Table 1 Summary of the papers cited in the review and showing alterations in different biological systems in BPD

HPA axis

Southwick et al. [26]

37 subjects with PTSD comorbid with BPD; 18 subjects only with PTSD

Higher 24 h urinary cortisol levels in patients with PTSD compared to patients with PTSD and comorbid BPD.

Wingenfeld et al. [27]

21 female patients with BPD; 24 healthy female controls.

Higher overnight urinary cortisol levels in BPD patients compared to controls. Very high cortisol levels were found only in BPD patients with a low number of PTSD symptoms.

Rinne et al. [28]

39 BPD patients (24 with and 15 without sustained childhood abuse and comorbid PTSD (n = 12) or MDD (n = 11)); 11 control subjects

Higher ACTH and cortisol levels in the blood of BPD females who had experienced childhood abuse during the DEX/CRH test.

Carvalho Fernando et al. [29]

32 female BPD patients; 32 healthy female

Acute cortisol levels decreased the reaction time to target stimuli in both BPD patients and controls.

Martin-Blanco et al. [30]

481 subjects with BPD; 442 controls

Case-control study focusing on 47 SNPs in 10 HPA axis genes. An association between polymorphic variants within the FKPB5 and the CRHR genes with the diagnosis of BPD was shown. Two FKBP5 SNPs were more frequently represented in patients with a history of childhood trauma.

Neurotransmission

Wagner et al. [42]

159 BPD patients

Association between stressful events and low impulsivity in BPD patients. 5-HTTLPR S-allele carriers showed higher impulsivity scores when exposed to stressful events than LL omozygotes.

Wagner et al. [47]

112 female BPD patients

COMT Val158Met SNP was associated with early life stressful events and impulsive aggression in female BPD patients

Wagner et al. [48]

159 BPD patients

The effect of COMT Val158Met SNP on the association between stressful life events and impulsivity was not confirmed.

Tadic et al. [49]

161 Caucasian BPD patients; 156 healthy controls.

The COMT Met158Met SNP was over-represented in BPD patients compared to controls. No differences in 5-HTTLPR genotype were found. An interaction between the COMT Met158 and the 5-HTTLPR s-allele was observed.

Martin-Blanco et al. [50]

481 BPD subjects; 442 controls

Genetic variants within COMT, DBH and SLC6A2 genes were associated with an enhanced risk to develop BPD

Endogenous Opioid System

Kalin et al. [57]

8 infant rhesus monkeys (4 males and 4 females)

The endogenous opioid system mediates separate-induced vocalizations and influences the HPA axis activation in rhesus monkeys using the mother-infant separation paradigm.

Prossin et al. [61]

18 un-medicated female BPD patients; 14 female controls

BPD patients had greater regional μ-opioid availability at baseline in the left necleus accumbens, the hypothalamus and the right hippocampus/parahippocampus relative to controls, showing an endogenous opioid system activation.

Neuroimaging studies

Driessen et al. [36]

21 female BPD patients; 21 female controls

Volume reduction in the hippocampus and in the amygdala in BPD patients compared to controls.

Schmahl et al. [38]

25 unmedicated female patients with BPD (10 with and 15 without comorbid PTSD); 25 female controls

Hippocampal volume reduction in patients with BPD and comorbid PTSD as compared to controls.

Kreisel et al. [70]

39 BPD patients; 39 controls

Smaller hippocampal volume in BPD patients with a lifetime history than those without comorbid PTSD.

Boen et al. [71]

18 women with BPD; 21 controls

Two hippocampal structures (DG-CA4 and CA2–3 subfields) were significantly smaller in patients with BPD than controls.

Kuhlmann et al. [73]

30 BPD patients; 33 controls

Patients with BPD showed lower hippocampal volumes than controls, but higher volumes in the hypothalamus.

Rodrigues et al. [63]

124 BPD patients; 147 controls

Both the left and the right sides of the hippocampus were reduced in BPD patients with PTSD when compared to controls.

Ruocco et al. [37]

205 BPD patients; 222 controls

Bilateral volume reductions of the amygdala and hippocampus were not related to comorbid MDD, PTSD or substance use disorders.

Epigenetics

Martin-Blanco et al. [88]

281 subjects with BPD

An association between NR3C1 methylation levels and childhood trauma was found in blood samples of BPD patients.

Dammann et al. [89]

26 BPD patients; 11 controls

An increase in the methylation levels of HTR2A,NR3C1,MAOA,MAOB and COMT was found in BPD patients as compared to controls.

Perroud et al. [91]

346 BD, BPD, and ADHD patients

Differential 5-HT3AR methylation levels were associated with the severity of childhood trauma, mainly found in BPD patients.

Teschler et al. [93]

24 female BPD patients; 11 female controls

Genome-wide methylation analyses revealed increased methylation levels of several genes (APBA2,APBA3,GATA4,KCNQ1,MCF2,NINJ2,TAAR5) in blood of BPD female patients and controls.

Prados et al. [94]

96 BPD subjects suffering from a high level of child adversity; 93 subjects suffering from MDD and reporting a low rate of child maltreatment

Several CpGs within or near genes involved in inflammation and in neuronal excitability were differentially methylated in BPD patients compared to MDD patients or in relation to the severity of childhood trauma.

Teschler et al. [95]

24 female BPD patients; 11 female controls

A significant aberrant methylation of rDNA and PRIMA1 was revealed for BPD patients using pyrosequencing. For the promoter of PRIMA1, the average methylation of six CpG sites was higher in BPD patients compared to controls. In contrast, the methylation levels of the rDNA promoter region and the 5′ETS were significantly lower in patients with BPD compared to controls.

Neuroplasticity

Koenigsberg et al. [109]

24 medication-free BPD patients; 18 healthy control subjects

Decrease of PKC and BDNF protein levels in the blood of BPD patients.

Tadic et al. [49]

161 Caucasian BPD patients; 156 healthy controls.

Association between HTR1B A-161 variant and the functional BDNF 196A allele in BPD patients.

Perroud et al. [90]

115 subjects with BPD; 52 controls

Higher methylation levels in BDNF CpG exons I and IV in BPD patients than in controls. Higher BDNF protein levels in plasma of BPD patients than in controls.

Thaler et al. [92]

64 women with bulimia nervosa and comorbid BPD; 32 controls

Hypermethylation within BDNF promoter region sites in women with bulimia nervosa and with a history of BPD and/or trauma events.