santa Rosa plum, and countless other varieties that founded the California plum industry. The fruit is large and heart-shaped to conical. The skin color can range from golden yellow, orange-red, or blood red to purple and black. Flesh color usually reflects a variation on the skin color. The taste is slightly acid over sweet. They are best eaten fresh. The flesh is juicy and unlike European plums they are not freestone, two notable exceptions being Satsuma and its improvement, Mariposa. These two varieties also feature less acidity and thus can be dried, a la prune plums. Japanese plums bloom abundantly early in the season , and thus fruit earlier than European plums . They generally produce heavy crops; if even 1–2% of the blooms set fruit, thinning is required. They tolerate milder winters, that is to say they bloom and set fruit with less chill hours than European plums. The trees tend to be vigorous, rambunctious growers, often exceeding 10 feet a year on standard rootstocks. They are very upright growers with the exception of the Satsuma and Mariposa varieties, trimming cannabis which again exhibit a prune plum-like growth habit. Their pollination needs are similar to European plums.Domestica plums should be pruned hard to stimulate continued vegetative growth throughout their life. As with peaches, when a plum branch goes flat it weakens and produces smaller and smaller fruit.

Prune to an inward or upward facing bud to redirect flat growth upward. Japanese plums should rarely be stimulated via heading cuts once established. Heading causes multiple narrow-angled , excessively vigorous regrowth. Pruning at maturation devolves to the occasional thinning cut and the renewal of the brushy lateral fruit bearing growth. Japanese flower buds have a cluster of 3–5 blossoms that live for 3–5 years. In any given pruning session 20% of these laterals should be stubbed back to 1–3 buds and regrown. They will fruit in the second year after renewal. Thinning for Japanese and European plums should be one to a cluster every 4–6 inches. Oversetting results in a nutrient sink that inhibits bloom and fruiting the next year . As with peaches they can and probably should be rescaffolded periodically .The current way we produce and consume food threatens both human health and environmental sustainability. In 2019, the EAT-Lancet Commission launched the planetary health diet, a global reference diet with focus on healthy diet produced in a sustainable way. The main objective of this diet is to increase the consumption of plant-based foods including vegetables, fruits, legumes, whole grains, and nuts while reducing the consumption of animal-sourced foods such as red and processed meat and dairy products. Another rapidly expanding area of research is understanding the complex relationships between diet, gut microbiome, and human health. Human gut microbiota refers to a complex community of trillions of different micro-organismsresiding in the human gut. Diet is considered one of the most important factors influencing composition and function of the gut microbiome, and thus, determining its metabolic outputs that may play a role in human health and disease. Consequently, many diet-associated conditions have been associated with the gut microbiome including obesity and several chronic diseases such as type 2 diabetes and cardiovascular diseases. Thus, it is important to examine the specific roles of different food groups on the gut microbiome.

The associations of plant-based foods, red and processed meat and dairy consumption with gut microbiome have not been extensively examined. Controlled small-scale human trials conducted mainly on individuals with obesity have demonstrated shifts in microbiome diversity or composition and adverse changes in microbial metabolites on diets high in animal-based foods and low in carbohydrates during 5-days to 8-weeks. Larger-scale observational studies on healthy French adults and on Chinese middle-aged and elderly have reported inconsistent results on the associations between individual plant-based foods, red meat or dairy with gut microbiome. These studies, however, lacked data on actual consumption of the foods due to the utilized dietary assessment method and on some foods in the core of this current study such as dairy subgroups . These apply also to our previous study, consisting partly of the same study population, where we also used frequency-based FPQ to examine diet quality-microbiome links . Furthermore, another Chinese study on middle-aged and elderly individuals utilized a more detailed dietary assessment method but focused on associations of vegetables or fruits with gut microbiome. To address these limitations, we used dietary recalls which capture wider range of foods and provide detailed dietary data on the quantitative consumption of these foods allowing for a more comprehensive assessment of the associations of plant-based foods, red meat, or dairy with gut microbiome. Furthermore, in contrast to our previous study where we examined genus-level microbiome associations we now examined species-level associations. The specific aims of the current study were to examine whether the consumption of plant-based foods , red and processed meat or dairy is related to individual gut microbiome diversity , inter-individual differences in gut microbiome composition , and differences in relative abundances of bacterial species in Finnish adults. We also examined how the functional properties of the microbiome relate to these food groups.We used data from the National FINDIET 2002 Study, a sub-study of the National FINRISK 2002 Study. The FINRISK Studies have been conducted by the Finnish Institute for Health and Welfare every five years from 1972 until 2012 to monitor risk factors for non-communicable diseases in Finnish adults. FINRISK 2002 comprised of a self-administered health questionnaire and a health examination, involving a random sample from six large geographical areas in Finland drawn from the national population information system . Stool shallow shotgun sequencing was successfully performed for a total of 7231 participants of which additional 20 participants were excluded due low read counts . One third of the FINRISK participants belonged to the FINDIET 2002 subsample where dietary habits of the participants were assessed by a 48-hour dietary recall. Of those invited, 2045 completed the recall and 2007 of the recalls were accepted. After excluding pregnant women and those who had a registered purchase of antibacterial medications for systemic use code: J01) within six months prior to the baseline examination , the final data included 1273 participants with available stool samples and dietary recalls.Food consumption was assessed with a 48-hour dietary recall. Dietary recalls were conducted during the health examination by trained nutritionists who interviewed participants and recorded all foods and beverages consumed. Portion sizes were estimated using commonly used food packaging, gardening rack household measures and a validated portion size picture booklet. The mean daily energy intake and consumption of food groups were assessed using the in-house calculation software Finessi and the Finnish national food composition database maintained by the THL.

Food consumption was calculated at the ingredient level by decomposing mixed dishes into individual ingredients using standard recipes. The main food groups and their subgroups used in the study are presented in Table 1. As we were unable to analyze nuts and seeds separately due to their very low consumption, we included them within the vegetables subgroup and for the same reason we also kept legumes within the vegetables. Similarly due its low consumption, we included ice cream within the other dairy products subgroup. The food variables were categorized based on the study specific consumption quartiles for principal coordinates analysis and distance-based redundancy analysis .All who participated in the health examination of FINRISK 2002 were asked to donate a stool sample. Those willing were given a stool sampling kit and instructions during the health examination to promptly gather the sample at home at their earliest convenience. Participants collected the samples into 50 ml Falcon tubes without a stabilizing solution and then sent them overnight under Finnish winter conditions to the study personnel preferably on Monday, Tuesday, Wednesday, or Thursday, to ensure optimal preservation of the sample. The samples were immediately stored at −20 °C and were kept unthawed until sequencing in 2017. The samples were sequenced based on whole-genome, untargeted shallow shotgun sequencing at the University of California San Diego. Normalizing of the samples to 5-ng inputs were done using an Echo 550 acoustic liquid handling robot and the samples were sequenced using Illumina Hi-Seq 4000 for paired-end 150-bp reads. The average read count was approximately 900,000 reads per sample. A more detailed description of protocols for DNA extraction and library preparation can be found elsewhere. Quality trimming of the sequences and removal of sequencing adapters was performed using Atropos. After removing human DNA reads by mapping them against the reference genome assembly GRCh38 using Bowtie2, the raw sequences were taxonomically annotated using SHallow shOtGUN profiler v1.0.5 by comparing them against complete archaeal, bacterial, and viral genomes in NCBI Reference Sequence Database v82 , U.S. National Library of Medicine, Bethesda, MD, USA; May 8, 2017). The classified microbial data were used in a compositional form, meaning their relative abundances were calculated by scaling their raw counts to the total sum of reads. For taxa analyses, the data were filtered to bacterial taxa and down to a core microbiome including any bacterial species with a minimum abundance of 0.01% and a prevalence of at least 1% across all samples, similar to Salosensaari et al..Trained nurses at the study site measured weight and height using standardized international protocols with participants wearing light clothing and no shoes . Height was measured to the nearest 0.1 cm using a wall attached stadiometer and weight to the nearest 0.1 kg using a beam balance scale. BMI was calculated as kg/ m². Participants’ age was calculated based on the birth date and study date, and sex was self-reported. The self-administered questionnaire included questions on participant´s smoking history and current smoking habits. For the analysis two groups were formed: current smokers and nonsmokers who had not smoked in the last 6 months. Information on medicines which could potentially affect the microbiome in addition to the excluded systemic antimicrobial medicines was acquired from the prescription medicine purchase register maintained by the Social Insurance Institution of Finland. Participants were linked to the register through the unique personal identifier assigned to each Finnish citizen. In contrast to assessing the use of systemic antimicrobial medication, an individual was flagged as using these other drugs if he/she had at least 3 separate purchase events including a purchase within 4 months prior to baseline investigation.The analyses were conducted jointly for men and women because the results in general were similar by sex. Characteristics of the study participants are reported as means with their standard deviations for continuous variables and as percentages for categorical variables. Alpha-diversity refers to intra-individual diversity of the microbiome and it was measured using the Shannon index. The associations between alpha-diversity and the main food groups and their subgroups were assessed using linear regression analysis. Beta-diversity refers to inter-individual diversity of the gut microbiome and thus acts as a measure of compositional difference. It was measured using the Bray-Curtis dissimilarity score. Permutational multivariate analysis of variance was used to assess the amount of compositional variation in microbiomes between individuals were explained by main food groups and their subgroups. Principal coordinates analysis was used to assess and visualize clustering of microbiomes in the highest and the lowest consumption quartile of each main food group. PCoA was paired with the function “factorfit” from the vegan package to test whether the averages of the PCoA ordination scores of the highest and the lowest consumption quartiles of the main food groups differ significantly .