Purpose

With increasing interest in predictive and personalised medicine, the concept of phenotyping an individual is evoking much attention. In this context, genome-wide analysis have shown a genetic basis for phenotypes mentioned in ayurveda, the indigenous medical system of Indian subcontinent^{1, 2}. Seven major phenotypes (labeled V, P, K, VP, PK, KV and VPK) based on a comprehensive set of physical, physiological and psychological parameters are mentioned in ayurveda³. This study has used an innovative combination of MRI for objective assessment of physical phenotyping indices and biochemical parameters for physiological indices and correlated them with ayurvedic phenotypes.

Methods

Phenotyping: Thirty four healthy volunteers (18 males, 16 females; age: 20-35 yrs) were recruited for the study. A validated questionnaire based on ayurvedic phenotyping indices was used to assess separately the physical, physiological and psychological constitution types for each volunteer. Ethical clearance was obtained from the Institute to carry out these studies.

MR studies: MRI evaluation of subcutaneous fat (SF) in abdomen (SF_Abd) and thigh (SF_Thigh) were carried out at 1.5 T (Avanto, Siemens): TR of 650 ms, TE of 11 ms, 256 x 256 matrix and 8 mm contiguous slices. T1-weighted (T1W) transverse images were obtained from abdomen (breath hold sequences) (T9 vertebra to the superior surface of hip joint) and thigh (superior surface of hip joint to the lower end of medial condyle) regions. Area of SF and fat mass were calculated for the entire region studied. Proton MRS was carried out in liver for assessing the lipid levels.

Bio-impedance Analyser (BIA): Tanita TBF-215 analyser (Japan) was used to assess Body Mass Index (BMI), Basal Metabolic Rate (BMR) and % body fat.

Biochemical factors: Total Cholesterol (TChol), Serum Triglycerides (TG) and Insulin Sensitivity (IS) were assessed using standard procedures⁴.

Human Leukocyte Antigen (HLA): PCR Sequence-Specific Oligonucleotide Probes (SSOP) was carried out in a PCR amplifier (Bio-Rad, USA) for HLA typing.

Data analysis: Pearson's correlation analysis was used for statistical evaluation. P < 0.05 was considered statistically significant. The occurrence of HLA-DRB1 alleles with volunteers’ phenotypes is presented as a frequency distribution analysis.

Results and Discussion

The volunteers were phenotyped as follows: Physical - KV (9), PK (11) and VP (14); Physiological - KV (11), PK (9) and VP (14); Psychological - KV (20), PK (7) and VP (7). KV also includes KP, PK includes PV, and VP that of VK. Figure 1 shows the entire data as a radial graph. It can be seen that different parameters correlate with different phenotypes.

Physical phenotyping

Physical correlates: SF_Abd , SF_Thigh and % body fat were maximum (p < 0.02) in K dominated phenotypes (KV and KP) followed by those in P and V. These observations are in agreement with the ayurvedic understanding of K and V dominated phenotypes as having contrasting fat distribution with P dominated as an intermediate. According to ayurveda, K dominated phenotypes are associated with presence of more fat (and predisposed to obesity and diabetes) as opposed to V types, who have very less fat and predisposed to diseases like osteoarthritis.

Biochemical correlates: K dominated phenotypes showed positive correlation with TG and TChol. This observation parallels the MRI correlation for SF in the K phenotypes. HLA- Figure 2 shows frequency distribution of HLA - DRB1 alleles in different phenotypes for all the 34 volunteers. It is interesting to observe that K dominated phenotypes showed an increased presence of the following HLA alleles - DRB1*11 (KV - 67 % and KP - 50%), DRB1*04 (KP - 50%) and DRB1*07 (KP - 60%). On the other hand, HLA -DRB*09 allele was observed more in VP individuals. This reiterates the genetic basis for the ayurvedic phenotyping^{1, 2}.

Physiological phenotyping

SF_Abd was significantly high (p< 0.05) in K dominated phenotypes which is in agreement with that observed in the physical phenotypes. Insulin sensitivity was also significantly high (p < 0.05) in VK and VP types and low in K predominant phenotypes. This is in line with known decreased insulin sensitivity in obesity. HLA- HLA - DRB1*11 allele was seen in 50% of KV and KP, DRB1*07 in 60% of KV and DRB1*09 in 60% VP phenotypes.

Conclusion

Considering that information on phenotypes are used in ayurveda to identify risk groups for conditions such as obesity and diabetes, this approach using MR and biochemical parameters for objective evaluation of ayurvedic phenotyping is innovative and could provide the much needed objective parameters for phenotyping. Further studies are under way.