**1. Introduction**

The impact of cognitive decline in a dramatically ageing population is one of the biggest challenges of the future, given its impact on both the individual and society [1]. Not only is deterioration of cognitive functioning amongs<sup>t</sup> the most feared aspects of growing old due to its ability to lower a person's quality of life, it is often accompanied by significant burden [2]. Accordingly, there is a growing need to identify practical methods of delaying cognitive decline to ensure independence in aged individuals.

Epidemiological and observational studies sugges<sup>t</sup> a relationship exists between nutrient deficiencies and cognitive decline in ageing adults [3–5]. In the case of vitamin B12, low levels (together with low folate levels) are a common cause of elevated total homocysteine (HcY), which has been shown to be an independent risk factor for cognitive decline and Alzheimer's disease [6–12]. Vitamin B12 deficiency is common in older adults, with prevalence increasing substantially with age [13–18]. In the UK, three surveys reported that around 1 in 20 people aged 65–74 years and at least 1 in 10 of those aged ≥75 years were deficient in vitamin B12 [18]. Moreover, an Australian survey of persons aged ≥50 years reported 23% of participants to have low vitamin B12 levels, with prevalence increasing to ~30% for men aged ≥70 years and women ≥80 years [13].

Over the past few decades, studies have produced inconsistent findings regarding vitamin B12 and its association with cognitive impairment in older adults [12]. A recent review of prospective cohort studies [19] found no association between serum vitamin B12 concentrations and cognitive decline. However, it was noted that the four studies that used alternate biomarkers of vitamin B12, that is, methylmalonic acid (MMA; a marker of progressed vitamin B12 deficiency) and holotranscobalamin (the biologically active fraction of this vitamin and hereafter referred to as active B12) did find evidence of an association with cognitive status [20–23]. Moreover, these markers of vitamin B12 deficiency appear to display a stronger relationship with cognition than serum B12 [16,24–26].

A 10-year longitudinal cohort study of 1648 individuals [20] found that a doubling of MMA was associated with a 50% faster rate of cognitive decline, with a doubling of active B12 associated with a 30% slower rate of decline in individuals aged over 65 years. Furthermore, a 5-year prospective study of 107 community-dwelling participants aged 61–87 years without cognitive impairment at enrolment showed that the decrease in brain volume was greatest among those with lower serum vitamin B12 and active B12 levels and higher plasma total homocysteine (HcY) and MMA levels at baseline [27]. These results are consistent with findings by Tangney et al. [28], in which high HcY and MMA levels were associated with rate of cognitive decline, along with a decrease in total brain volume.

A recent comprehensive review of the literature on the association of dairy foods and cognitive function in older adults concluded that "low-fat dairy products, when consumed regularly as part of a balanced diet, may have a number of beneficial outcomes for neuro-cognitive health during ageing" and identified the whey fraction as being particularly rich in a range of bioactives, including proline-rich polypeptides, α-lactalbumin and B12, which may support brain health in diverse ways [10]. Whey is a natural by-product of the manufacture of cheese, with 100 g of dried powder providing 2.5 μg of vitamin B12, equivalent to 100% of the Recommended Dietary Allowance (RDA). Acid whey powder contains high amounts of lactose (60–70 g per 100 g), which is problematic given the increased risk of lactose malabsorption in older adults [29]. Whey protein isolate (WPI), on the other hand, has less than 1 g lactose per 100 g and is also richer in vitamin B12 (6 μg/100 g). Additionally, WPI has an 80% lower level of sodium and 14-fold higher level of natural folate relative to acid whey powder. While WPI has already been shown to be beneficial for improving cognitive performance in the context of memory tasks in stress-vulnerable subjects [30], the bioavailability and bioefficacy of vitamin B12 from WPI remains relatively unexplored, particularly in relation to age-related cognitive decline.

The purpose of the present study was to consider the role that dairy-based products such as WPI may have on nutritional markers associated with age-related cognitive decline. Primary outcomes including the impact of WPI supplementation on vitamin B12 status and related factors in individuals with subclinical B12 levels at baseline have already been reported [31] but are briefly revisited herein. This study specifically explores the impact of WPI supplementation on cognitive function measured during the intervention. Furthermore, it explores the relationship between changes in various biomarkers of B12 status (serum B12, active B12, MMA and HcY) and changes in cognitive function during the course of the active intervention.

#### **2. Materials and Methods**

#### *2.1. Screening and Recruitment of Participants*

The study was advertised in local newspapers in the Adelaide metropolitan area and on the CSIRO clinic website. Interested participants contacted the clinic to arrange an appointment for screening. Fifty-six eligible participants attended an information session and read the study information sheet before providing written informed consent to participate in the study. Figure 1 provides an overview of screening and recruitment and Table 1 provides an overview of baseline characteristics of study completers. All data collection visits for this study occurred between September 2014 and June 2015.

Inclusion criteria were: healthy subjects aged between 45 and 75 years, not taking vitamin B12/choline/antioxidant vitamins at doses that exceed 25% of the Recommended Dietary Allowance, subclinical vitamin B12 deficiency as defined previously [31] (using the following parameters: serum concentration of B12 in the range of 100–350 pmol/L, plasma MMA >0.20 μmol/L and serum creatinine concentration of 120 μmol/L or less) and willing to consume the quantities of WPI or SPI specified for the trial.

Exclusion criteria were: cognitive deficiencies indicated by Mini Mental State Examination Score ≤24, current smokers, people who habitually consume more than two standard alcoholic drinks per day, BMI ≥35 kg/m2, diagnosed with diabetes and/or lactose intolerance, history of pernicious anaemia or atrophic gastritis and regular users of antacids.

**Figure 1.** Consort diagram of the intervention trial.

**Table 1.** Baseline characteristics of participants who completed the study.


SPI: soy protein isolate; WPI: whey protein isolate.
