Page 213 - Biobehavioral Resilence to Stress

190 Biobehavioral Resilience to Stress

stimulus or “prepulse” preceding an intense startling stimulus results in an
inhibited startle response (Graham, 1975). This behavior is a simple opera-

tional measure of startle inhibition and sensorimotor gating. PPI is defi cient
in individuals with some neuropsychiatric disorders such as schizophrenia,
Tourette’s disease, Huntington’s disorder, panic disorder, and PTSD (Braff
et al., 2001; Ludewig et al., 2002, 2005). Researchers who perform twin
studies report concordance as high as 50% and 70% for PPI and baseline
startle (Anokhin, Heath, Myers, Ralano & Wood, 2003; Kumari, Das,
Zachariah, Ettinger & Sharma, 2005). A recent study of mice found that
PPI was the only one of several genetically linked types of behavior that
was relatively insensitive to the epigenetic influence of maternal environ-

ment (Francis, Szegda, Campbell, Martin & Insel, 2003). PPI is currently
under investigation as a potential marker for vulnerability to the develop-

ment of schizophrenia (Cadenhead & Braff, 2002). PPI has also been shown

to be reduced in PTSD patients, although the effect size is smaller than
that shown for other neuropsychiatric disorders (Grillon, Morgan, Davis
& Southwick, 1998; Grillon, Morgan, Southwick, Davis & Charney, 1996;
Ornitz & Pynoos, 1989).
It is important to note that genetic studies of startle response alone do not
specifically address PTSD-related phenotypes such as anxiety and fear (see

Grillon & Baas, 2002, 2003; DSMIV). In animal studies, baseline abnormali-
ties of the startle response (increased startle, reduced PPI, reduced habituation)
are often observed outside the context of stress or anxiety (Geyer, McIlwain &

Paylor, 2002). In human studies, changes in startle plasticity are also not spe-
cific to anxiety disorders, but rather are observed in many other neuropsychi-

atric disorders as well (see review by Braff et al., 2001). Thus, baseline startle or

PPI phenotypes alone are not specific probes for the discovery of gene candi-

dates involved in anxiety or PTSD (Joober, Zarate, Rouleau, Skamene & Boksa,
2002; Liu et al., 2003; Palmer et al., 2003; Swerdlow, Talledo & Braff , 2005).
However, the startle response can be used in conjunction with the study of
anxiety-related endophenotypes such as increased amygdala activation (Nutt
& Malizia, 2004; Protopopescu et al., 2005; Rauch et al., 2000), reductions in
hippocampal volume (Gilbertson et al., 2002; Gurvits et al., 1996), and exces-
sive central CRF release (Baker et al., 1999; Bremner et al., 1997; Sautter et al.,

2003). This approach is more likely to reveal genes or systems of independent
or combined relevance to PTSD. It is important to note, however, that each
behavioral measure of anxiety (e.g., in animals: startle responding, avoidance

behavior, freezing) may probe only one specific aspect of a complex construct.
Hence, each “anxiety-like” phenotype may survey a diff erent genetic system
(Crabbe, Metten, Cameron & Wahlsten, 2005).
Another important consideration is that baseline startle diff erences
may vary between studies due to differences in the type or extent of uncon-

trolled experimental stressors (Grillon, 2002) or other nonspecifi c eff ects

1/22/2008 9:36:18 PM
CRC_71777_Ch008.indd 190 1/22/2008 9:36:18 PM
CRC_71777_Ch008.indd 190